[Met_help] [rt.rap.ucar.edu #84316] History for MET 6.1 bugfix
John Halley Gotway via RT
met_help at ucar.edu
Tue Mar 13 10:36:01 MDT 2018
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
I built the MET 6.1, but the tests won't run. My user is getting the same error in trying to open the file for writing using the tutorial:
cd scripts; make
make[1]: Entering directory `/users/ramos/met/met-6.1_bugfix/scripts'
*** Running ASCII2NC to reformat ASCII point observations into NetCDF ***
../src/tools/other/ascii2nc/ascii2nc \
../data/sample_obs/ascii/sample_ascii_obs.txt \
../out/ascii2nc/sample_ascii.nc \
-v 2
DEBUG 1: Config File: /software/depot/met-6.1_bugfix/share/met/config/Ascii2NcConfig_default
DEBUG 2: Rejected 0 observations off the masking grid.
DEBUG 2: Rejected 0 observations outside the masking polyline.
DEBUG 2: Rejected 0 observations not matched with station ID's.
DEBUG 2: Processing observations for 5 headers.
DEBUG 1: Creating NetCDF Observation file: ../out/ascii2nc/sample_ascii.nc
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
make[1]: *** [ascii2nc] Error 1
make[1]: Leaving directory `/users/ramos/met/met-6.1_bugfix/scripts'
make: *** [test] Error 2
[ramos at maury2 met-6.1_bugfix]$
I am guessing this is a common error.
Greg Ramos
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: MET 6.1 bugfix
From: Howard Soh
Time: Tue Mar 06 09:23:16 2018
Would you check if ../out/ascii2nc exists?
The common cause are 1) when the output directory does not exist or 2)
the
user does not have the proper privileges (write privilege).
Would you check these?
ls ../out/ascii2nc/
touch ../out/ascii2nc/test.txt
rm ../out/ascii2nc/test.txt
Cheers,
Howard
On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> Queue: met_help
> Subject: MET 6.1 bugfix
> Owner: Nobody
> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
>
> I built the MET 6.1, but the tests won't run. My user is getting the
same
> error in trying to open the file for writing using the tutorial:
>
>
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF ***
> ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-
6.1_bugfix/share/met/config/
> Ascii2NcConfig_default
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
../out/ascii2nc/sample_ascii.nc
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-6.1_bugfix/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-6.1_bugfix]$
>
> I am guessing this is a common error.
>
> Greg Ramos
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 09:25:23 2018
The output I sent you was from "make test". So I can't even get the
tests to run from the build. Thank you.
Greg Ramos
-----Original Message-----
From: Howard Soh via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 8:23 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Would you check if ../out/ascii2nc exists?
The common cause are 1) when the output directory does not exist or 2)
the user does not have the proper privileges (write privilege).
Would you check these?
ls ../out/ascii2nc/
touch ../out/ascii2nc/test.txt
rm ../out/ascii2nc/test.txt
Cheers,
Howard
On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> Queue: met_help
> Subject: MET 6.1 bugfix
> Owner: Nobody
> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >
>
>
> I built the MET 6.1, but the tests won't run. My user is getting the
> same error in trying to open the file for writing using the
tutorial:
>
>
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-
6.1_bugfix/share/met/config/
> Ascii2NcConfig_default
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-6.1_bugfix/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-6.1_bugfix]$
>
> I am guessing this is a common error.
>
> Greg Ramos
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 09:49:01 2018
This morning I downloaded the latest version of MET. The tests still
don't run:
[ramos at maury2 met-7.0]$ make test
cd scripts; make
make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
*** Running ASCII2NC to reformat ASCII point observations into NetCDF
***
../src/tools/other/ascii2nc/ascii2nc \
../data/sample_obs/ascii/sample_ascii_obs.txt \
../out/ascii2nc/sample_ascii.nc \
-v 2
DEBUG 1: Config File: /software/depot/met-
7.0/share/met/config/Ascii2NcConfig_default
DEBUG 2: Rejected 0 observations off the masking grid.
DEBUG 2: Rejected 0 observations outside the masking polyline.
DEBUG 2: Rejected 0 observations not matched with station ID's.
DEBUG 2: Processing observations for 5 headers.
DEBUG 1: Creating NetCDF Observation file:
../out/ascii2nc/sample_ascii.nc
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
make[1]: *** [ascii2nc] Error 1
make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
make: *** [test] Error 2
[ramos at maury2 met-7.0]$
-----Original Message-----
From: Howard Soh via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 8:23 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Would you check if ../out/ascii2nc exists?
The common cause are 1) when the output directory does not exist or 2)
the user does not have the proper privileges (write privilege).
Would you check these?
ls ../out/ascii2nc/
touch ../out/ascii2nc/test.txt
rm ../out/ascii2nc/test.txt
Cheers,
Howard
On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> Queue: met_help
> Subject: MET 6.1 bugfix
> Owner: Nobody
> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >
>
>
> I built the MET 6.1, but the tests won't run. My user is getting the
> same error in trying to open the file for writing using the
tutorial:
>
>
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-
6.1_bugfix/share/met/config/
> Ascii2NcConfig_default
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-6.1_bugfix/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-6.1_bugfix]$
>
> I am guessing this is a common error.
>
> Greg Ramos
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 10:12:00 2018
Hi Greg,
This is John, I work with Howard. Well I'm glad to hear that the
behavior
is consistent across versions. If not, I would have been confused.
So the very first command that's being run by the test script is
failing
because it isn't able to create an output file.
As Howard mentioned, the most likely cause for this type of error is a
permissions issue. So let's rule that out first. Please try running
the
following commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
The touch command just creates an empty file. If you get an error
message
about permissions, then we've found the problem. I tried running the
exact
same ascii2nc command but writing my output to "/usr/bin"... where of
course I don't have permission to write. Here's the error message I
got:
cd met-7.0/scripts
../src/tools/other/ascii2nc/ascii2nc
../data/sample_obs/ascii/sample_ascii_obs.txt
/usr/bin/sample_ascii_obs.txt
-v 2
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"/usr/bin/sample_ascii_obs.txt" for writing
ERROR :
Do a long listing of that directory to see the permissions:
ls -la /users/ramos/met/met-7.0/out/ascii2nc
What do you see?
Thanks,
John
On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This morning I downloaded the latest version of MET. The tests still
don't
> run:
>
> [ramos at maury2 met-7.0]$ make test
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF ***
> ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
../out/ascii2nc/sample_ascii.nc
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 8:23 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Would you check if ../out/ascii2nc exists?
>
> The common cause are 1) when the output directory does not exist or
2) the
> user does not have the proper privileges (write privilege).
>
> Would you check these?
>
> ls ../out/ascii2nc/
> touch ../out/ascii2nc/test.txt
> rm ../out/ascii2nc/test.txt
>
> Cheers,
> Howard
>
>
> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > Queue: met_help
> > Subject: MET 6.1 bugfix
> > Owner: Nobody
> > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >
> >
> >
> > I built the MET 6.1, but the tests won't run. My user is getting
the
> > same error in trying to open the file for writing using the
tutorial:
> >
> >
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-
6.1_bugfix/share/met/config/
> > Ascii2NcConfig_default
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-6.1_bugfix]$
> >
> > I am guessing this is a common error.
> >
> > Greg Ramos
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 10:15:39 2018
[ramos at maury2 scripts]$ pwd
/users/ramos/met/met-7.0/scripts
[ramos at maury2 scripts]$ ls -la /users/ramos/met/met-
7.0/out/ascii2nc
total 8
drwxrwxr-x 2 ramos ramos 4096 Mar 5 22:01 .
drwxrwxr-x 17 ramos ramos 4096 Mar 5 22:01 ..
[ramos at maury2 scripts]$
I own the entire tree.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:12 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Hi Greg,
This is John, I work with Howard. Well I'm glad to hear that the
behavior is consistent across versions. If not, I would have been
confused.
So the very first command that's being run by the test script is
failing because it isn't able to create an output file.
As Howard mentioned, the most likely cause for this type of error is a
permissions issue. So let's rule that out first. Please try running
the following commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
The touch command just creates an empty file. If you get an error
message about permissions, then we've found the problem. I tried
running the exact same ascii2nc command but writing my output to
"/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
cd met-7.0/scripts
../src/tools/other/ascii2nc/ascii2nc
../data/sample_obs/ascii/sample_ascii_obs.txt
/usr/bin/sample_ascii_obs.txt -v 2
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"/usr/bin/sample_ascii_obs.txt" for writing ERROR :
Do a long listing of that directory to see the permissions:
ls -la /users/ramos/met/met-7.0/out/ascii2nc
What do you see?
Thanks,
John
On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This morning I downloaded the latest version of MET. The tests still
> don't
> run:
>
> [ramos at maury2 met-7.0]$ make test
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 8:23 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Would you check if ../out/ascii2nc exists?
>
> The common cause are 1) when the output directory does not exist or
2)
> the user does not have the proper privileges (write privilege).
>
> Would you check these?
>
> ls ../out/ascii2nc/
> touch ../out/ascii2nc/test.txt
> rm ../out/ascii2nc/test.txt
>
> Cheers,
> Howard
>
>
> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > Queue: met_help
> > Subject: MET 6.1 bugfix
> > Owner: Nobody
> > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >
> >
> >
> > I built the MET 6.1, but the tests won't run. My user is getting
the
> > same error in trying to open the file for writing using the
tutorial:
> >
> >
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File:
> > /software/depot/met-6.1_bugfix/share/met/config/
> > Ascii2NcConfig_default
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-6.1_bugfix]$
> >
> > I am guessing this is a common error.
> >
> > Greg Ramos
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 10:20:54 2018
Greg,
Yes, I see that you do own the tree. I'm surprised that I don't see
the
file in that directory named "test_output_file". That would have been
the
output of these commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
So was the "touch" command unsuccessful?
Thanks,
John
On Tue, Mar 6, 2018 at 10:15 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> [ramos at maury2 scripts]$ pwd
> /users/ramos/met/met-7.0/scripts
> [ramos at maury2 scripts]$ ls -la /users/ramos/met/met-
7.0/out/ascii2nc
> total 8
> drwxrwxr-x 2 ramos ramos 4096 Mar 5 22:01 .
> drwxrwxr-x 17 ramos ramos 4096 Mar 5 22:01 ..
> [ramos at maury2 scripts]$
>
>
> I own the entire tree.
>
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:12 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Hi Greg,
>
> This is John, I work with Howard. Well I'm glad to hear that the
behavior
> is consistent across versions. If not, I would have been confused.
>
> So the very first command that's being run by the test script is
failing
> because it isn't able to create an output file.
>
> As Howard mentioned, the most likely cause for this type of error is
a
> permissions issue. So let's rule that out first. Please try
running the
> following commands:
> cd /users/ramos/met/met-7.0/scripts
> touch ../out/ascii2nc/test_output_file
>
> The touch command just creates an empty file. If you get an error
message
> about permissions, then we've found the problem. I tried running
the exact
> same ascii2nc command but writing my output to "/usr/bin"... where
of
> course I don't have permission to write. Here's the error message I
got:
>
> cd met-7.0/scripts
> ../src/tools/other/ascii2nc/ascii2nc
> ../data/sample_obs/ascii/sample_ascii_obs.txt
> /usr/bin/sample_ascii_obs.txt -v 2
>
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>
> Do a long listing of that directory to see the permissions:
> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>
> What do you see?
>
> Thanks,
> John
>
>
> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This morning I downloaded the latest version of MET. The tests
still
> > don't
> > run:
> >
> > [ramos at maury2 met-7.0]$ make test
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> > the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 8:23 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Would you check if ../out/ascii2nc exists?
> >
> > The common cause are 1) when the output directory does not exist
or 2)
> > the user does not have the proper privileges (write privilege).
> >
> > Would you check these?
> >
> > ls ../out/ascii2nc/
> > touch ../out/ascii2nc/test.txt
> > rm ../out/ascii2nc/test.txt
> >
> > Cheers,
> > Howard
> >
> >
> > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > Queue: met_help
> > > Subject: MET 6.1 bugfix
> > > Owner: Nobody
> > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > Status: new
> > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > >
> > >
> > >
> > > I built the MET 6.1, but the tests won't run. My user is getting
the
> > > same error in trying to open the file for writing using the
tutorial:
> > >
> > >
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File:
> > > /software/depot/met-6.1_bugfix/share/met/config/
> > > Ascii2NcConfig_default
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-6.1_bugfix]$
> > >
> > > I am guessing this is a common error.
> > >
> > > Greg Ramos
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 10:21:00 2018
[ramos at maury2 met-7.0]$ cd scripts/
[ramos at maury2 scripts]$ touch ../out/ascii2nc/test_output_file
[ramos at maury2 scripts]$ ls -larth ../out/ascii2nc/test_output_file
-rw-rw-r-- 1 ramos ramos 0 Mar 6 17:19
../out/ascii2nc/test_output_file
[ramos at maury2 scripts]$
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:12 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Hi Greg,
This is John, I work with Howard. Well I'm glad to hear that the
behavior is consistent across versions. If not, I would have been
confused.
So the very first command that's being run by the test script is
failing because it isn't able to create an output file.
As Howard mentioned, the most likely cause for this type of error is a
permissions issue. So let's rule that out first. Please try running
the following commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
The touch command just creates an empty file. If you get an error
message about permissions, then we've found the problem. I tried
running the exact same ascii2nc command but writing my output to
"/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
cd met-7.0/scripts
../src/tools/other/ascii2nc/ascii2nc
../data/sample_obs/ascii/sample_ascii_obs.txt
/usr/bin/sample_ascii_obs.txt -v 2
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"/usr/bin/sample_ascii_obs.txt" for writing ERROR :
Do a long listing of that directory to see the permissions:
ls -la /users/ramos/met/met-7.0/out/ascii2nc
What do you see?
Thanks,
John
On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This morning I downloaded the latest version of MET. The tests still
> don't
> run:
>
> [ramos at maury2 met-7.0]$ make test
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 8:23 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Would you check if ../out/ascii2nc exists?
>
> The common cause are 1) when the output directory does not exist or
2)
> the user does not have the proper privileges (write privilege).
>
> Would you check these?
>
> ls ../out/ascii2nc/
> touch ../out/ascii2nc/test.txt
> rm ../out/ascii2nc/test.txt
>
> Cheers,
> Howard
>
>
> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > Queue: met_help
> > Subject: MET 6.1 bugfix
> > Owner: Nobody
> > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >
> >
> >
> > I built the MET 6.1, but the tests won't run. My user is getting
the
> > same error in trying to open the file for writing using the
tutorial:
> >
> >
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File:
> > /software/depot/met-6.1_bugfix/share/met/config/
> > Ascii2NcConfig_default
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-6.1_bugfix]$
> >
> > I am guessing this is a common error.
> >
> > Greg Ramos
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 10:26:11 2018
The touch command was successful, I forgot to include it. This is the
same error my user is getting trying to run the examples in the
tutorial, but in her own directories.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:21 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Yes, I see that you do own the tree. I'm surprised that I don't see
the file in that directory named "test_output_file". That would have
been the output of these commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
So was the "touch" command unsuccessful?
Thanks,
John
On Tue, Mar 6, 2018 at 10:15 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> [ramos at maury2 scripts]$ pwd
> /users/ramos/met/met-7.0/scripts
> [ramos at maury2 scripts]$ ls -la /users/ramos/met/met-
7.0/out/ascii2nc
> total 8
> drwxrwxr-x 2 ramos ramos 4096 Mar 5 22:01 .
> drwxrwxr-x 17 ramos ramos 4096 Mar 5 22:01 ..
> [ramos at maury2 scripts]$
>
>
> I own the entire tree.
>
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:12 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Hi Greg,
>
> This is John, I work with Howard. Well I'm glad to hear that the
> behavior is consistent across versions. If not, I would have been
confused.
>
> So the very first command that's being run by the test script is
> failing because it isn't able to create an output file.
>
> As Howard mentioned, the most likely cause for this type of error is
a
> permissions issue. So let's rule that out first. Please try
running
> the following commands:
> cd /users/ramos/met/met-7.0/scripts
> touch ../out/ascii2nc/test_output_file
>
> The touch command just creates an empty file. If you get an error
> message about permissions, then we've found the problem. I tried
> running the exact same ascii2nc command but writing my output to
> "/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
>
> cd met-7.0/scripts
> ../src/tools/other/ascii2nc/ascii2nc
> ../data/sample_obs/ascii/sample_ascii_obs.txt
> /usr/bin/sample_ascii_obs.txt -v 2
>
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>
> Do a long listing of that directory to see the permissions:
> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>
> What do you see?
>
> Thanks,
> John
>
>
> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This morning I downloaded the latest version of MET. The tests
still
> > don't
> > run:
> >
> > [ramos at maury2 met-7.0]$ make test
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
> > off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 8:23 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Would you check if ../out/ascii2nc exists?
> >
> > The common cause are 1) when the output directory does not exist
or
> > 2) the user does not have the proper privileges (write privilege).
> >
> > Would you check these?
> >
> > ls ../out/ascii2nc/
> > touch ../out/ascii2nc/test.txt
> > rm ../out/ascii2nc/test.txt
> >
> > Cheers,
> > Howard
> >
> >
> > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > Queue: met_help
> > > Subject: MET 6.1 bugfix
> > > Owner: Nobody
> > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > Status: new
> > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > >
> > >
> > >
> > > I built the MET 6.1, but the tests won't run. My user is getting
> > > the same error in trying to open the file for writing using the
tutorial:
> > >
> > >
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File:
> > > /software/depot/met-6.1_bugfix/share/met/config/
> > > Ascii2NcConfig_default
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-6.1_bugfix]$
> > >
> > > I am guessing this is a common error.
> > >
> > > Greg Ramos
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 10:38:55 2018
These are the binaries that were built. It did not build all of the
binaries listed in the README:
[ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
total 208M
drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
-rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
-rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 grid_stat
-rwxr-xr-x 1 ramos ramos 13M Mar 6 16:44 mode
-rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44 mode_analysis
-rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44 pcp_combine
-rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat
-rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
-rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat
-rwxr-xr-x 1 ramos ramos 9.8M Mar 6 16:44 series_analysis
-rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 gen_vx_mask
-rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44 plot_data_plane
-rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 plot_point_obs
-rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44 ascii2nc
-rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
-rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc
-rwxr-xr-x 1 ramos ramos 8.4M Mar 6 16:44 wwmca_regrid
-rwxr-xr-x 1 ramos ramos 8.9M Mar 6 16:44 wwmca_plot
-rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
-rwxr-xr-x 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr
-rwxr-xr-x 1 ramos ramos 7.5M Mar 6 16:44 gsidens2orank
-rwxr-xr-x 1 ramos ramos 7.6M Mar 6 16:44 regrid_data_plane
-rwxr-xr-x 1 ramos ramos 7.2M Mar 6 16:44 shift_data_plane
-rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44 tc_dland
-rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs
drwxrwxr-x 2 ramos ramos 4.0K Mar 6 16:44 .
-rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
[ramos at maury2 met-7.0]$
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:12 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Hi Greg,
This is John, I work with Howard. Well I'm glad to hear that the
behavior is consistent across versions. If not, I would have been
confused.
So the very first command that's being run by the test script is
failing because it isn't able to create an output file.
As Howard mentioned, the most likely cause for this type of error is a
permissions issue. So let's rule that out first. Please try running
the following commands:
cd /users/ramos/met/met-7.0/scripts
touch ../out/ascii2nc/test_output_file
The touch command just creates an empty file. If you get an error
message about permissions, then we've found the problem. I tried
running the exact same ascii2nc command but writing my output to
"/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
cd met-7.0/scripts
../src/tools/other/ascii2nc/ascii2nc
../data/sample_obs/ascii/sample_ascii_obs.txt
/usr/bin/sample_ascii_obs.txt -v 2
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"/usr/bin/sample_ascii_obs.txt" for writing ERROR :
Do a long listing of that directory to see the permissions:
ls -la /users/ramos/met/met-7.0/out/ascii2nc
What do you see?
Thanks,
John
On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This morning I downloaded the latest version of MET. The tests still
> don't
> run:
>
> [ramos at maury2 met-7.0]$ make test
> cd scripts; make
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 8:23 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Would you check if ../out/ascii2nc exists?
>
> The common cause are 1) when the output directory does not exist or
2)
> the user does not have the proper privileges (write privilege).
>
> Would you check these?
>
> ls ../out/ascii2nc/
> touch ../out/ascii2nc/test.txt
> rm ../out/ascii2nc/test.txt
>
> Cheers,
> Howard
>
>
> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > Queue: met_help
> > Subject: MET 6.1 bugfix
> > Owner: Nobody
> > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >
> >
> >
> > I built the MET 6.1, but the tests won't run. My user is getting
the
> > same error in trying to open the file for writing using the
tutorial:
> >
> >
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File:
> > /software/depot/met-6.1_bugfix/share/met/config/
> > Ascii2NcConfig_default
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-6.1_bugfix]$
> >
> > I am guessing this is a common error.
> >
> > Greg Ramos
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 10:57:50 2018
Greg,
The number of executables built depends on the way the build was
configured. Options which require "extra" dependencies are disabled
by
default. Here's an excerpt from the top-level README file:
To configure MET to install all of the available tools in the "bin"
subdirectory of your current directory, you would use the following
command:
./configure --prefix=`pwd` --enable-grib2 --enable-modis \
--enable-mode_graphics --enable-lidar2nc
So it's fine that you don't have all possible executables built.
That's
unrelated to the test script issue.
My next guess would be an issue with the NetCDF library to which MET
is
linking. Let's replicate the error by trying to write to some other
output
directory, like your home directory:
/users/ramos/met/met-7.0/bin/ascii2nc \
/users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
~/sample_ascii.nc \
-v 2
I expect that this will fail with the same error message. If it does,
you
could try recompiling NetCDF... and then recompiling MET. Perhaps
there's
some incompatibility in how NetCDF and MET are built?
Here's some info about that:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
Thanks,
John
On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> These are the binaries that were built. It did not build all of the
> binaries listed in the README:
>
> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> total 208M
> drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 grid_stat
> -rwxr-xr-x 1 ramos ramos 13M Mar 6 16:44 mode
> -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44 mode_analysis
> -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44 pcp_combine
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat
> -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
> -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat
> -rwxr-xr-x 1 ramos ramos 9.8M Mar 6 16:44 series_analysis
> -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 gen_vx_mask
> -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44 plot_data_plane
> -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 plot_point_obs
> -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44 ascii2nc
> -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc
> -rwxr-xr-x 1 ramos ramos 8.4M Mar 6 16:44 wwmca_regrid
> -rwxr-xr-x 1 ramos ramos 8.9M Mar 6 16:44 wwmca_plot
> -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
> -rwxr-xr-x 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr
> -rwxr-xr-x 1 ramos ramos 7.5M Mar 6 16:44 gsidens2orank
> -rwxr-xr-x 1 ramos ramos 7.6M Mar 6 16:44 regrid_data_plane
> -rwxr-xr-x 1 ramos ramos 7.2M Mar 6 16:44 shift_data_plane
> -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44 tc_dland
> -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs
> drwxrwxr-x 2 ramos ramos 4.0K Mar 6 16:44 .
> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:12 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Hi Greg,
>
> This is John, I work with Howard. Well I'm glad to hear that the
behavior
> is consistent across versions. If not, I would have been confused.
>
> So the very first command that's being run by the test script is
failing
> because it isn't able to create an output file.
>
> As Howard mentioned, the most likely cause for this type of error is
a
> permissions issue. So let's rule that out first. Please try
running the
> following commands:
> cd /users/ramos/met/met-7.0/scripts
> touch ../out/ascii2nc/test_output_file
>
> The touch command just creates an empty file. If you get an error
message
> about permissions, then we've found the problem. I tried running
the exact
> same ascii2nc command but writing my output to "/usr/bin"... where
of
> course I don't have permission to write. Here's the error message I
got:
>
> cd met-7.0/scripts
> ../src/tools/other/ascii2nc/ascii2nc
> ../data/sample_obs/ascii/sample_ascii_obs.txt
> /usr/bin/sample_ascii_obs.txt -v 2
>
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>
> Do a long listing of that directory to see the permissions:
> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>
> What do you see?
>
> Thanks,
> John
>
>
> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This morning I downloaded the latest version of MET. The tests
still
> > don't
> > run:
> >
> > [ramos at maury2 met-7.0]$ make test
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> > the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 8:23 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Would you check if ../out/ascii2nc exists?
> >
> > The common cause are 1) when the output directory does not exist
or 2)
> > the user does not have the proper privileges (write privilege).
> >
> > Would you check these?
> >
> > ls ../out/ascii2nc/
> > touch ../out/ascii2nc/test.txt
> > rm ../out/ascii2nc/test.txt
> >
> > Cheers,
> > Howard
> >
> >
> > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > Queue: met_help
> > > Subject: MET 6.1 bugfix
> > > Owner: Nobody
> > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > Status: new
> > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > >
> > >
> > >
> > > I built the MET 6.1, but the tests won't run. My user is getting
the
> > > same error in trying to open the file for writing using the
tutorial:
> > >
> > >
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File:
> > > /software/depot/met-6.1_bugfix/share/met/config/
> > > Ascii2NcConfig_default
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-6.1_bugfix]$
> > >
> > > I am guessing this is a common error.
> > >
> > > Greg Ramos
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 11:06:54 2018
I had to modify your example slightly, but it didn't work:
[ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
/users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
~/sample_ascii.nc \ -v 2
DEBUG 1: Config File: /software/depot/met-
7.0/share/met/config/Ascii2NcConfig_default
ERROR :
ERROR : FileHandler::processFiles() -> can't open input ASCII file "
~/sample_ascii.nc" for reading
ERROR :
DEBUG 2: Rejected 0 observations off the masking grid.
DEBUG 2: Rejected 0 observations outside the masking polyline.
DEBUG 2: Rejected 0 observations not matched with station ID's.
DEBUG 2: Processing observations for 5 headers.
DEBUG 1: Creating NetCDF Observation file: 2
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"2" for writing
ERROR :
[ramos at maury2 met-7.0]$
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:58 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
The number of executables built depends on the way the build was
configured. Options which require "extra" dependencies are disabled
by default. Here's an excerpt from the top-level README file:
To configure MET to install all of the available tools in the "bin"
subdirectory of your current directory, you would use the following
command:
./configure --prefix=`pwd` --enable-grib2 --enable-modis \
--enable-mode_graphics --enable-lidar2nc
So it's fine that you don't have all possible executables built.
That's unrelated to the test script issue.
My next guess would be an issue with the NetCDF library to which MET
is linking. Let's replicate the error by trying to write to some
other output directory, like your home directory:
/users/ramos/met/met-7.0/bin/ascii2nc \
/users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
~/sample_ascii.nc \ -v 2
I expect that this will fail with the same error message. If it does,
you could try recompiling NetCDF... and then recompiling MET. Perhaps
there's some incompatibility in how NetCDF and MET are built?
Here's some info about that:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
Thanks,
John
On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> These are the binaries that were built. It did not build all of the
> binaries listed in the README:
>
> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
total
> 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-xr-x
1
> ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos ramos
13M
> Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44 pcp_combine
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x 1
> ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos ramos
> 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar 6
> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
plot_data_plane
> -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 plot_point_obs -rwxr-xr-x
1
> ramos ramos 8.3M Mar 6 16:44 ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M
> Mar 6 16:44 madis2nc -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44
pb2nc
> -rwxr-xr-x 1 ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
> ramos ramos 8.9M Mar 6 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos
10M
> Mar 6 16:44 mtd -rwxr-xr-x 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr
> -rwxr-xr-x 1 ramos ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x
1
> ramos ramos 7.6M Mar 6 16:44 regrid_data_plane -rwxr-xr-x 1 ramos
> ramos 7.2M Mar 6 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos
3.6M
> Mar 6 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> tc_pairs drwxrwxr-x 2 ramos ramos 4.0K Mar 6 16:44 .
> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:12 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Hi Greg,
>
> This is John, I work with Howard. Well I'm glad to hear that the
> behavior is consistent across versions. If not, I would have been
confused.
>
> So the very first command that's being run by the test script is
> failing because it isn't able to create an output file.
>
> As Howard mentioned, the most likely cause for this type of error is
a
> permissions issue. So let's rule that out first. Please try
running
> the following commands:
> cd /users/ramos/met/met-7.0/scripts
> touch ../out/ascii2nc/test_output_file
>
> The touch command just creates an empty file. If you get an error
> message about permissions, then we've found the problem. I tried
> running the exact same ascii2nc command but writing my output to
> "/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
>
> cd met-7.0/scripts
> ../src/tools/other/ascii2nc/ascii2nc
> ../data/sample_obs/ascii/sample_ascii_obs.txt
> /usr/bin/sample_ascii_obs.txt -v 2
>
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>
> Do a long listing of that directory to see the permissions:
> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>
> What do you see?
>
> Thanks,
> John
>
>
> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This morning I downloaded the latest version of MET. The tests
still
> > don't
> > run:
> >
> > [ramos at maury2 met-7.0]$ make test
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
> > off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 8:23 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Would you check if ../out/ascii2nc exists?
> >
> > The common cause are 1) when the output directory does not exist
or
> > 2) the user does not have the proper privileges (write privilege).
> >
> > Would you check these?
> >
> > ls ../out/ascii2nc/
> > touch ../out/ascii2nc/test.txt
> > rm ../out/ascii2nc/test.txt
> >
> > Cheers,
> > Howard
> >
> >
> > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > Queue: met_help
> > > Subject: MET 6.1 bugfix
> > > Owner: Nobody
> > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > Status: new
> > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > >
> > >
> > >
> > > I built the MET 6.1, but the tests won't run. My user is getting
> > > the same error in trying to open the file for writing using the
tutorial:
> > >
> > >
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File:
> > > /software/depot/met-6.1_bugfix/share/met/config/
> > > Ascii2NcConfig_default
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-6.1_bugfix]$
> > >
> > > I am guessing this is a common error.
> > >
> > > Greg Ramos
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 11:13:40 2018
Greg,
To clarify, those backslashes '\' are line continuation marks...
telling
the shell that the command is continued on the next line. Including
them
on one line like that causes the odd behavior of creating an output
file
named "2". Removing them and running:
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
Will still result in an error... but the requested output file will be
what
you expect.
So I'd try recompiling NetCDF using the same family/version number of
the
compilers you're using to compile MET. Sorry I don't have an easier
solution for you.
Thanks,
John
On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I had to modify your example slightly, but it didn't work:
>
>
> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\ ~/
> sample_ascii.nc \ -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default
> ERROR :
> ERROR : FileHandler::processFiles() -> can't open input ASCII file
" ~/
> sample_ascii.nc" for reading
> ERROR :
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file: 2
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"2"
> for writing
> ERROR :
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:58 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> The number of executables built depends on the way the build was
> configured. Options which require "extra" dependencies are disabled
by
> default. Here's an excerpt from the top-level README file:
>
> To configure MET to install all of the available tools in the
"bin"
> subdirectory of your current directory, you would use the
following
> command:
>
> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> --enable-mode_graphics --enable-lidar2nc
>
> So it's fine that you don't have all possible executables built.
That's
> unrelated to the test script issue.
>
> My next guess would be an issue with the NetCDF library to which MET
is
> linking. Let's replicate the error by trying to write to some other
output
> directory, like your home directory:
>
> /users/ramos/met/met-7.0/bin/ascii2nc \
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\ ~/
> sample_ascii.nc \ -v 2
>
> I expect that this will fail with the same error message. If it
does, you
> could try recompiling NetCDF... and then recompiling MET. Perhaps
there's
> some incompatibility in how NetCDF and MET are built?
>
> Here's some info about that:
> https://dtcenter.org/met/users/support/online_tutorial/
> METv6.1/tutorial.php?name=compilation&category=req_libs
>
> Thanks,
> John
>
>
>
>
> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > These are the binaries that were built. It did not build all of
the
> > binaries listed in the README:
> >
> > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
total
> > 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-xr-
x 1
> > ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos ramos
13M
> > Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
pcp_combine
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x 1
> > ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos
ramos
> > 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar 6
> > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
plot_data_plane
> > -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 plot_point_obs -rwxr-
xr-x 1
> > ramos ramos 8.3M Mar 6 16:44 ascii2nc -rwxr-xr-x 1 ramos ramos
6.6M
> > Mar 6 16:44 madis2nc -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44
pb2nc
> > -rwxr-xr-x 1 ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x
1
> > ramos ramos 8.9M Mar 6 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos
10M
> > Mar 6 16:44 mtd -rwxr-xr-x 1 ramos ramos 6.9M Mar 6 16:44
gsid2mpr
> > -rwxr-xr-x 1 ramos ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-
x 1
> > ramos ramos 7.6M Mar 6 16:44 regrid_data_plane -rwxr-xr-x 1 ramos
> > ramos 7.2M Mar 6 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos
3.6M
> > Mar 6 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > tc_pairs drwxrwxr-x 2 ramos ramos 4.0K Mar 6 16:44 .
> > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:12 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Hi Greg,
> >
> > This is John, I work with Howard. Well I'm glad to hear that the
> > behavior is consistent across versions. If not, I would have been
> confused.
> >
> > So the very first command that's being run by the test script is
> > failing because it isn't able to create an output file.
> >
> > As Howard mentioned, the most likely cause for this type of error
is a
> > permissions issue. So let's rule that out first. Please try
running
> > the following commands:
> > cd /users/ramos/met/met-7.0/scripts
> > touch ../out/ascii2nc/test_output_file
> >
> > The touch command just creates an empty file. If you get an error
> > message about permissions, then we've found the problem. I tried
> > running the exact same ascii2nc command but writing my output to
> > "/usr/bin"... where of course I don't have permission to write.
Here's
> the error message I got:
> >
> > cd met-7.0/scripts
> > ../src/tools/other/ascii2nc/ascii2nc
> > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > /usr/bin/sample_ascii_obs.txt -v 2
> >
> > ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> >
> > Do a long listing of that directory to see the permissions:
> > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> >
> > What do you see?
> >
> > Thanks,
> > John
> >
> >
> > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > This morning I downloaded the latest version of MET. The tests
still
> > > don't
> > > run:
> > >
> > > [ramos at maury2 met-7.0]$ make test
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Would you check if ../out/ascii2nc exists?
> > >
> > > The common cause are 1) when the output directory does not exist
or
> > > 2) the user does not have the proper privileges (write
privilege).
> > >
> > > Would you check these?
> > >
> > > ls ../out/ascii2nc/
> > > touch ../out/ascii2nc/test.txt
> > > rm ../out/ascii2nc/test.txt
> > >
> > > Cheers,
> > > Howard
> > >
> > >
> > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Queue: met_help
> > > > Subject: MET 6.1 bugfix
> > > > Owner: Nobody
> > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Status: new
> > > > Ticket <URL:
> > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > >
> > > >
> > > >
> > > > I built the MET 6.1, but the tests won't run. My user is
getting
> > > > the same error in trying to open the file for writing using
the
> tutorial:
> > > >
> > > >
> > > > cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File:
> > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > Ascii2NcConfig_default
> > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-6.1_bugfix]$
> > > >
> > > > I am guessing this is a common error.
> > > >
> > > > Greg Ramos
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 11:18:14 2018
I built the C and C++ versions of NetCDF. I also ran the "make check",
but I would be inclined to agree with you, that NetCDF might be the
problem.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 9:58 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
The number of executables built depends on the way the build was
configured. Options which require "extra" dependencies are disabled
by default. Here's an excerpt from the top-level README file:
To configure MET to install all of the available tools in the "bin"
subdirectory of your current directory, you would use the following
command:
./configure --prefix=`pwd` --enable-grib2 --enable-modis \
--enable-mode_graphics --enable-lidar2nc
So it's fine that you don't have all possible executables built.
That's unrelated to the test script issue.
My next guess would be an issue with the NetCDF library to which MET
is linking. Let's replicate the error by trying to write to some
other output directory, like your home directory:
/users/ramos/met/met-7.0/bin/ascii2nc \
/users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
~/sample_ascii.nc \ -v 2
I expect that this will fail with the same error message. If it does,
you could try recompiling NetCDF... and then recompiling MET. Perhaps
there's some incompatibility in how NetCDF and MET are built?
Here's some info about that:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
Thanks,
John
On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> These are the binaries that were built. It did not build all of the
> binaries listed in the README:
>
> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
total
> 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-xr-x
1
> ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos ramos
13M
> Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44 pcp_combine
> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x 1
> ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos ramos
> 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar 6
> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
plot_data_plane
> -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44 plot_point_obs -rwxr-xr-x
1
> ramos ramos 8.3M Mar 6 16:44 ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M
> Mar 6 16:44 madis2nc -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44
pb2nc
> -rwxr-xr-x 1 ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
> ramos ramos 8.9M Mar 6 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos
10M
> Mar 6 16:44 mtd -rwxr-xr-x 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr
> -rwxr-xr-x 1 ramos ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x
1
> ramos ramos 7.6M Mar 6 16:44 regrid_data_plane -rwxr-xr-x 1 ramos
> ramos 7.2M Mar 6 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos
3.6M
> Mar 6 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> tc_pairs drwxrwxr-x 2 ramos ramos 4.0K Mar 6 16:44 .
> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:12 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Hi Greg,
>
> This is John, I work with Howard. Well I'm glad to hear that the
> behavior is consistent across versions. If not, I would have been
confused.
>
> So the very first command that's being run by the test script is
> failing because it isn't able to create an output file.
>
> As Howard mentioned, the most likely cause for this type of error is
a
> permissions issue. So let's rule that out first. Please try
running
> the following commands:
> cd /users/ramos/met/met-7.0/scripts
> touch ../out/ascii2nc/test_output_file
>
> The touch command just creates an empty file. If you get an error
> message about permissions, then we've found the problem. I tried
> running the exact same ascii2nc command but writing my output to
> "/usr/bin"... where of course I don't have permission to write.
Here's the error message I got:
>
> cd met-7.0/scripts
> ../src/tools/other/ascii2nc/ascii2nc
> ../data/sample_obs/ascii/sample_ascii_obs.txt
> /usr/bin/sample_ascii_obs.txt -v 2
>
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>
> Do a long listing of that directory to see the permissions:
> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>
> What do you see?
>
> Thanks,
> John
>
>
> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This morning I downloaded the latest version of MET. The tests
still
> > don't
> > run:
> >
> > [ramos at maury2 met-7.0]$ make test
> > cd scripts; make
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
> > off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 8:23 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Would you check if ../out/ascii2nc exists?
> >
> > The common cause are 1) when the output directory does not exist
or
> > 2) the user does not have the proper privileges (write privilege).
> >
> > Would you check these?
> >
> > ls ../out/ascii2nc/
> > touch ../out/ascii2nc/test.txt
> > rm ../out/ascii2nc/test.txt
> >
> > Cheers,
> > Howard
> >
> >
> > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > Queue: met_help
> > > Subject: MET 6.1 bugfix
> > > Owner: Nobody
> > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > Status: new
> > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > >
> > >
> > >
> > > I built the MET 6.1, but the tests won't run. My user is getting
> > > the same error in trying to open the file for writing using the
tutorial:
> > >
> > >
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File:
> > > /software/depot/met-6.1_bugfix/share/met/config/
> > > Ascii2NcConfig_default
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-6.1_bugfix]$
> > >
> > > I am guessing this is a common error.
> > >
> > > Greg Ramos
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 11:23:46 2018
I know what the backslashes are for. What I meant was your example
included a bin directory that didn't exist.
Does NetCDF need to be built with HDF?
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 10:14 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
To clarify, those backslashes '\' are line continuation marks...
telling the shell that the command is continued on the next line.
Including them on one line like that causes the odd behavior of
creating an output file named "2". Removing them and running:
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2 Will
still result in an error... but the requested output file will be what
you expect.
So I'd try recompiling NetCDF using the same family/version number of
the compilers you're using to compile MET. Sorry I don't have an
easier solution for you.
Thanks,
John
On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I had to modify your example slightly, but it didn't work:
>
>
> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\
> ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
ERROR
> :
> ERROR : FileHandler::processFiles() -> can't open input ASCII file
"
> ~/ sample_ascii.nc" for reading ERROR :
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"2"
> for writing
> ERROR :
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:58 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> The number of executables built depends on the way the build was
> configured. Options which require "extra" dependencies are disabled
> by default. Here's an excerpt from the top-level README file:
>
> To configure MET to install all of the available tools in the
"bin"
> subdirectory of your current directory, you would use the
following
> command:
>
> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> --enable-mode_graphics --enable-lidar2nc
>
> So it's fine that you don't have all possible executables built.
> That's unrelated to the test script issue.
>
> My next guess would be an issue with the NetCDF library to which MET
> is linking. Let's replicate the error by trying to write to some
> other output directory, like your home directory:
>
> /users/ramos/met/met-7.0/bin/ascii2nc \
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\
> ~/ sample_ascii.nc \ -v 2
>
> I expect that this will fail with the same error message. If it
does,
> you could try recompiling NetCDF... and then recompiling MET.
Perhaps
> there's some incompatibility in how NetCDF and MET are built?
>
> Here's some info about that:
> https://dtcenter.org/met/users/support/online_tutorial/
> METv6.1/tutorial.php?name=compilation&category=req_libs
>
> Thanks,
> John
>
>
>
>
> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > These are the binaries that were built. It did not build all of
the
> > binaries listed in the README:
> >
> > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-xr-
x
> > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos ramos
> > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
pcp_combine
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x 1
> > ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos
ramos
> > 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar 6
> > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44 ascii2nc
> > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x 1
> > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos 8.4M
> > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
16:44
> > wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd -rwxr-
xr-x
> > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos ramos
> > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M Mar
6
> > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6 16:44
> > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
tc_dland
> > -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs drwxrwxr-x 2
ramos ramos 4.0K Mar 6 16:44 .
> > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:12 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Hi Greg,
> >
> > This is John, I work with Howard. Well I'm glad to hear that the
> > behavior is consistent across versions. If not, I would have been
> confused.
> >
> > So the very first command that's being run by the test script is
> > failing because it isn't able to create an output file.
> >
> > As Howard mentioned, the most likely cause for this type of error
is
> > a permissions issue. So let's rule that out first. Please try
> > running the following commands:
> > cd /users/ramos/met/met-7.0/scripts
> > touch ../out/ascii2nc/test_output_file
> >
> > The touch command just creates an empty file. If you get an error
> > message about permissions, then we've found the problem. I tried
> > running the exact same ascii2nc command but writing my output to
> > "/usr/bin"... where of course I don't have permission to write.
> > Here's
> the error message I got:
> >
> > cd met-7.0/scripts
> > ../src/tools/other/ascii2nc/ascii2nc
> > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > /usr/bin/sample_ascii_obs.txt -v 2
> >
> > ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> >
> > Do a long listing of that directory to see the permissions:
> > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> >
> > What do you see?
> >
> > Thanks,
> > John
> >
> >
> > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > This morning I downloaded the latest version of MET. The tests
> > > still don't
> > > run:
> > >
> > > [ramos at maury2 met-7.0]$ make test
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Would you check if ../out/ascii2nc exists?
> > >
> > > The common cause are 1) when the output directory does not exist
> > > or
> > > 2) the user does not have the proper privileges (write
privilege).
> > >
> > > Would you check these?
> > >
> > > ls ../out/ascii2nc/
> > > touch ../out/ascii2nc/test.txt
> > > rm ../out/ascii2nc/test.txt
> > >
> > > Cheers,
> > > Howard
> > >
> > >
> > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Queue: met_help
> > > > Subject: MET 6.1 bugfix
> > > > Owner: Nobody
> > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Status: new
> > > > Ticket <URL:
> > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > >
> > > >
> > > >
> > > > I built the MET 6.1, but the tests won't run. My user is
getting
> > > > the same error in trying to open the file for writing using
the
> tutorial:
> > > >
> > > >
> > > > cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File:
> > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > Ascii2NcConfig_default
> > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-6.1_bugfix]$
> > > >
> > > > I am guessing this is a common error.
> > > >
> > > > Greg Ramos
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 13:55:13 2018
I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
*** Running ASCII2NC to reformat ASCII point observations into NetCDF
***
../src/tools/other/ascii2nc/ascii2nc \
../data/sample_obs/ascii/sample_ascii_obs.txt \
../out/ascii2nc/sample_ascii.nc \
-v 2
DEBUG 1: Config File: /software/depot/met-
7.0/share/met/config/Ascii2NcConfig_default
DEBUG 2: Rejected 0 observations off the masking grid.
DEBUG 2: Rejected 0 observations outside the masking polyline.
DEBUG 2: Rejected 0 observations not matched with station ID's.
DEBUG 2: Processing observations for 5 headers.
DEBUG 1: Creating NetCDF Observation file:
../out/ascii2nc/sample_ascii.nc
ERROR :
ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
make[1]: *** [ascii2nc] Error 1
make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
make: *** [test] Error 2
[ramos at maury2 met-7.0]$
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 10:14 AM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
To clarify, those backslashes '\' are line continuation marks...
telling the shell that the command is continued on the next line.
Including them on one line like that causes the odd behavior of
creating an output file named "2". Removing them and running:
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2 Will
still result in an error... but the requested output file will be what
you expect.
So I'd try recompiling NetCDF using the same family/version number of
the compilers you're using to compile MET. Sorry I don't have an
easier solution for you.
Thanks,
John
On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I had to modify your example slightly, but it didn't work:
>
>
> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\
> ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
ERROR
> :
> ERROR : FileHandler::processFiles() -> can't open input ASCII file
"
> ~/ sample_ascii.nc" for reading ERROR :
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
"2"
> for writing
> ERROR :
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 9:58 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> The number of executables built depends on the way the build was
> configured. Options which require "extra" dependencies are disabled
> by default. Here's an excerpt from the top-level README file:
>
> To configure MET to install all of the available tools in the
"bin"
> subdirectory of your current directory, you would use the
following
> command:
>
> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> --enable-mode_graphics --enable-lidar2nc
>
> So it's fine that you don't have all possible executables built.
> That's unrelated to the test script issue.
>
> My next guess would be an issue with the NetCDF library to which MET
> is linking. Let's replicate the error by trying to write to some
> other output directory, like your home directory:
>
> /users/ramos/met/met-7.0/bin/ascii2nc \
> /users/ramos/met/met-7.0/data/sample_obs/ascii/sample_ascii_obs.txt
\
> ~/ sample_ascii.nc \ -v 2
>
> I expect that this will fail with the same error message. If it
does,
> you could try recompiling NetCDF... and then recompiling MET.
Perhaps
> there's some incompatibility in how NetCDF and MET are built?
>
> Here's some info about that:
> https://dtcenter.org/met/users/support/online_tutorial/
> METv6.1/tutorial.php?name=compilation&category=req_libs
>
> Thanks,
> John
>
>
>
>
> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > These are the binaries that were built. It did not build all of
the
> > binaries listed in the README:
> >
> > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-xr-
x
> > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos ramos
> > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
pcp_combine
> > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x 1
> > ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos
ramos
> > 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar 6
> > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44 ascii2nc
> > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x 1
> > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos 8.4M
> > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
16:44
> > wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd -rwxr-
xr-x
> > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos ramos
> > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M Mar
6
> > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6 16:44
> > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
tc_dland
> > -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs drwxrwxr-x 2
ramos ramos 4.0K Mar 6 16:44 .
> > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:12 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Hi Greg,
> >
> > This is John, I work with Howard. Well I'm glad to hear that the
> > behavior is consistent across versions. If not, I would have been
> confused.
> >
> > So the very first command that's being run by the test script is
> > failing because it isn't able to create an output file.
> >
> > As Howard mentioned, the most likely cause for this type of error
is
> > a permissions issue. So let's rule that out first. Please try
> > running the following commands:
> > cd /users/ramos/met/met-7.0/scripts
> > touch ../out/ascii2nc/test_output_file
> >
> > The touch command just creates an empty file. If you get an error
> > message about permissions, then we've found the problem. I tried
> > running the exact same ascii2nc command but writing my output to
> > "/usr/bin"... where of course I don't have permission to write.
> > Here's
> the error message I got:
> >
> > cd met-7.0/scripts
> > ../src/tools/other/ascii2nc/ascii2nc
> > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > /usr/bin/sample_ascii_obs.txt -v 2
> >
> > ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> >
> > Do a long listing of that directory to see the permissions:
> > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> >
> > What do you see?
> >
> > Thanks,
> > John
> >
> >
> > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > This morning I downloaded the latest version of MET. The tests
> > > still don't
> > > run:
> > >
> > > [ramos at maury2 met-7.0]$ make test
> > > cd scripts; make
> > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Would you check if ../out/ascii2nc exists?
> > >
> > > The common cause are 1) when the output directory does not exist
> > > or
> > > 2) the user does not have the proper privileges (write
privilege).
> > >
> > > Would you check these?
> > >
> > > ls ../out/ascii2nc/
> > > touch ../out/ascii2nc/test.txt
> > > rm ../out/ascii2nc/test.txt
> > >
> > > Cheers,
> > > Howard
> > >
> > >
> > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Queue: met_help
> > > > Subject: MET 6.1 bugfix
> > > > Owner: Nobody
> > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > Status: new
> > > > Ticket <URL:
> > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > >
> > > >
> > > >
> > > > I built the MET 6.1, but the tests won't run. My user is
getting
> > > > the same error in trying to open the file for writing using
the
> tutorial:
> > > >
> > > >
> > > > cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File:
> > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > Ascii2NcConfig_default
> > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-
6.1_bugfix/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-6.1_bugfix]$
> > > >
> > > > I am guessing this is a common error.
> > > >
> > > > Greg Ramos
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 14:17:50 2018
Greg,
Another thing to test would be the runtime linker. You set the
MET_NETCDF
environment variable to tell MET where to find the header and library
files
for compilation. By default, MET is compiled using dynamic linking.
I
suppose its possible that at runtime the MET executables are linking
to
some other version of NetCDF. You could set the LD_LIBRARY_PATH
environment variable to define the search path for the linker. Let's
try
this:
# Using c-shell syntax here. And MET_NETCDF is the top-level directory
for
your NetCDF compilations
setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
Any difference in results?
Thanks,
John
On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
>
>
>
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF ***
> ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default
> DEBUG 2: Rejected 0 observations off the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
../out/ascii2nc/sample_ascii.nc
> ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 10:14 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> To clarify, those backslashes '\' are line continuation marks...
telling
> the shell that the command is continued on the next line. Including
them
> on one line like that causes the odd behavior of creating an output
file
> named "2". Removing them and running:
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2 Will
still
> result in an error... but the requested output file will be what you
expect.
>
> So I'd try recompiling NetCDF using the same family/version number
of the
> compilers you're using to compile MET. Sorry I don't have an easier
> solution for you.
>
> Thanks,
> John
>
>
>
> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I had to modify your example slightly, but it didn't work:
> >
> >
> > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
> > ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
ERROR
> > :
> > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > ~/ sample_ascii.nc" for reading ERROR :
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file "2"
> > for writing
> > ERROR :
> > [ramos at maury2 met-7.0]$
> >
> > Greg
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:58 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > The number of executables built depends on the way the build was
> > configured. Options which require "extra" dependencies are
disabled
> > by default. Here's an excerpt from the top-level README file:
> >
> > To configure MET to install all of the available tools in the
"bin"
> > subdirectory of your current directory, you would use the
following
> > command:
> >
> > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > --enable-mode_graphics --enable-lidar2nc
> >
> > So it's fine that you don't have all possible executables built.
> > That's unrelated to the test script issue.
> >
> > My next guess would be an issue with the NetCDF library to which
MET
> > is linking. Let's replicate the error by trying to write to some
> > other output directory, like your home directory:
> >
> > /users/ramos/met/met-7.0/bin/ascii2nc \
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt \
> > ~/ sample_ascii.nc \ -v 2
> >
> > I expect that this will fail with the same error message. If it
does,
> > you could try recompiling NetCDF... and then recompiling MET.
Perhaps
> > there's some incompatibility in how NetCDF and MET are built?
> >
> > Here's some info about that:
> > https://dtcenter.org/met/users/support/online_tutorial/
> > METv6.1/tutorial.php?name=compilation&category=req_libs
> >
> > Thanks,
> > John
> >
> >
> >
> >
> > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > These are the binaries that were built. It did not build all of
the
> > > binaries listed in the README:
> > >
> > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat -rwxr-
xr-x
> > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
ramos
> > > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
pcp_combine
> > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 point_stat -rwxr-xr-x
1
> > > ramos ramos 9.7M Mar 6 16:44 stat_analysis -rwxr-xr-x 1 ramos
ramos
> > > 12M Mar 6 16:44 wavelet_stat -rwxr-xr-x 1 ramos ramos 9.8M Mar
6
> > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
ascii2nc
> > > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x 1
> > > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos
8.4M
> > > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
16:44
> > > wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd -rwxr-
xr-x
> > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
> > > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar 6
> > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
16:44
> > > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
tc_dland
> > > -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs drwxrwxr-x 2
ramos
> ramos 4.0K Mar 6 16:44 .
> > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Hi Greg,
> > >
> > > This is John, I work with Howard. Well I'm glad to hear that
the
> > > behavior is consistent across versions. If not, I would have
been
> > confused.
> > >
> > > So the very first command that's being run by the test script is
> > > failing because it isn't able to create an output file.
> > >
> > > As Howard mentioned, the most likely cause for this type of
error is
> > > a permissions issue. So let's rule that out first. Please try
> > > running the following commands:
> > > cd /users/ramos/met/met-7.0/scripts
> > > touch ../out/ascii2nc/test_output_file
> > >
> > > The touch command just creates an empty file. If you get an
error
> > > message about permissions, then we've found the problem. I
tried
> > > running the exact same ascii2nc command but writing my output to
> > > "/usr/bin"... where of course I don't have permission to write.
> > > Here's
> > the error message I got:
> > >
> > > cd met-7.0/scripts
> > > ../src/tools/other/ascii2nc/ascii2nc
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > /usr/bin/sample_ascii_obs.txt -v 2
> > >
> > > ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > > "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > >
> > > Do a long listing of that directory to see the permissions:
> > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >
> > > What do you see?
> > >
> > > Thanks,
> > > John
> > >
> > >
> > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > This morning I downloaded the latest version of MET. The tests
> > > > still don't
> > > > run:
> > > >
> > > > [ramos at maury2 met-7.0]$ make test
> > > > cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > > off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Would you check if ../out/ascii2nc exists?
> > > >
> > > > The common cause are 1) when the output directory does not
exist
> > > > or
> > > > 2) the user does not have the proper privileges (write
privilege).
> > > >
> > > > Would you check these?
> > > >
> > > > ls ../out/ascii2nc/
> > > > touch ../out/ascii2nc/test.txt
> > > > rm ../out/ascii2nc/test.txt
> > > >
> > > > Cheers,
> > > > Howard
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Queue: met_help
> > > > > Subject: MET 6.1 bugfix
> > > > > Owner: Nobody
> > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Status: new
> > > > > Ticket <URL:
> > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > >
> > > > > I built the MET 6.1, but the tests won't run. My user is
getting
> > > > > the same error in trying to open the file for writing using
the
> > tutorial:
> > > > >
> > > > >
> > > > > cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File:
> > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > Ascii2NcConfig_default
> > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > >
> > > > > I am guessing this is a common error.
> > > > >
> > > > > Greg Ramos
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 14:25:30 2018
We use modules here. LD_LIBRAY_PATH is set via the loading of the
modules. You do bring up a good point. But regardless of that, should
'make test' use what is in the build tree? I figure if I can't get
'make test' to work, the user won't be able to get her tests to run.
We have someone coming out to give a seminar on MET. I have been
building MET since version 5.0, first time I have had issues.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 1:18 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Another thing to test would be the runtime linker. You set the
MET_NETCDF environment variable to tell MET where to find the header
and library files for compilation. By default, MET is compiled using
dynamic linking. I suppose its possible that at runtime the MET
executables are linking to some other version of NetCDF. You could
set the LD_LIBRARY_PATH environment variable to define the search path
for the linker. Let's try
this:
# Using c-shell syntax here. And MET_NETCDF is the top-level directory
for your NetCDF compilations
setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
Any difference in results?
Thanks,
John
On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
>
>
>
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 10:14 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> To clarify, those backslashes '\' are line continuation marks...
> telling the shell that the command is continued on the next line.
> Including them on one line like that causes the odd behavior of
> creating an output file named "2". Removing them and running:
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2 Will
> still result in an error... but the requested output file will be
what you expect.
>
> So I'd try recompiling NetCDF using the same family/version number
of
> the compilers you're using to compile MET. Sorry I don't have an
> easier solution for you.
>
> Thanks,
> John
>
>
>
> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I had to modify your example slightly, but it didn't work:
> >
> >
> > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
> > ERROR
> > :
> > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > ~/ sample_ascii.nc" for reading ERROR :
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file "2"
> > for writing
> > ERROR :
> > [ramos at maury2 met-7.0]$
> >
> > Greg
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:58 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > The number of executables built depends on the way the build was
> > configured. Options which require "extra" dependencies are
disabled
> > by default. Here's an excerpt from the top-level README file:
> >
> > To configure MET to install all of the available tools in the
"bin"
> > subdirectory of your current directory, you would use the
> > following
> > command:
> >
> > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > --enable-mode_graphics --enable-lidar2nc
> >
> > So it's fine that you don't have all possible executables built.
> > That's unrelated to the test script issue.
> >
> > My next guess would be an issue with the NetCDF library to which
MET
> > is linking. Let's replicate the error by trying to write to some
> > other output directory, like your home directory:
> >
> > /users/ramos/met/met-7.0/bin/ascii2nc \
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > \ ~/ sample_ascii.nc \ -v 2
> >
> > I expect that this will fail with the same error message. If it
> > does, you could try recompiling NetCDF... and then recompiling
MET.
> > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> >
> > Here's some info about that:
> > https://dtcenter.org/met/users/support/online_tutorial/
> > METv6.1/tutorial.php?name=compilation&category=req_libs
> >
> > Thanks,
> > John
> >
> >
> >
> >
> > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
> > Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > These are the binaries that were built. It did not build all of
> > > the binaries listed in the README:
> > >
> > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > -rwxr-xr-x
> > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
ramos
> > > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
> > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
> > > -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
> > > -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-xr-
x
> > > 1 ramos ramos 9.8M Mar 6
> > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
ascii2nc
> > > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x 1
> > > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos
8.4M
> > > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
> > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
> > > -rwxr-xr-x
> > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
> > > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
> > > 6
> > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
16:44
> > > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
> > > tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs
> > > drwxrwxr-x 2 ramos
> ramos 4.0K Mar 6 16:44 .
> > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Hi Greg,
> > >
> > > This is John, I work with Howard. Well I'm glad to hear that
the
> > > behavior is consistent across versions. If not, I would have
been
> > confused.
> > >
> > > So the very first command that's being run by the test script is
> > > failing because it isn't able to create an output file.
> > >
> > > As Howard mentioned, the most likely cause for this type of
error
> > > is a permissions issue. So let's rule that out first. Please
try
> > > running the following commands:
> > > cd /users/ramos/met/met-7.0/scripts
> > > touch ../out/ascii2nc/test_output_file
> > >
> > > The touch command just creates an empty file. If you get an
error
> > > message about permissions, then we've found the problem. I
tried
> > > running the exact same ascii2nc command but writing my output to
> > > "/usr/bin"... where of course I don't have permission to write.
> > > Here's
> > the error message I got:
> > >
> > > cd met-7.0/scripts
> > > ../src/tools/other/ascii2nc/ascii2nc
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > /usr/bin/sample_ascii_obs.txt -v 2
> > >
> > > ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > >
> > > Do a long listing of that directory to see the permissions:
> > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >
> > > What do you see?
> > >
> > > Thanks,
> > > John
> > >
> > >
> > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > This morning I downloaded the latest version of MET. The tests
> > > > still don't
> > > > run:
> > > >
> > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Would you check if ../out/ascii2nc exists?
> > > >
> > > > The common cause are 1) when the output directory does not
exist
> > > > or
> > > > 2) the user does not have the proper privileges (write
privilege).
> > > >
> > > > Would you check these?
> > > >
> > > > ls ../out/ascii2nc/
> > > > touch ../out/ascii2nc/test.txt
> > > > rm ../out/ascii2nc/test.txt
> > > >
> > > > Cheers,
> > > > Howard
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Queue: met_help
> > > > > Subject: MET 6.1 bugfix
> > > > > Owner: Nobody
> > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Status: new
> > > > > Ticket <URL:
> > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > >
> > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > getting the same error in trying to open the file for
writing
> > > > > using the
> > tutorial:
> > > > >
> > > > >
> > > > > cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File:
> > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > Ascii2NcConfig_default
> > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > >
> > > > > I am guessing this is a common error.
> > > > >
> > > > > Greg Ramos
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 14:36:08 2018
BTW, here are my build settings:
[ramos at maury2 met-7.0]$ env |grep MET
MET_GSL=/software/depot/gsl-1.16
MET_ROOT=/software/depot/met-6.1_bugfix
MET_NETCDF=/software/depot/netcdf-3.6.2018
MET_NETCDFINC=
MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
[ramos at maury2 met-7.0]$
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 1:18 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Another thing to test would be the runtime linker. You set the
MET_NETCDF environment variable to tell MET where to find the header
and library files for compilation. By default, MET is compiled using
dynamic linking. I suppose its possible that at runtime the MET
executables are linking to some other version of NetCDF. You could
set the LD_LIBRARY_PATH environment variable to define the search path
for the linker. Let's try
this:
# Using c-shell syntax here. And MET_NETCDF is the top-level directory
for your NetCDF compilations
setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
/software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
Any difference in results?
Thanks,
John
On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
>
>
>
> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>
> *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> *** ../src/tools/other/ascii2nc/ascii2nc \
> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> ../out/ascii2nc/sample_ascii.nc \
> -v 2
> DEBUG 1: Config File: /software/depot/met-7.0/share/
> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> the masking grid.
> DEBUG 2: Rejected 0 observations outside the masking polyline.
> DEBUG 2: Rejected 0 observations not matched with station ID's.
> DEBUG 2: Processing observations for 5 headers.
> DEBUG 1: Creating NetCDF Observation file:
> ../out/ascii2nc/sample_ascii.nc ERROR :
> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF file
> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> make[1]: *** [ascii2nc] Error 1
> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> make: *** [test] Error 2
> [ramos at maury2 met-7.0]$
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 10:14 AM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> To clarify, those backslashes '\' are line continuation marks...
> telling the shell that the command is continued on the next line.
> Including them on one line like that causes the odd behavior of
> creating an output file named "2". Removing them and running:
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2 Will
> still result in an error... but the requested output file will be
what you expect.
>
> So I'd try recompiling NetCDF using the same family/version number
of
> the compilers you're using to compile MET. Sorry I don't have an
> easier solution for you.
>
> Thanks,
> John
>
>
>
> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I had to modify your example slightly, but it didn't work:
> >
> >
> > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
> > ERROR
> > :
> > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > ~/ sample_ascii.nc" for reading ERROR :
> > DEBUG 2: Rejected 0 observations off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file "2"
> > for writing
> > ERROR :
> > [ramos at maury2 met-7.0]$
> >
> > Greg
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 9:58 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > The number of executables built depends on the way the build was
> > configured. Options which require "extra" dependencies are
disabled
> > by default. Here's an excerpt from the top-level README file:
> >
> > To configure MET to install all of the available tools in the
"bin"
> > subdirectory of your current directory, you would use the
> > following
> > command:
> >
> > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > --enable-mode_graphics --enable-lidar2nc
> >
> > So it's fine that you don't have all possible executables built.
> > That's unrelated to the test script issue.
> >
> > My next guess would be an issue with the NetCDF library to which
MET
> > is linking. Let's replicate the error by trying to write to some
> > other output directory, like your home directory:
> >
> > /users/ramos/met/met-7.0/bin/ascii2nc \
> > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > \ ~/ sample_ascii.nc \ -v 2
> >
> > I expect that this will fail with the same error message. If it
> > does, you could try recompiling NetCDF... and then recompiling
MET.
> > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> >
> > Here's some info about that:
> > https://dtcenter.org/met/users/support/online_tutorial/
> > METv6.1/tutorial.php?name=compilation&category=req_libs
> >
> > Thanks,
> > John
> >
> >
> >
> >
> > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
> > Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > These are the binaries that were built. It did not build all of
> > > the binaries listed in the README:
> > >
> > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
> > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > -rwxr-xr-x
> > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
ramos
> > > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6 16:44
> > > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
> > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
> > > -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
> > > -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-xr-
x
> > > 1 ramos ramos 9.8M Mar 6
> > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
ascii2nc
> > > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x 1
> > > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos
8.4M
> > > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
> > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
> > > -rwxr-xr-x
> > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
> > > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
> > > 6
> > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
16:44
> > > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
> > > tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs
> > > drwxrwxr-x 2 ramos
> ramos 4.0K Mar 6 16:44 .
> > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Hi Greg,
> > >
> > > This is John, I work with Howard. Well I'm glad to hear that
the
> > > behavior is consistent across versions. If not, I would have
been
> > confused.
> > >
> > > So the very first command that's being run by the test script is
> > > failing because it isn't able to create an output file.
> > >
> > > As Howard mentioned, the most likely cause for this type of
error
> > > is a permissions issue. So let's rule that out first. Please
try
> > > running the following commands:
> > > cd /users/ramos/met/met-7.0/scripts
> > > touch ../out/ascii2nc/test_output_file
> > >
> > > The touch command just creates an empty file. If you get an
error
> > > message about permissions, then we've found the problem. I
tried
> > > running the exact same ascii2nc command but writing my output to
> > > "/usr/bin"... where of course I don't have permission to write.
> > > Here's
> > the error message I got:
> > >
> > > cd met-7.0/scripts
> > > ../src/tools/other/ascii2nc/ascii2nc
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > /usr/bin/sample_ascii_obs.txt -v 2
> > >
> > > ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > >
> > > Do a long listing of that directory to see the permissions:
> > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >
> > > What do you see?
> > >
> > > Thanks,
> > > John
> > >
> > >
> > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > This morning I downloaded the latest version of MET. The tests
> > > > still don't
> > > > run:
> > > >
> > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > > >
> > > > *** Running ASCII2NC to reformat ASCII point observations into
> > > > NetCDF
> > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > ../out/ascii2nc/sample_ascii.nc \
> > > > -v 2
> > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file:
> > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > make[1]: *** [ascii2nc] Error 1
> > > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > > make: *** [test] Error 2
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Would you check if ../out/ascii2nc exists?
> > > >
> > > > The common cause are 1) when the output directory does not
exist
> > > > or
> > > > 2) the user does not have the proper privileges (write
privilege).
> > > >
> > > > Would you check these?
> > > >
> > > > ls ../out/ascii2nc/
> > > > touch ../out/ascii2nc/test.txt
> > > > rm ../out/ascii2nc/test.txt
> > > >
> > > > Cheers,
> > > > Howard
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Queue: met_help
> > > > > Subject: MET 6.1 bugfix
> > > > > Owner: Nobody
> > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > Status: new
> > > > > Ticket <URL:
> > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > >
> > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > getting the same error in trying to open the file for
writing
> > > > > using the
> > tutorial:
> > > > >
> > > > >
> > > > > cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File:
> > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > Ascii2NcConfig_default
> > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> bugfix/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > >
> > > > > I am guessing this is a common error.
> > > > >
> > > > > Greg Ramos
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 14:44:42 2018
Greg,
Ah, OK, I see that you're using NetCDF version 3.6. That's likely the
problem. As of MET version 6.0 we switched from NetCDF version 3 to
NetCDF
version 4:
https://dtcenter.org/met/users/support/release_notes/METv6.0_release_notes.php
Please try compiling against NetCDF4 instead of 3:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
I'm surprised MET compiled without error and that the problem showed
up at
runtime.
John
On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> BTW, here are my build settings:
>
> [ramos at maury2 met-7.0]$ env |grep MET
> MET_GSL=/software/depot/gsl-1.16
> MET_ROOT=/software/depot/met-6.1_bugfix
> MET_NETCDF=/software/depot/netcdf-3.6.2018
> MET_NETCDFINC=
> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 1:18 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Another thing to test would be the runtime linker. You set the
MET_NETCDF
> environment variable to tell MET where to find the header and
library files
> for compilation. By default, MET is compiled using dynamic linking.
I
> suppose its possible that at runtime the MET executables are linking
to
> some other version of NetCDF. You could set the LD_LIBRARY_PATH
> environment variable to define the search path for the linker.
Let's try
> this:
>
> # Using c-shell syntax here. And MET_NETCDF is the top-level
directory for
> your NetCDF compilations
>
> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>
> Any difference in results?
>
> Thanks,
> John
>
> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
> >
> >
> >
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
off
> > the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 10:14 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > To clarify, those backslashes '\' are line continuation marks...
> > telling the shell that the command is continued on the next line.
> > Including them on one line like that causes the odd behavior of
> > creating an output file named "2". Removing them and running:
> > /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> > sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
Will
> > still result in an error... but the requested output file will be
what
> you expect.
> >
> > So I'd try recompiling NetCDF using the same family/version number
of
> > the compilers you're using to compile MET. Sorry I don't have an
> > easier solution for you.
> >
> > Thanks,
> > John
> >
> >
> >
> > On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > I had to modify your example slightly, but it didn't work:
> > >
> > >
> > > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > > \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
> > > ERROR
> > > :
> > > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > > ~/ sample_ascii.nc" for reading ERROR :
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> "2"
> > > for writing
> > > ERROR :
> > > [ramos at maury2 met-7.0]$
> > >
> > > Greg
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:58 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Greg,
> > >
> > > The number of executables built depends on the way the build was
> > > configured. Options which require "extra" dependencies are
disabled
> > > by default. Here's an excerpt from the top-level README file:
> > >
> > > To configure MET to install all of the available tools in the
"bin"
> > > subdirectory of your current directory, you would use the
> > > following
> > > command:
> > >
> > > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > > --enable-mode_graphics --enable-lidar2nc
> > >
> > > So it's fine that you don't have all possible executables built.
> > > That's unrelated to the test script issue.
> > >
> > > My next guess would be an issue with the NetCDF library to which
MET
> > > is linking. Let's replicate the error by trying to write to
some
> > > other output directory, like your home directory:
> > >
> > > /users/ramos/met/met-7.0/bin/ascii2nc \
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.txt
> > > \ ~/ sample_ascii.nc \ -v 2
> > >
> > > I expect that this will fail with the same error message. If it
> > > does, you could try recompiling NetCDF... and then recompiling
MET.
> > > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> > >
> > > Here's some info about that:
> > > https://dtcenter.org/met/users/support/online_tutorial/
> > > METv6.1/tutorial.php?name=compilation&category=req_libs
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> > > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > These are the binaries that were built. It did not build all
of
> > > > the binaries listed in the README:
> > > >
> > > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> > > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
ramos
> > > > 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar 6
16:44
> > > > mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
> > > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
> > > > -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
> > > > -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-
xr-x
> > > > 1 ramos ramos 9.8M Mar 6
> > > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
> > > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
ascii2nc
> > > > -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc -rwxr-xr-x
1
> > > > ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1 ramos ramos
8.4M
> > > > Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos ramos 8.9M Mar 6
> > > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
> > > > 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
> > > > 6
> > > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
16:44
> > > > shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6 16:44
> > > > tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44 tc_pairs
> > > > drwxrwxr-x 2 ramos
> > ramos 4.0K Mar 6 16:44 .
> > > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Hi Greg,
> > > >
> > > > This is John, I work with Howard. Well I'm glad to hear that
the
> > > > behavior is consistent across versions. If not, I would have
been
> > > confused.
> > > >
> > > > So the very first command that's being run by the test script
is
> > > > failing because it isn't able to create an output file.
> > > >
> > > > As Howard mentioned, the most likely cause for this type of
error
> > > > is a permissions issue. So let's rule that out first. Please
try
> > > > running the following commands:
> > > > cd /users/ramos/met/met-7.0/scripts
> > > > touch ../out/ascii2nc/test_output_file
> > > >
> > > > The touch command just creates an empty file. If you get an
error
> > > > message about permissions, then we've found the problem. I
tried
> > > > running the exact same ascii2nc command but writing my output
to
> > > > "/usr/bin"... where of course I don't have permission to
write.
> > > > Here's
> > > the error message I got:
> > > >
> > > > cd met-7.0/scripts
> > > > ../src/tools/other/ascii2nc/ascii2nc
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > > /usr/bin/sample_ascii_obs.txt -v 2
> > > >
> > > > ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > > >
> > > > Do a long listing of that directory to see the permissions:
> > > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > > >
> > > > What do you see?
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > > > >
> > > > > This morning I downloaded the latest version of MET. The
tests
> > > > > still don't
> > > > > run:
> > > > >
> > > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > > observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-7.0]$
> > > > >
> > > > > -----Original Message-----
> > > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > >
> > > > > Would you check if ../out/ascii2nc exists?
> > > > >
> > > > > The common cause are 1) when the output directory does not
exist
> > > > > or
> > > > > 2) the user does not have the proper privileges (write
privilege).
> > > > >
> > > > > Would you check these?
> > > > >
> > > > > ls ../out/ascii2nc/
> > > > > touch ../out/ascii2nc/test.txt
> > > > > rm ../out/ascii2nc/test.txt
> > > > >
> > > > > Cheers,
> > > > > Howard
> > > > >
> > > > >
> > > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
> > > > > Code
> > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Queue: met_help
> > > > > > Subject: MET 6.1 bugfix
> > > > > > Owner: Nobody
> > > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Status: new
> > > > > > Ticket <URL:
> > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > >
> > > > > >
> > > > > >
> > > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > > getting the same error in trying to open the file for
writing
> > > > > > using the
> > > tutorial:
> > > > > >
> > > > > >
> > > > > > cd scripts; make
> > > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > > bugfix/scripts'
> > > > > >
> > > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > > NetCDF
> > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > -v 2
> > > > > > DEBUG 1: Config File:
> > > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > > Ascii2NcConfig_default
> > > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > > > make: *** [test] Error 2
> > > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > > >
> > > > > > I am guessing this is a common error.
> > > > > >
> > > > > > Greg Ramos
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 14:49:17 2018
No, I labeled the install directory with today's date, after you
suggested I rebuild NetCDF
drwxr-xr-x 24 ramos ramos 8.0K Mar 6 19:22 netcdf-4.6.0
drwxr-xr-x 6 ramos ramos 4.0K Mar 6 20:01 netcdf-cxx4-4.3.0
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 1:45 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Ah, OK, I see that you're using NetCDF version 3.6. That's likely the
problem. As of MET version 6.0 we switched from NetCDF version 3 to
NetCDF version 4:
https://dtcenter.org/met/users/support/release_notes/METv6.0_release_notes.php
Please try compiling against NetCDF4 instead of 3:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
I'm surprised MET compiled without error and that the problem showed
up at runtime.
John
On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> BTW, here are my build settings:
>
> [ramos at maury2 met-7.0]$ env |grep MET
> MET_GSL=/software/depot/gsl-1.16
> MET_ROOT=/software/depot/met-6.1_bugfix
> MET_NETCDF=/software/depot/netcdf-3.6.2018
> MET_NETCDFINC=
> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 1:18 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Another thing to test would be the runtime linker. You set the
> MET_NETCDF environment variable to tell MET where to find the header
> and library files for compilation. By default, MET is compiled
using
> dynamic linking. I suppose its possible that at runtime the MET
> executables are linking to some other version of NetCDF. You could
> set the LD_LIBRARY_PATH environment variable to define the search
path
> for the linker. Let's try
> this:
>
> # Using c-shell syntax here. And MET_NETCDF is the top-level
directory
> for your NetCDF compilations
>
> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>
> Any difference in results?
>
> Thanks,
> John
>
> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
> >
> >
> >
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
> > off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 10:14 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > To clarify, those backslashes '\' are line continuation marks...
> > telling the shell that the command is continued on the next line.
> > Including them on one line like that causes the odd behavior of
> > creating an output file named "2". Removing them and running:
> > /software/depot/met-7.0/bin/ascii2nc
> > /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt
> > ~/sample_ascii.nc -v 2 Will still result in an error... but the
> > requested output file will be what
> you expect.
> >
> > So I'd try recompiling NetCDF using the same family/version number
> > of the compilers you're using to compile MET. Sorry I don't have
an
> > easier solution for you.
> >
> > Thanks,
> > John
> >
> >
> >
> > On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
> > Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > I had to modify your example slightly, but it didn't work:
> > >
> > >
> > > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
> > > ERROR
> > > :
> > > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > > ~/ sample_ascii.nc" for reading ERROR :
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file
> "2"
> > > for writing
> > > ERROR :
> > > [ramos at maury2 met-7.0]$
> > >
> > > Greg
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:58 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Greg,
> > >
> > > The number of executables built depends on the way the build was
> > > configured. Options which require "extra" dependencies are
> > > disabled by default. Here's an excerpt from the top-level
README file:
> > >
> > > To configure MET to install all of the available tools in the
"bin"
> > > subdirectory of your current directory, you would use the
> > > following
> > > command:
> > >
> > > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > > --enable-mode_graphics --enable-lidar2nc
> > >
> > > So it's fine that you don't have all possible executables built.
> > > That's unrelated to the test script issue.
> > >
> > > My next guess would be an issue with the NetCDF library to which
> > > MET is linking. Let's replicate the error by trying to write to
> > > some other output directory, like your home directory:
> > >
> > > /users/ramos/met/met-7.0/bin/ascii2nc \
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > t
> > > \ ~/ sample_ascii.nc \ -v 2
> > >
> > > I expect that this will fail with the same error message. If it
> > > does, you could try recompiling NetCDF... and then recompiling
MET.
> > > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> > >
> > > Here's some info about that:
> > > https://dtcenter.org/met/users/support/online_tutorial/
> > > METv6.1/tutorial.php?name=compilation&category=req_libs
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> > > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > These are the binaries that were built. It did not build all
of
> > > > the binaries listed in the README:
> > > >
> > > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> > > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> > > > ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
> > > > 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
> > > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> > > > point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > > > stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > > > wavelet_stat -rwxr-xr-x
> > > > 1 ramos ramos 9.8M Mar 6
> > > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
> > > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > > > ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> > > > -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
> > > > ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
> > > > ramos 8.9M Mar 6
> > > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > > > ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos
> > > > 7.6M Mar
> > > > 6
> > > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> > > > 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > > > 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > > > tc_pairs drwxrwxr-x 2 ramos
> > ramos 4.0K Mar 6 16:44 .
> > > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Hi Greg,
> > > >
> > > > This is John, I work with Howard. Well I'm glad to hear that
> > > > the behavior is consistent across versions. If not, I would
> > > > have been
> > > confused.
> > > >
> > > > So the very first command that's being run by the test script
is
> > > > failing because it isn't able to create an output file.
> > > >
> > > > As Howard mentioned, the most likely cause for this type of
> > > > error is a permissions issue. So let's rule that out first.
> > > > Please try running the following commands:
> > > > cd /users/ramos/met/met-7.0/scripts
> > > > touch ../out/ascii2nc/test_output_file
> > > >
> > > > The touch command just creates an empty file. If you get an
> > > > error message about permissions, then we've found the problem.
> > > > I tried running the exact same ascii2nc command but writing my
> > > > output to "/usr/bin"... where of course I don't have
permission to write.
> > > > Here's
> > > the error message I got:
> > > >
> > > > cd met-7.0/scripts
> > > > ../src/tools/other/ascii2nc/ascii2nc
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > > /usr/bin/sample_ascii_obs.txt -v 2
> > > >
> > > > ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > > >
> > > > Do a long listing of that directory to see the permissions:
> > > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > > >
> > > > What do you see?
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > > This morning I downloaded the latest version of MET. The
tests
> > > > > still don't
> > > > > run:
> > > > >
> > > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > > observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-7.0]$
> > > > >
> > > > > -----Original Message-----
> > > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > >
> > > > > Would you check if ../out/ascii2nc exists?
> > > > >
> > > > > The common cause are 1) when the output directory does not
> > > > > exist or
> > > > > 2) the user does not have the proper privileges (write
privilege).
> > > > >
> > > > > Would you check these?
> > > > >
> > > > > ls ../out/ascii2nc/
> > > > > touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
> > > > >
> > > > > Cheers,
> > > > > Howard
> > > > >
> > > > >
> > > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > > > > Contractor, Code
> > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Queue: met_help
> > > > > > Subject: MET 6.1 bugfix
> > > > > > Owner: Nobody
> > > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Status: new
> > > > > > Ticket <URL:
> > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > >
> > > > > >
> > > > > >
> > > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > > getting the same error in trying to open the file for
> > > > > > writing using the
> > > tutorial:
> > > > > >
> > > > > >
> > > > > > cd scripts; make
> > > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > > bugfix/scripts'
> > > > > >
> > > > > > *** Running ASCII2NC to reformat ASCII point observations
> > > > > > into NetCDF
> > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > -v 2
> > > > > > DEBUG 1: Config File:
> > > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > > Ascii2NcConfig_default
> > > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > > > make: *** [test] Error 2
> > > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > > >
> > > > > > I am guessing this is a common error.
> > > > > >
> > > > > > Greg Ramos
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 14:57:15 2018
Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
[ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
total 576K
drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
-rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3
-rwxr-xr-x 1 ramos ramos 260K Mar 6 19:57 ncgen
-rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 ncdump
-rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
-rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config
drwxr-xr-x 2 ramos ramos 4.0K Mar 6 20:15 .
-rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
[ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin/ncdump
-rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.2018/bin/ncdump
[ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
linux-vdso.so.1 => (0x00007ffd951ea000)
libnetcdf.so.13 => /software/depot/netcdf-
3.6.2018/lib/libnetcdf.so.13 (0x00002b3eb98fc000)
libmfhdf.so.0 => /software/depot/hdf-4.2.10-rhel6-
aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
libdf.so.0 => /software/depot/hdf-4.2.10-rhel6-
aerosol/lib/libdf.so.0 (0x00002b3eb9e1a000)
libjpeg.so.62 => /usr/lib64/libjpeg.so.62 (0x00000032a6200000)
libhdf5_hl.so.100 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5_hl.so.100 (0x00002b3eba0e3000)
libhdf5.so.101 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5.so.101 (0x00002b3eba306000)
librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x0000003299600000)
/lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
[ramos at maury2 ~]$
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 1:45 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Ah, OK, I see that you're using NetCDF version 3.6. That's likely the
problem. As of MET version 6.0 we switched from NetCDF version 3 to
NetCDF version 4:
https://dtcenter.org/met/users/support/release_notes/METv6.0_release_notes.php
Please try compiling against NetCDF4 instead of 3:
https://dtcenter.org/met/users/support/online_tutorial/METv6.1/tutorial.php?name=compilation&category=req_libs
I'm surprised MET compiled without error and that the problem showed
up at runtime.
John
On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> BTW, here are my build settings:
>
> [ramos at maury2 met-7.0]$ env |grep MET
> MET_GSL=/software/depot/gsl-1.16
> MET_ROOT=/software/depot/met-6.1_bugfix
> MET_NETCDF=/software/depot/netcdf-3.6.2018
> MET_NETCDFINC=
> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> [ramos at maury2 met-7.0]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 1:18 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Another thing to test would be the runtime linker. You set the
> MET_NETCDF environment variable to tell MET where to find the header
> and library files for compilation. By default, MET is compiled
using
> dynamic linking. I suppose its possible that at runtime the MET
> executables are linking to some other version of NetCDF. You could
> set the LD_LIBRARY_PATH environment variable to define the search
path
> for the linker. Let's try
> this:
>
> # Using c-shell syntax here. And MET_NETCDF is the top-level
directory
> for your NetCDF compilations
>
> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-7.0/data/
> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>
> Any difference in results?
>
> Thanks,
> John
>
> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
error:
> >
> >
> >
> > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >
> > *** Running ASCII2NC to reformat ASCII point observations into
> > NetCDF
> > *** ../src/tools/other/ascii2nc/ascii2nc \
> > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > ../out/ascii2nc/sample_ascii.nc \
> > -v 2
> > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0 observations
> > off the masking grid.
> > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > DEBUG 2: Processing observations for 5 headers.
> > DEBUG 1: Creating NetCDF Observation file:
> > ../out/ascii2nc/sample_ascii.nc ERROR :
> > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > make[1]: *** [ascii2nc] Error 1
> > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > make: *** [test] Error 2
> > [ramos at maury2 met-7.0]$
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 10:14 AM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > To clarify, those backslashes '\' are line continuation marks...
> > telling the shell that the command is continued on the next line.
> > Including them on one line like that causes the odd behavior of
> > creating an output file named "2". Removing them and running:
> > /software/depot/met-7.0/bin/ascii2nc
> > /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt
> > ~/sample_ascii.nc -v 2 Will still result in an error... but the
> > requested output file will be what
> you expect.
> >
> > So I'd try recompiling NetCDF using the same family/version number
> > of the compilers you're using to compile MET. Sorry I don't have
an
> > easier solution for you.
> >
> > Thanks,
> > John
> >
> >
> >
> > On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
> > Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > I had to modify your example slightly, but it didn't work:
> > >
> > >
> > > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > > /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
> > > ERROR
> > > :
> > > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > > ~/ sample_ascii.nc" for reading ERROR :
> > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file
> "2"
> > > for writing
> > > ERROR :
> > > [ramos at maury2 met-7.0]$
> > >
> > > Greg
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 9:58 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Greg,
> > >
> > > The number of executables built depends on the way the build was
> > > configured. Options which require "extra" dependencies are
> > > disabled by default. Here's an excerpt from the top-level
README file:
> > >
> > > To configure MET to install all of the available tools in the
"bin"
> > > subdirectory of your current directory, you would use the
> > > following
> > > command:
> > >
> > > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > > --enable-mode_graphics --enable-lidar2nc
> > >
> > > So it's fine that you don't have all possible executables built.
> > > That's unrelated to the test script issue.
> > >
> > > My next guess would be an issue with the NetCDF library to which
> > > MET is linking. Let's replicate the error by trying to write to
> > > some other output directory, like your home directory:
> > >
> > > /users/ramos/met/met-7.0/bin/ascii2nc \
> > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > t
> > > \ ~/ sample_ascii.nc \ -v 2
> > >
> > > I expect that this will fail with the same error message. If it
> > > does, you could try recompiling NetCDF... and then recompiling
MET.
> > > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> > >
> > > Here's some info about that:
> > > https://dtcenter.org/met/users/support/online_tutorial/
> > > METv6.1/tutorial.php?name=compilation&category=req_libs
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> > > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > These are the binaries that were built. It did not build all
of
> > > > the binaries listed in the README:
> > > >
> > > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> > > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> > > > ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
> > > > 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
> > > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> > > > point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > > > stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > > > wavelet_stat -rwxr-xr-x
> > > > 1 ramos ramos 9.8M Mar 6
> > > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
> > > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > > > ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> > > > -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
> > > > ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
> > > > ramos 8.9M Mar 6
> > > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> > > > -rwxr-xr-x
> > > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > > > ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos
> > > > 7.6M Mar
> > > > 6
> > > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> > > > 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > > > 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > > > tc_pairs drwxrwxr-x 2 ramos
> > ramos 4.0K Mar 6 16:44 .
> > > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > -----Original Message-----
> > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Hi Greg,
> > > >
> > > > This is John, I work with Howard. Well I'm glad to hear that
> > > > the behavior is consistent across versions. If not, I would
> > > > have been
> > > confused.
> > > >
> > > > So the very first command that's being run by the test script
is
> > > > failing because it isn't able to create an output file.
> > > >
> > > > As Howard mentioned, the most likely cause for this type of
> > > > error is a permissions issue. So let's rule that out first.
> > > > Please try running the following commands:
> > > > cd /users/ramos/met/met-7.0/scripts
> > > > touch ../out/ascii2nc/test_output_file
> > > >
> > > > The touch command just creates an empty file. If you get an
> > > > error message about permissions, then we've found the problem.
> > > > I tried running the exact same ascii2nc command but writing my
> > > > output to "/usr/bin"... where of course I don't have
permission to write.
> > > > Here's
> > > the error message I got:
> > > >
> > > > cd met-7.0/scripts
> > > > ../src/tools/other/ascii2nc/ascii2nc
> > > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > > /usr/bin/sample_ascii_obs.txt -v 2
> > > >
> > > > ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > > >
> > > > Do a long listing of that directory to see the permissions:
> > > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > > >
> > > > What do you see?
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > > This morning I downloaded the latest version of MET. The
tests
> > > > > still don't
> > > > > run:
> > > > >
> > > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > > make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > > > >
> > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > NetCDF
> > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > -v 2
> > > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > > observations off the masking grid.
> > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
ERROR :
> > > > > make[1]: *** [ascii2nc] Error 1
> > > > > make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > > > > make: *** [test] Error 2
> > > > > [ramos at maury2 met-7.0]$
> > > > >
> > > > > -----Original Message-----
> > > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > >
> > > > > Would you check if ../out/ascii2nc exists?
> > > > >
> > > > > The common cause are 1) when the output directory does not
> > > > > exist or
> > > > > 2) the user does not have the proper privileges (write
privilege).
> > > > >
> > > > > Would you check these?
> > > > >
> > > > > ls ../out/ascii2nc/
> > > > > touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
> > > > >
> > > > > Cheers,
> > > > > Howard
> > > > >
> > > > >
> > > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > > > > Contractor, Code
> > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Queue: met_help
> > > > > > Subject: MET 6.1 bugfix
> > > > > > Owner: Nobody
> > > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > Status: new
> > > > > > Ticket <URL:
> > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > >
> > > > > >
> > > > > >
> > > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > > getting the same error in trying to open the file for
> > > > > > writing using the
> > > tutorial:
> > > > > >
> > > > > >
> > > > > > cd scripts; make
> > > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > > bugfix/scripts'
> > > > > >
> > > > > > *** Running ASCII2NC to reformat ASCII point observations
> > > > > > into NetCDF
> > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > -v 2
> > > > > > DEBUG 1: Config File:
> > > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > > Ascii2NcConfig_default
> > > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > bugfix/scripts'
> > > > > > make: *** [test] Error 2
> > > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > > >
> > > > > > I am guessing this is a common error.
> > > > > >
> > > > > > Greg Ramos
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Tue Mar 06 15:02:46 2018
Greg,
No, the MET tools do not use ncdump. They use the C++ NetCDF
interface for
reading/writing NetCDF files.
John
On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
>
> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
> total 576K
> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3
> -rwxr-xr-x 1 ramos ramos 260K Mar 6 19:57 ncgen
> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 ncdump
> -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config
> drwxr-xr-x 2 ramos ramos 4.0K Mar 6 20:15 .
> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin/ncdump
> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.
> 2018/bin/ncdump
> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
> linux-vdso.so.1 => (0x00007ffd951ea000)
> libnetcdf.so.13 => /software/depot/netcdf-
3.6.2018/lib/libnetcdf.so.13
> (0x00002b3eb98fc000)
> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> libdf.so.0 => /software/depot/hdf-4.2.10-rhel6-
aerosol/lib/libdf.so.0
> (0x00002b3eb9e1a000)
> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
> libhdf5_hl.so.100 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5_hl.so.100
> (0x00002b3eba0e3000)
> libhdf5.so.101 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5.so.101
> (0x00002b3eba306000)
> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> [ramos at maury2 ~]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 1:45 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Ah, OK, I see that you're using NetCDF version 3.6. That's likely
the
> problem. As of MET version 6.0 we switched from NetCDF version 3 to
NetCDF
> version 4:
>
> https://dtcenter.org/met/users/support/release_notes/
> METv6.0_release_notes.php
>
> Please try compiling against NetCDF4 instead of 3:
>
> https://dtcenter.org/met/users/support/online_tutorial/
> METv6.1/tutorial.php?name=compilation&category=req_libs
>
> I'm surprised MET compiled without error and that the problem showed
up at
> runtime.
>
> John
>
> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > BTW, here are my build settings:
> >
> > [ramos at maury2 met-7.0]$ env |grep MET
> > MET_GSL=/software/depot/gsl-1.16
> > MET_ROOT=/software/depot/met-6.1_bugfix
> > MET_NETCDF=/software/depot/netcdf-3.6.2018
> > MET_NETCDFINC=
> > MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> > MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> > MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> > [ramos at maury2 met-7.0]$
> >
> > Greg
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 1:18 PM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > Another thing to test would be the runtime linker. You set the
> > MET_NETCDF environment variable to tell MET where to find the
header
> > and library files for compilation. By default, MET is compiled
using
> > dynamic linking. I suppose its possible that at runtime the MET
> > executables are linking to some other version of NetCDF. You
could
> > set the LD_LIBRARY_PATH environment variable to define the search
path
> > for the linker. Let's try
> > this:
> >
> > # Using c-shell syntax here. And MET_NETCDF is the top-level
directory
> > for your NetCDF compilations
> >
> > setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> > /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> > sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> >
> > Any difference in results?
> >
> > Thanks,
> > John
> >
> > On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > I rebuilt NetCDF C and NetCDF C++ and I am still getting the
same
> error:
> > >
> > >
> > >
> > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
> > > "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 10:14 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Greg,
> > >
> > > To clarify, those backslashes '\' are line continuation marks...
> > > telling the shell that the command is continued on the next
line.
> > > Including them on one line like that causes the odd behavior of
> > > creating an output file named "2". Removing them and running:
> > > /software/depot/met-7.0/bin/ascii2nc
> > > /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt
> > > ~/sample_ascii.nc -v 2 Will still result in an error... but the
> > > requested output file will be what
> > you expect.
> > >
> > > So I'd try recompiling NetCDF using the same family/version
number
> > > of the compilers you're using to compile MET. Sorry I don't
have an
> > > easier solution for you.
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > > On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > I had to modify your example slightly, but it didn't work:
> > > >
> > > >
> > > > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > > t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > > > /software/depot/met-7.0/share/
met/config/Ascii2NcConfig_default
> > > > ERROR
> > > > :
> > > > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > > > ~/ sample_ascii.nc" for reading ERROR :
> > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file
> > "2"
> > > > for writing
> > > > ERROR :
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > Greg
> > > >
> > > > -----Original Message-----
> > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 9:58 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Greg,
> > > >
> > > > The number of executables built depends on the way the build
was
> > > > configured. Options which require "extra" dependencies are
> > > > disabled by default. Here's an excerpt from the top-level
README
> file:
> > > >
> > > > To configure MET to install all of the available tools in
the "bin"
> > > > subdirectory of your current directory, you would use the
> > > > following
> > > > command:
> > > >
> > > > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > > > --enable-mode_graphics --enable-lidar2nc
> > > >
> > > > So it's fine that you don't have all possible executables
built.
> > > > That's unrelated to the test script issue.
> > > >
> > > > My next guess would be an issue with the NetCDF library to
which
> > > > MET is linking. Let's replicate the error by trying to write
to
> > > > some other output directory, like your home directory:
> > > >
> > > > /users/ramos/met/met-7.0/bin/ascii2nc \
> > > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> > > > t
> > > > \ ~/ sample_ascii.nc \ -v 2
> > > >
> > > > I expect that this will fail with the same error message. If
it
> > > > does, you could try recompiling NetCDF... and then recompiling
MET.
> > > > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> > > >
> > > > Here's some info about that:
> > > > https://dtcenter.org/met/users/support/online_tutorial/
> > > > METv6.1/tutorial.php?name=compilation&category=req_libs
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > > > >
> > > > > These are the binaries that were built. It did not build all
of
> > > > > the binaries listed in the README:
> > > > >
> > > > > [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> > > > > total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > > > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > > > -rwxr-xr-x
> > > > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> > > > > ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M
Mar 6
> > > > > 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
16:44
> > > > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> > > > > point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > > > > stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > > > > wavelet_stat -rwxr-xr-x
> > > > > 1 ramos ramos 9.8M Mar 6
> > > > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
> > > > > gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > > > > ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> > > > > -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x
1
> > > > > ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
ramos
> > > > > ramos 8.9M Mar 6
> > > > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> > > > > -rwxr-xr-x
> > > > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > > > > ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
ramos
> > > > > 7.6M Mar
> > > > > 6
> > > > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> > > > > 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > > > > 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > > > > tc_pairs drwxrwxr-x 2 ramos
> > > ramos 4.0K Mar 6 16:44 .
> > > > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > > > [ramos at maury2 met-7.0]$
> > > > >
> > > > > -----Original Message-----
> > > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > >
> > > > > Hi Greg,
> > > > >
> > > > > This is John, I work with Howard. Well I'm glad to hear
that
> > > > > the behavior is consistent across versions. If not, I would
> > > > > have been
> > > > confused.
> > > > >
> > > > > So the very first command that's being run by the test
script is
> > > > > failing because it isn't able to create an output file.
> > > > >
> > > > > As Howard mentioned, the most likely cause for this type of
> > > > > error is a permissions issue. So let's rule that out first.
> > > > > Please try running the following commands:
> > > > > cd /users/ramos/met/met-7.0/scripts
> > > > > touch ../out/ascii2nc/test_output_file
> > > > >
> > > > > The touch command just creates an empty file. If you get an
> > > > > error message about permissions, then we've found the
problem.
> > > > > I tried running the exact same ascii2nc command but writing
my
> > > > > output to "/usr/bin"... where of course I don't have
permission to
> write.
> > > > > Here's
> > > > the error message I got:
> > > > >
> > > > > cd met-7.0/scripts
> > > > > ../src/tools/other/ascii2nc/ascii2nc
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > > > /usr/bin/sample_ascii_obs.txt -v 2
> > > > >
> > > > > ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > > file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > > > >
> > > > > Do a long listing of that directory to see the permissions:
> > > > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > > > >
> > > > > What do you see?
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > >
> > > > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> > > > > Code
> > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > >
> > > > > >
> > > > > > This morning I downloaded the latest version of MET. The
tests
> > > > > > still don't
> > > > > > run:
> > > > > >
> > > > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > > > make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > > > > >
> > > > > > *** Running ASCII2NC to reformat ASCII point observations
into
> > > > > > NetCDF
> > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > -v 2
> > > > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > > > observations off the masking grid.
> > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > > > > > make: *** [test] Error 2
> > > > > > [ramos at maury2 met-7.0]$
> > > > > >
> > > > > > -----Original Message-----
> > > > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > > >
> > > > > > Would you check if ../out/ascii2nc exists?
> > > > > >
> > > > > > The common cause are 1) when the output directory does not
> > > > > > exist or
> > > > > > 2) the user does not have the proper privileges (write
> privilege).
> > > > > >
> > > > > > Would you check these?
> > > > > >
> > > > > > ls ../out/ascii2nc/
> > > > > > touch ../out/ascii2nc/test.txt rm
../out/ascii2nc/test.txt
> > > > > >
> > > > > > Cheers,
> > > > > > Howard
> > > > > >
> > > > > >
> > > > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > > > > > Contractor, Code
> > > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > > >
> > > > > > >
> > > > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > > Queue: met_help
> > > > > > > Subject: MET 6.1 bugfix
> > > > > > > Owner: Nobody
> > > > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > > Status: new
> > > > > > > Ticket <URL:
> > > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > > > getting the same error in trying to open the file for
> > > > > > > writing using the
> > > > tutorial:
> > > > > > >
> > > > > > >
> > > > > > > cd scripts; make
> > > > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > > > bugfix/scripts'
> > > > > > >
> > > > > > > *** Running ASCII2NC to reformat ASCII point
observations
> > > > > > > into NetCDF
> > > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > > -v 2
> > > > > > > DEBUG 1: Config File:
> > > > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > > > Ascii2NcConfig_default
> > > > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > > DEBUG 2: Rejected 0 observations not matched with
station ID's.
> > > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for
writing
> > ERROR :
> > > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > > bugfix/scripts'
> > > > > > > make: *** [test] Error 2
> > > > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > > > >
> > > > > > > I am guessing this is a common error.
> > > > > > >
> > > > > > > Greg Ramos
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 15:05:27 2018
So I rebuilt NetCDF C and NetCDF C++ and used the same install
directory, was that right?
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 2:03 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
No, the MET tools do not use ncdump. They use the C++ NetCDF
interface for reading/writing NetCDF files.
John
On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
>
> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
> total 576K
> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1 ramos
> ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos 109K Mar 6
> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x 2
> ramos ramos 4.0K Mar 6 20:15 .
> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> [ramos at maury2 ~]$ ls -larth
> /software/depot/netcdf-3.6.2018/bin/ncdump
> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.
> 2018/bin/ncdump
> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
> linux-vdso.so.1 => (0x00007ffd951ea000)
> libnetcdf.so.13 =>
> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> (0x00002b3eb98fc000)
> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> libdf.so.0 =>
> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> (0x00002b3eb9e1a000)
> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
> libhdf5_hl.so.100 =>
> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> (0x00002b3eba0e3000)
> libhdf5.so.101 =>
> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> (0x00002b3eba306000)
> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> [ramos at maury2 ~]$
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 1:45 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Ah, OK, I see that you're using NetCDF version 3.6. That's likely
the
> problem. As of MET version 6.0 we switched from NetCDF version 3 to
> NetCDF version 4:
>
> https://dtcenter.org/met/users/support/release_notes/
> METv6.0_release_notes.php
>
> Please try compiling against NetCDF4 instead of 3:
>
> https://dtcenter.org/met/users/support/online_tutorial/
> METv6.1/tutorial.php?name=compilation&category=req_libs
>
> I'm surprised MET compiled without error and that the problem showed
> up at runtime.
>
> John
>
> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > BTW, here are my build settings:
> >
> > [ramos at maury2 met-7.0]$ env |grep MET
> > MET_GSL=/software/depot/gsl-1.16
> > MET_ROOT=/software/depot/met-6.1_bugfix
> > MET_NETCDF=/software/depot/netcdf-3.6.2018
> > MET_NETCDFINC=
> > MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> > MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> > MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> > [ramos at maury2 met-7.0]$
> >
> > Greg
> >
> > -----Original Message-----
> > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 1:18 PM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > Greg,
> >
> > Another thing to test would be the runtime linker. You set the
> > MET_NETCDF environment variable to tell MET where to find the
header
> > and library files for compilation. By default, MET is compiled
> > using dynamic linking. I suppose its possible that at runtime the
> > MET executables are linking to some other version of NetCDF. You
> > could set the LD_LIBRARY_PATH environment variable to define the
> > search path for the linker. Let's try
> > this:
> >
> > # Using c-shell syntax here. And MET_NETCDF is the top-level
> > directory for your NetCDF compilations
> >
> > setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> > /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> > sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> >
> > Any difference in results?
> >
> > Thanks,
> > John
> >
> > On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > I rebuilt NetCDF C and NetCDF C++ and I am still getting the
same
> error:
> > >
> > >
> > >
> > > make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> > >
> > > *** Running ASCII2NC to reformat ASCII point observations into
> > > NetCDF
> > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > ../out/ascii2nc/sample_ascii.nc \
> > > -v 2
> > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> > > off the masking grid.
> > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > DEBUG 2: Rejected 0 observations not matched with station ID's.
> > > DEBUG 2: Processing observations for 5 headers.
> > > DEBUG 1: Creating NetCDF Observation file:
> > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> > > file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > > make[1]: *** [ascii2nc] Error 1
> > > make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > > make: *** [test] Error 2
> > > [ramos at maury2 met-7.0]$
> > >
> > > -----Original Message-----
> > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > Sent: Tuesday, March 6, 2018 10:14 AM
> > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >
> > > Greg,
> > >
> > > To clarify, those backslashes '\' are line continuation marks...
> > > telling the shell that the command is continued on the next
line.
> > > Including them on one line like that causes the odd behavior of
> > > creating an output file named "2". Removing them and running:
> > > /software/depot/met-7.0/bin/ascii2nc
> > > /users/ramos/met/met-7.0/data/
> > > sample_obs/ascii/sample_ascii_obs.txt
> > > ~/sample_ascii.nc -v 2 Will still result in an error... but the
> > > requested output file will be what
> > you expect.
> > >
> > > So I'd try recompiling NetCDF using the same family/version
number
> > > of the compilers you're using to compile MET. Sorry I don't
have
> > > an easier solution for you.
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > > On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > > >
> > > > I had to modify your example slightly, but it didn't work:
> > > >
> > > >
> > > > [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > > > tx t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > > > /software/depot/met-7.0/share/
met/config/Ascii2NcConfig_default
> > > > ERROR
> > > > :
> > > > ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> > > > ~/ sample_ascii.nc" for reading ERROR :
> > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > DEBUG 2: Rejected 0 observations outside the masking polyline.
> > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > DEBUG 2: Processing observations for 5 headers.
> > > > DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > > > ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > > > file
> > "2"
> > > > for writing
> > > > ERROR :
> > > > [ramos at maury2 met-7.0]$
> > > >
> > > > Greg
> > > >
> > > > -----Original Message-----
> > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > Sent: Tuesday, March 6, 2018 9:58 AM
> > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > >
> > > > Greg,
> > > >
> > > > The number of executables built depends on the way the build
was
> > > > configured. Options which require "extra" dependencies are
> > > > disabled by default. Here's an excerpt from the top-level
> > > > README
> file:
> > > >
> > > > To configure MET to install all of the available tools in
the "bin"
> > > > subdirectory of your current directory, you would use the
> > > > following
> > > > command:
> > > >
> > > > ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> > > > --enable-mode_graphics --enable-lidar2nc
> > > >
> > > > So it's fine that you don't have all possible executables
built.
> > > > That's unrelated to the test script issue.
> > > >
> > > > My next guess would be an issue with the NetCDF library to
which
> > > > MET is linking. Let's replicate the error by trying to write
to
> > > > some other output directory, like your home directory:
> > > >
> > > > /users/ramos/met/met-7.0/bin/ascii2nc \
> > > > /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > > > tx
> > > > t
> > > > \ ~/ sample_ascii.nc \ -v 2
> > > >
> > > > I expect that this will fail with the same error message. If
it
> > > > does, you could try recompiling NetCDF... and then recompiling
MET.
> > > > Perhaps there's some incompatibility in how NetCDF and MET are
built?
> > > >
> > > > Here's some info about that:
> > > > https://dtcenter.org/met/users/support/online_tutorial/
> > > > METv6.1/tutorial.php?name=compilation&category=req_libs
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > >
> > > >
> > > > On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
Contractor,
> > > > Code
> > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > >
> > > > >
> > > > > These are the binaries that were built. It did not build all
> > > > > of the binaries listed in the README:
> > > > >
> > > > > [ramos at maury2 met-7.0]$ ls -larth
> > > > > /software/depot/met-7.0/bin/ total 208M drwxr-xr-x 4 ramos
ramos 4.0K Mar 6 16:44 ..
> > > > > -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > > > > -rwxr-xr-x
> > > > > 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> > > > > ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M
Mar
> > > > > 6
> > > > > 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
16:44
> > > > > pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> > > > > point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > > > > stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > > > > wavelet_stat -rwxr-xr-x
> > > > > 1 ramos ramos 9.8M Mar 6
> > > > > 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
> > > > > 16:44 gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> > > > > plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > > > > plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > > > > ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> > > > > -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x
1
> > > > > ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
ramos
> > > > > ramos 8.9M Mar 6
> > > > > 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
> > > > > mtd -rwxr-xr-x
> > > > > 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > > > > ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
ramos
> > > > > 7.6M Mar
> > > > > 6
> > > > > 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> > > > > 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > > > > 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > > > > tc_pairs drwxrwxr-x 2 ramos
> > > ramos 4.0K Mar 6 16:44 .
> > > > > -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > > > > [ramos at maury2 met-7.0]$
> > > > >
> > > > > -----Original Message-----
> > > > > From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > > > > Sent: Tuesday, March 6, 2018 9:12 AM
> > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > >
> > > > > Hi Greg,
> > > > >
> > > > > This is John, I work with Howard. Well I'm glad to hear
that
> > > > > the behavior is consistent across versions. If not, I would
> > > > > have been
> > > > confused.
> > > > >
> > > > > So the very first command that's being run by the test
script
> > > > > is failing because it isn't able to create an output file.
> > > > >
> > > > > As Howard mentioned, the most likely cause for this type of
> > > > > error is a permissions issue. So let's rule that out first.
> > > > > Please try running the following commands:
> > > > > cd /users/ramos/met/met-7.0/scripts
> > > > > touch ../out/ascii2nc/test_output_file
> > > > >
> > > > > The touch command just creates an empty file. If you get an
> > > > > error message about permissions, then we've found the
problem.
> > > > > I tried running the exact same ascii2nc command but writing
my
> > > > > output to "/usr/bin"... where of course I don't have
> > > > > permission to
> write.
> > > > > Here's
> > > > the error message I got:
> > > > >
> > > > > cd met-7.0/scripts
> > > > > ../src/tools/other/ascii2nc/ascii2nc
> > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt
> > > > > /usr/bin/sample_ascii_obs.txt -v 2
> > > > >
> > > > > ERROR :
> > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > NetCDF file "/usr/bin/sample_ascii_obs.txt" for writing
ERROR :
> > > > >
> > > > > Do a long listing of that directory to see the permissions:
> > > > > ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > > > >
> > > > > What do you see?
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > >
> > > > > On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
> > > > > Contractor, Code
> > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > >
> > > > > >
> > > > > > <URL:
> > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > >
> > > > > >
> > > > > > This morning I downloaded the latest version of MET. The
> > > > > > tests still don't
> > > > > > run:
> > > > > >
> > > > > > [ramos at maury2 met-7.0]$ make test cd scripts; make
> > > > > > make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > > > > >
> > > > > > *** Running ASCII2NC to reformat ASCII point observations
> > > > > > into NetCDF
> > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > -v 2
> > > > > > DEBUG 1: Config File: /software/depot/met-7.0/share/
> > > > > > met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > > > > > observations off the masking grid.
> > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> ERROR :
> > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > > > > > make: *** [test] Error 2
> > > > > > [ramos at maury2 met-7.0]$
> > > > > >
> > > > > > -----Original Message-----
> > > > > > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > > > > > Sent: Tuesday, March 6, 2018 8:23 AM
> > > > > > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > > > > > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > > > > >
> > > > > > Would you check if ../out/ascii2nc exists?
> > > > > >
> > > > > > The common cause are 1) when the output directory does not
> > > > > > exist or
> > > > > > 2) the user does not have the proper privileges (write
> privilege).
> > > > > >
> > > > > > Would you check these?
> > > > > >
> > > > > > ls ../out/ascii2nc/
> > > > > > touch ../out/ascii2nc/test.txt rm
../out/ascii2nc/test.txt
> > > > > >
> > > > > > Cheers,
> > > > > > Howard
> > > > > >
> > > > > >
> > > > > > On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > > > > > Contractor, Code
> > > > > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > > > > >
> > > > > > >
> > > > > > > Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > > > > > > Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > > Queue: met_help
> > > > > > > Subject: MET 6.1 bugfix
> > > > > > > Owner: Nobody
> > > > > > > Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > > > > > > Status: new
> > > > > > > Ticket <URL:
> > > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > I built the MET 6.1, but the tests won't run. My user is
> > > > > > > getting the same error in trying to open the file for
> > > > > > > writing using the
> > > > tutorial:
> > > > > > >
> > > > > > >
> > > > > > > cd scripts; make
> > > > > > > make[1]: Entering directory `/users/ramos/met/met-6.1_
> > > > bugfix/scripts'
> > > > > > >
> > > > > > > *** Running ASCII2NC to reformat ASCII point
observations
> > > > > > > into NetCDF
> > > > > > > *** ../src/tools/other/ascii2nc/ascii2nc \
> > > > > > > ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > > > > > > ../out/ascii2nc/sample_ascii.nc \
> > > > > > > -v 2
> > > > > > > DEBUG 1: Config File:
> > > > > > > /software/depot/met-6.1_bugfix/share/met/config/
> > > > > > > Ascii2NcConfig_default
> > > > > > > DEBUG 2: Rejected 0 observations off the masking grid.
> > > > > > > DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > > > > > > DEBUG 2: Rejected 0 observations not matched with
station ID's.
> > > > > > > DEBUG 2: Processing observations for 5 headers.
> > > > > > > DEBUG 1: Creating NetCDF Observation file:
> > > > > > > ../out/ascii2nc/sample_ascii.nc ERROR :
> > > > > > > ERROR : FileHandler::_openNetcdf() -> can't open output
> > > > > > > NetCDF file "../out/ascii2nc/sample_ascii.nc" for
writing
> > ERROR :
> > > > > > > make[1]: *** [ascii2nc] Error 1
> > > > > > > make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > > bugfix/scripts'
> > > > > > > make: *** [test] Error 2
> > > > > > > [ramos at maury2 met-6.1_bugfix]$
> > > > > > >
> > > > > > > I am guessing this is a common error.
> > > > > > >
> > > > > > > Greg Ramos
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Howard Soh
Time: Tue Mar 06 16:09:25 2018
John, ldd ascii2nc will show the NetCDF library.
NetCDF4 can be built without NetCDF4 features. This can be another
case
that NetCDF file can not be opened (created ) because NetCDF was
created
with NetCDF4 format.
Other case will be different configuration on building MET and
executing
MET.
Building MET with NetCDF3 puzzles me. It should be detected by:
Step 1. compile time with header files
Step 2. link time with the library files.
Cheers,
Howard
On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Greg,
>
> No, the MET tools do not use ncdump. They use the C++ NetCDF
interface for
> reading/writing NetCDF files.
>
> John
>
> On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
> via RT <met_help at ucar.edu> wrote:
>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>
>> Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
>>
>> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
>> total 576K
>> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
>> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3
>> -rwxr-xr-x 1 ramos ramos 260K Mar 6 19:57 ncgen
>> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 ncdump
>> -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
>> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config
>> drwxr-xr-x 2 ramos ramos 4.0K Mar 6 20:15 .
>> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
>> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin/ncdump
>> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.
>> 2018/bin/ncdump
>> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
>> linux-vdso.so.1 => (0x00007ffd951ea000)
>> libnetcdf.so.13 => /software/depot/netcdf-
3.6.2018/lib/libnetcdf.so.13
>> (0x00002b3eb98fc000)
>> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
>> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
>> libdf.so.0 => /software/depot/hdf-4.2.10-rhel6-
aerosol/lib/libdf.so.0
>> (0x00002b3eb9e1a000)
>> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
>> libhdf5_hl.so.100 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5_hl.so.100
>> (0x00002b3eba0e3000)
>> libhdf5.so.101 => /software/depot/hdf5-1.10.1-
gcc/lib/libhdf5.so.101
>> (0x00002b3eba306000)
>> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
>> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
>> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
>> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
>> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
>> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
>> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
>> [ramos at maury2 ~]$
>>
>> Greg
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Tuesday, March 6, 2018 1:45 PM
>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>
>> Greg,
>>
>> Ah, OK, I see that you're using NetCDF version 3.6. That's likely
the
>> problem. As of MET version 6.0 we switched from NetCDF version 3
to NetCDF
>> version 4:
>>
>> https://dtcenter.org/met/users/support/release_notes/
>> METv6.0_release_notes.php
>>
>> Please try compiling against NetCDF4 instead of 3:
>>
>> https://dtcenter.org/met/users/support/online_tutorial/
>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>
>> I'm surprised MET compiled without error and that the problem
showed up at
>> runtime.
>>
>> John
>>
>> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
Code
>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>
>>> BTW, here are my build settings:
>>>
>>> [ramos at maury2 met-7.0]$ env |grep MET
>>> MET_GSL=/software/depot/gsl-1.16
>>> MET_ROOT=/software/depot/met-6.1_bugfix
>>> MET_NETCDF=/software/depot/netcdf-3.6.2018
>>> MET_NETCDFINC=
>>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
>>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
>>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
>>> [ramos at maury2 met-7.0]$
>>>
>>> Greg
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Tuesday, March 6, 2018 1:18 PM
>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>
>>> Greg,
>>>
>>> Another thing to test would be the runtime linker. You set the
>>> MET_NETCDF environment variable to tell MET where to find the
header
>>> and library files for compilation. By default, MET is compiled
using
>>> dynamic linking. I suppose its possible that at runtime the MET
>>> executables are linking to some other version of NetCDF. You
could
>>> set the LD_LIBRARY_PATH environment variable to define the search
path
>>> for the linker. Let's try
>>> this:
>>>
>>> # Using c-shell syntax here. And MET_NETCDF is the top-level
directory
>>> for your NetCDF compilations
>>>
>>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
>>> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
>>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>>>
>>> Any difference in results?
>>>
>>> Thanks,
>>> John
>>>
>>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>
>>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
>> error:
>>>>
>>>>
>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>
>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>> NetCDF
>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>> ../out/ascii2nc/sample_ascii.nc \
>>>> -v 2
>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
>>>> off the masking grid.
>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>> DEBUG 2: Processing observations for 5 headers.
>>>> DEBUG 1: Creating NetCDF Observation file:
>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
file
>>>> "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
>>>> make[1]: *** [ascii2nc] Error 1
>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>> make: *** [test] Error 2
>>>> [ramos at maury2 met-7.0]$
>>>>
>>>> -----Original Message-----
>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>> Sent: Tuesday, March 6, 2018 10:14 AM
>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>
>>>> Greg,
>>>>
>>>> To clarify, those backslashes '\' are line continuation marks...
>>>> telling the shell that the command is continued on the next line.
>>>> Including them on one line like that causes the odd behavior of
>>>> creating an output file named "2". Removing them and running:
>>>> /software/depot/met-7.0/bin/ascii2nc
>>>> /users/ramos/met/met-7.0/data/
sample_obs/ascii/sample_ascii_obs.txt
>>>> ~/sample_ascii.nc -v 2 Will still result in an error... but the
>>>> requested output file will be what
>>> you expect.
>>>> So I'd try recompiling NetCDF using the same family/version
number
>>>> of the compilers you're using to compile MET. Sorry I don't have
an
>>>> easier solution for you.
>>>>
>>>> Thanks,
>>>> John
>>>>
>>>>
>>>>
>>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
>>>> Code
>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>
>>>>> I had to modify your example slightly, but it didn't work:
>>>>>
>>>>>
>>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
>>>>> /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
>>>>> ERROR
>>>>> :
>>>>> ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
>>>>> ~/ sample_ascii.nc" for reading ERROR :
>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>> file
>>> "2"
>>>>> for writing
>>>>> ERROR :
>>>>> [ramos at maury2 met-7.0]$
>>>>>
>>>>> Greg
>>>>>
>>>>> -----Original Message-----
>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>> Sent: Tuesday, March 6, 2018 9:58 AM
>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>
>>>>> Greg,
>>>>>
>>>>> The number of executables built depends on the way the build was
>>>>> configured. Options which require "extra" dependencies are
>>>>> disabled by default. Here's an excerpt from the top-level
README
>> file:
>>>>> To configure MET to install all of the available tools in the
"bin"
>>>>> subdirectory of your current directory, you would use the
>>>>> following
>>>>> command:
>>>>>
>>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
>>>>> --enable-mode_graphics --enable-lidar2nc
>>>>>
>>>>> So it's fine that you don't have all possible executables built.
>>>>> That's unrelated to the test script issue.
>>>>>
>>>>> My next guess would be an issue with the NetCDF library to which
>>>>> MET is linking. Let's replicate the error by trying to write to
>>>>> some other output directory, like your home directory:
>>>>>
>>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t
>>>>> \ ~/ sample_ascii.nc \ -v 2
>>>>>
>>>>> I expect that this will fail with the same error message. If it
>>>>> does, you could try recompiling NetCDF... and then recompiling
MET.
>>>>> Perhaps there's some incompatibility in how NetCDF and MET are
built?
>>>>>
>>>>> Here's some info about that:
>>>>> https://dtcenter.org/met/users/support/online_tutorial/
>>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>>>>
>>>>> Thanks,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
>>>>> Code
>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>
>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>>
>>>>>> These are the binaries that were built. It did not build all of
>>>>>> the binaries listed in the README:
>>>>>>
>>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
>>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
>>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
>>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
>>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
>>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
>>>>>> point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
>>>>>> stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
>>>>>> wavelet_stat -rwxr-xr-x
>>>>>> 1 ramos ramos 9.8M Mar 6
>>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
>>>>>> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
>>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
>>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
>>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
>>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
>>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
>>>>>> ramos 8.9M Mar 6
>>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
>>>>>> ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos
>>>>>> 7.6M Mar
>>>>>> 6
>>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
>>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
>>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
>>>>>> tc_pairs drwxrwxr-x 2 ramos
>>>> ramos 4.0K Mar 6 16:44 .
>>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
>>>>>> [ramos at maury2 met-7.0]$
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>
>>>>>> Hi Greg,
>>>>>>
>>>>>> This is John, I work with Howard. Well I'm glad to hear that
>>>>>> the behavior is consistent across versions. If not, I would
>>>>>> have been
>>>>> confused.
>>>>>> So the very first command that's being run by the test script
is
>>>>>> failing because it isn't able to create an output file.
>>>>>>
>>>>>> As Howard mentioned, the most likely cause for this type of
>>>>>> error is a permissions issue. So let's rule that out first.
>>>>>> Please try running the following commands:
>>>>>> cd /users/ramos/met/met-7.0/scripts
>>>>>> touch ../out/ascii2nc/test_output_file
>>>>>>
>>>>>> The touch command just creates an empty file. If you get an
>>>>>> error message about permissions, then we've found the problem.
>>>>>> I tried running the exact same ascii2nc command but writing my
>>>>>> output to "/usr/bin"... where of course I don't have permission
to
>> write.
>>>>>> Here's
>>>>> the error message I got:
>>>>>> cd met-7.0/scripts
>>>>>> ../src/tools/other/ascii2nc/ascii2nc
>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
>>>>>> /usr/bin/sample_ascii_obs.txt -v 2
>>>>>>
>>>>>> ERROR :
>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>>>>>>
>>>>>> Do a long listing of that directory to see the permissions:
>>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>>>>>>
>>>>>> What do you see?
>>>>>>
>>>>>> Thanks,
>>>>>> John
>>>>>>
>>>>>>
>>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
>>>>>> Code
>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>
>>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>> This morning I downloaded the latest version of MET. The tests
>>>>>>> still don't
>>>>>>> run:
>>>>>>>
>>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
>>>>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>>>>
>>>>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>>>>> NetCDF
>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>> -v 2
>>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
>>>>>>> observations off the masking grid.
>>>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
>> ERROR :
>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>>>>> make: *** [test] Error 2
>>>>>>> [ramos at maury2 met-7.0]$
>>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
>>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
>>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>>
>>>>>>> Would you check if ../out/ascii2nc exists?
>>>>>>>
>>>>>>> The common cause are 1) when the output directory does not
>>>>>>> exist or
>>>>>>> 2) the user does not have the proper privileges (write
>> privilege).
>>>>>>> Would you check these?
>>>>>>>
>>>>>>> ls ../out/ascii2nc/
>>>>>>> touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Howard
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
>>>>>>> Contractor, Code
>>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>>
>>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
>>>>>>>> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Queue: met_help
>>>>>>>> Subject: MET 6.1 bugfix
>>>>>>>> Owner: Nobody
>>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Status: new
>>>>>>>> Ticket <URL:
>>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>>>
>>>>>>>> I built the MET 6.1, but the tests won't run. My user is
>>>>>>>> getting the same error in trying to open the file for
>>>>>>>> writing using the
>>>>> tutorial:
>>>>>>>>
>>>>>>>> cd scripts; make
>>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
>>>>> bugfix/scripts'
>>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
>>>>>>>> into NetCDF
>>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>>> -v 2
>>>>>>>> DEBUG 1: Config File:
>>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
>>>>>>>> Ascii2NcConfig_default
>>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
>>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
>>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
>>> ERROR :
>>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
>>>> bugfix/scripts'
>>>>>>>> make: *** [test] Error 2
>>>>>>>> [ramos at maury2 met-6.1_bugfix]$
>>>>>>>>
>>>>>>>> I am guessing this is a common error.
>>>>>>>>
>>>>>>>> Greg Ramos
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 16:47:39 2018
No, I did not use NetCDF 3. That was the name of today's date, the
install directory.
[ramos at maury2 ~]$ ldd /software/depot/met-7.0/bin/ascii2nc
linux-vdso.so.1 => (0x00007ffc3adce000)
libnetcdf_c++4.so.1 => /software/depot/netcdf-
both/lib/libnetcdf_c++4.so.1 (0x00002ab363f39000)
libnetcdf.so.13 => /software/depot/netcdf-
both/lib/libnetcdf.so.13 (0x00002ab364186000)
libgsl.so.0 => /software/depot/gsl-1.16/lib/libgsl.so.0
(0x00002ab364492000)
libgslcblas.so.0 => /software/depot/gsl-
1.16/lib/libgslcblas.so.0 (0x00002ab3648b8000)
libstdc++.so.6 => /software/depot/gcc-
6.3.0/lib64/libstdc++.so.6 (0x00002ab364af2000)
libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
libgcc_s.so.1 => /software/depot/gcc-6.3.0/lib64/libgcc_s.so.1
(0x00002ab364ea0000)
libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
libnetcdf.so.6 => /usr/lib64/libnetcdf.so.6
(0x0000003dd2400000)
libcurl.so.4 => /usr/lib64/libcurl.so.4 (0x00000036d0c00000)
/lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
libhdf5_hl.so.6 => /usr/lib64/libhdf5_hl.so.6
(0x0000003dd2000000)
libhdf5.so.6 => /usr/lib64/libhdf5.so.6 (0x0000003dd1a00000)
libidn.so.11 => /lib64/libidn.so.11 (0x000000329a600000)
libldap-2.4.so.2 => /lib64/libldap-2.4.so.2
(0x00000036d0800000)
librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
libgssapi_krb5.so.2 => /lib64/libgssapi_krb5.so.2
(0x00000032a8a00000)
libkrb5.so.3 => /lib64/libkrb5.so.3 (0x00000032a7200000)
libk5crypto.so.3 => /lib64/libk5crypto.so.3
(0x00000032a6e00000)
libcom_err.so.2 => /lib64/libcom_err.so.2 (0x00000032a5e00000)
libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
libssl3.so => /usr/lib64/libssl3.so (0x00000036d0000000)
libsmime3.so => /usr/lib64/libsmime3.so (0x00000036cfc00000)
libnss3.so => /usr/lib64/libnss3.so (0x00000036cf800000)
libnssutil3.so => /usr/lib64/libnssutil3.so
(0x00000032a7a00000)
libplds4.so => /lib64/libplds4.so (0x00000032a8200000)
libplc4.so => /lib64/libplc4.so (0x00000032a9200000)
libnspr4.so => /lib64/libnspr4.so (0x00000032a8600000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x0000003299600000)
libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
libssh2.so.1 => /usr/lib64/libssh2.so.1 (0x000000329d600000)
liblber-2.4.so.2 => /lib64/liblber-2.4.so.2
(0x00000036d0400000)
libresolv.so.2 => /lib64/libresolv.so.2 (0x000000329b600000)
libsasl2.so.2 => /usr/lib64/libsasl2.so.2 (0x00000032a6200000)
libkrb5support.so.0 => /lib64/libkrb5support.so.0
(0x00000032a7600000)
libkeyutils.so.1 => /lib64/libkeyutils.so.1
(0x00000032a7e00000)
libssl.so.10 => /usr/lib64/libssl.so.10 (0x00000032a9e00000)
libcrypto.so.10 => /usr/lib64/libcrypto.so.10
(0x00000032a6600000)
libcrypt.so.1 => /lib64/libcrypt.so.1 (0x00000032a5200000)
libselinux.so.1 => /lib64/libselinux.so.1 (0x000000329aa00000)
libfreebl3.so => /lib64/libfreebl3.so (0x00000032a5600000)
[ramos at maury2 ~]$
Greg
-----Original Message-----
From: Howard Soh via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 3:09 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
John, ldd ascii2nc will show the NetCDF library.
NetCDF4 can be built without NetCDF4 features. This can be another
case that NetCDF file can not be opened (created ) because NetCDF was
created with NetCDF4 format.
Other case will be different configuration on building MET and
executing MET.
Building MET with NetCDF3 puzzles me. It should be detected by:
Step 1. compile time with header files
Step 2. link time with the library files.
Cheers,
Howard
On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Greg,
>
> No, the MET tools do not use ncdump. They use the C++ NetCDF
> interface for reading/writing NetCDF files.
>
> John
>
> On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>
>> Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
>>
>> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
>> total 576K
>> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
>> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
ramos
>> ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos 109K Mar 6
>> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
>> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x 2
>> ramos ramos 4.0K Mar 6 20:15 .
>> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
>> [ramos at maury2 ~]$ ls -larth
>> /software/depot/netcdf-3.6.2018/bin/ncdump
>> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.
>> 2018/bin/ncdump
>> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
>> linux-vdso.so.1 => (0x00007ffd951ea000)
>> libnetcdf.so.13 =>
>> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
>> (0x00002b3eb98fc000)
>> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
>> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
>> libdf.so.0 =>
>> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
>> (0x00002b3eb9e1a000)
>> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
>> libhdf5_hl.so.100 =>
>> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
>> (0x00002b3eba0e3000)
>> libhdf5.so.101 =>
>> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
>> (0x00002b3eba306000)
>> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
>> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
>> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
>> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
>> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
>> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
>> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
>> [ramos at maury2 ~]$
>>
>> Greg
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Tuesday, March 6, 2018 1:45 PM
>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>
>> Greg,
>>
>> Ah, OK, I see that you're using NetCDF version 3.6. That's likely
>> the problem. As of MET version 6.0 we switched from NetCDF version
3
>> to NetCDF version 4:
>>
>> https://dtcenter.org/met/users/support/release_notes/
>> METv6.0_release_notes.php
>>
>> Please try compiling against NetCDF4 instead of 3:
>>
>> https://dtcenter.org/met/users/support/online_tutorial/
>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>
>> I'm surprised MET compiled without error and that the problem
showed
>> up at runtime.
>>
>> John
>>
>> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
Code
>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>
>>> BTW, here are my build settings:
>>>
>>> [ramos at maury2 met-7.0]$ env |grep MET
>>> MET_GSL=/software/depot/gsl-1.16
>>> MET_ROOT=/software/depot/met-6.1_bugfix
>>> MET_NETCDF=/software/depot/netcdf-3.6.2018
>>> MET_NETCDFINC=
>>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
>>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
>>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
>>> [ramos at maury2 met-7.0]$
>>>
>>> Greg
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Tuesday, March 6, 2018 1:18 PM
>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>
>>> Greg,
>>>
>>> Another thing to test would be the runtime linker. You set the
>>> MET_NETCDF environment variable to tell MET where to find the
header
>>> and library files for compilation. By default, MET is compiled
>>> using dynamic linking. I suppose its possible that at runtime the
>>> MET executables are linking to some other version of NetCDF. You
>>> could set the LD_LIBRARY_PATH environment variable to define the
>>> search path for the linker. Let's try
>>> this:
>>>
>>> # Using c-shell syntax here. And MET_NETCDF is the top-level
>>> directory for your NetCDF compilations
>>>
>>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
>>> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
>>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>>>
>>> Any difference in results?
>>>
>>> Thanks,
>>> John
>>>
>>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>
>>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
>> error:
>>>>
>>>>
>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>
>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>> NetCDF
>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>> ../out/ascii2nc/sample_ascii.nc \
>>>> -v 2
>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
>>>> off the masking grid.
>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>> DEBUG 2: Processing observations for 5 headers.
>>>> DEBUG 1: Creating NetCDF Observation file:
>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
>>>> make[1]: *** [ascii2nc] Error 1
>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>> make: *** [test] Error 2
>>>> [ramos at maury2 met-7.0]$
>>>>
>>>> -----Original Message-----
>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>> Sent: Tuesday, March 6, 2018 10:14 AM
>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>
>>>> Greg,
>>>>
>>>> To clarify, those backslashes '\' are line continuation marks...
>>>> telling the shell that the command is continued on the next line.
>>>> Including them on one line like that causes the odd behavior of
>>>> creating an output file named "2". Removing them and running:
>>>> /software/depot/met-7.0/bin/ascii2nc
>>>> /users/ramos/met/met-7.0/data/
>>>> sample_obs/ascii/sample_ascii_obs.txt
>>>> ~/sample_ascii.nc -v 2 Will still result in an error... but the
>>>> requested output file will be what
>>> you expect.
>>>> So I'd try recompiling NetCDF using the same family/version
number
>>>> of the compilers you're using to compile MET. Sorry I don't have
>>>> an easier solution for you.
>>>>
>>>> Thanks,
>>>> John
>>>>
>>>>
>>>>
>>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
>>>> Code
>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>
>>>>> I had to modify your example slightly, but it didn't work:
>>>>>
>>>>>
>>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
>>>>> /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
>>>>> ERROR
>>>>> :
>>>>> ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
>>>>> ~/ sample_ascii.nc" for reading ERROR :
>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>> file
>>> "2"
>>>>> for writing
>>>>> ERROR :
>>>>> [ramos at maury2 met-7.0]$
>>>>>
>>>>> Greg
>>>>>
>>>>> -----Original Message-----
>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>> Sent: Tuesday, March 6, 2018 9:58 AM
>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>
>>>>> Greg,
>>>>>
>>>>> The number of executables built depends on the way the build was
>>>>> configured. Options which require "extra" dependencies are
>>>>> disabled by default. Here's an excerpt from the top-level
README
>> file:
>>>>> To configure MET to install all of the available tools in the
"bin"
>>>>> subdirectory of your current directory, you would use the
>>>>> following
>>>>> command:
>>>>>
>>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
>>>>> --enable-mode_graphics --enable-lidar2nc
>>>>>
>>>>> So it's fine that you don't have all possible executables built.
>>>>> That's unrelated to the test script issue.
>>>>>
>>>>> My next guess would be an issue with the NetCDF library to which
>>>>> MET is linking. Let's replicate the error by trying to write to
>>>>> some other output directory, like your home directory:
>>>>>
>>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t
>>>>> \ ~/ sample_ascii.nc \ -v 2
>>>>>
>>>>> I expect that this will fail with the same error message. If it
>>>>> does, you could try recompiling NetCDF... and then recompiling
MET.
>>>>> Perhaps there's some incompatibility in how NetCDF and MET are
built?
>>>>>
>>>>> Here's some info about that:
>>>>> https://dtcenter.org/met/users/support/online_tutorial/
>>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>>>>
>>>>> Thanks,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
>>>>> Code
>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>
>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>>
>>>>>> These are the binaries that were built. It did not build all of
>>>>>> the binaries listed in the README:
>>>>>>
>>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
>>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
>>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
>>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
>>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
>>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
>>>>>> -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
>>>>>> -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-
xr-x
>>>>>> 1 ramos ramos 9.8M Mar 6
>>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
>>>>>> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
>>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
>>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
>>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
>>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
>>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
>>>>>> ramos 8.9M Mar 6
>>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
>>>>>> 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
>>>>>> 6
>>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
>>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
>>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
>>>>>> tc_pairs drwxrwxr-x 2 ramos
>>>> ramos 4.0K Mar 6 16:44 .
>>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
>>>>>> [ramos at maury2 met-7.0]$
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>
>>>>>> Hi Greg,
>>>>>>
>>>>>> This is John, I work with Howard. Well I'm glad to hear that
the
>>>>>> behavior is consistent across versions. If not, I would have
>>>>>> been
>>>>> confused.
>>>>>> So the very first command that's being run by the test script
is
>>>>>> failing because it isn't able to create an output file.
>>>>>>
>>>>>> As Howard mentioned, the most likely cause for this type of
error
>>>>>> is a permissions issue. So let's rule that out first.
>>>>>> Please try running the following commands:
>>>>>> cd /users/ramos/met/met-7.0/scripts
>>>>>> touch ../out/ascii2nc/test_output_file
>>>>>>
>>>>>> The touch command just creates an empty file. If you get an
>>>>>> error message about permissions, then we've found the problem.
>>>>>> I tried running the exact same ascii2nc command but writing my
>>>>>> output to "/usr/bin"... where of course I don't have permission
>>>>>> to
>> write.
>>>>>> Here's
>>>>> the error message I got:
>>>>>> cd met-7.0/scripts
>>>>>> ../src/tools/other/ascii2nc/ascii2nc
>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
>>>>>> /usr/bin/sample_ascii_obs.txt -v 2
>>>>>>
>>>>>> ERROR :
>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>>>>>>
>>>>>> Do a long listing of that directory to see the permissions:
>>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>>>>>>
>>>>>> What do you see?
>>>>>>
>>>>>> Thanks,
>>>>>> John
>>>>>>
>>>>>>
>>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
>>>>>> Code
>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>
>>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>> This morning I downloaded the latest version of MET. The tests
>>>>>>> still don't
>>>>>>> run:
>>>>>>>
>>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
>>>>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>>>>
>>>>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>>>>> NetCDF
>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>> -v 2
>>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
>>>>>>> observations off the masking grid.
>>>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
>>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
>> ERROR :
>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>>>>> make: *** [test] Error 2
>>>>>>> [ramos at maury2 met-7.0]$
>>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
>>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
>>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>>
>>>>>>> Would you check if ../out/ascii2nc exists?
>>>>>>>
>>>>>>> The common cause are 1) when the output directory does not
exist
>>>>>>> or
>>>>>>> 2) the user does not have the proper privileges (write
>> privilege).
>>>>>>> Would you check these?
>>>>>>>
>>>>>>> ls ../out/ascii2nc/
>>>>>>> touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Howard
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
>>>>>>> Code
>>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>>
>>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
>>>>>>>> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Queue: met_help
>>>>>>>> Subject: MET 6.1 bugfix
>>>>>>>> Owner: Nobody
>>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Status: new
>>>>>>>> Ticket <URL:
>>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>>>
>>>>>>>> I built the MET 6.1, but the tests won't run. My user is
>>>>>>>> getting the same error in trying to open the file for writing
>>>>>>>> using the
>>>>> tutorial:
>>>>>>>>
>>>>>>>> cd scripts; make
>>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
>>>>> bugfix/scripts'
>>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
>>>>>>>> NetCDF
>>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>>> -v 2
>>>>>>>> DEBUG 1: Config File:
>>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
>>>>>>>> Ascii2NcConfig_default
>>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
>>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
>>>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
>>> ERROR :
>>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
>>>> bugfix/scripts'
>>>>>>>> make: *** [test] Error 2
>>>>>>>> [ramos at maury2 met-6.1_bugfix]$
>>>>>>>>
>>>>>>>> I am guessing this is a common error.
>>>>>>>>
>>>>>>>> Greg Ramos
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Tue Mar 06 21:17:18 2018
This is the output of nc-config --all
[ramos at maury2 ncview-2.1.7]$ /software/depot/netcdf-3.6.2018/bin/nc-
config --all
This netCDF 4.6.0 has been built with the following features:
--cc -> gcc
--cflags -> -I/software/depot/netcdf-3.6.2018/include
-I/software/depot/hdf5-1.10.1-gcc/include -I/software/depot/hdf-
4.2.10-rhel6-aerosol/include
--libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
--has-c++ -> no
--cxx ->
--has-c++4 -> yes
--cxx4 -> g++
--cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
--cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib -lnetcdf_c++4
-lnetcdf
--has-fortran-> no
--has-dap -> no
--has-dap4 -> no
--has-nc2 -> yes
--has-nc4 -> yes
--has-hdf5 -> yes
--has-hdf4 -> yes
--has-logging-> no
--has-pnetcdf-> no
--has-szlib -> no
--has-parallel -> no
--has-cdf5 -> no
--prefix -> /software/depot/netcdf-3.6.2018
--includedir-> /software/depot/netcdf-3.6.2018/include
--libdir -> /software/depot/netcdf-3.6.2018/lib
--version -> netCDF 4.6.0
[ramos at maury2 ncview-2.1.7]$
-----Original Message-----
From: Howard Soh via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, March 6, 2018 3:09 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
John, ldd ascii2nc will show the NetCDF library.
NetCDF4 can be built without NetCDF4 features. This can be another
case that NetCDF file can not be opened (created ) because NetCDF was
created with NetCDF4 format.
Other case will be different configuration on building MET and
executing MET.
Building MET with NetCDF3 puzzles me. It should be detected by:
Step 1. compile time with header files
Step 2. link time with the library files.
Cheers,
Howard
On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Greg,
>
> No, the MET tools do not use ncdump. They use the C++ NetCDF
> interface for reading/writing NetCDF files.
>
> John
>
> On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>
>> Does ascii2nc use ncdump? Here is my NetCDF buld binary directory:
>>
>> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
>> total 576K
>> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
>> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
ramos
>> ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos 109K Mar 6
>> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
>> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x 2
>> ramos ramos 4.0K Mar 6 20:15 .
>> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
>> [ramos at maury2 ~]$ ls -larth
>> /software/depot/netcdf-3.6.2018/bin/ncdump
>> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58 /software/depot/netcdf-
3.6.
>> 2018/bin/ncdump
>> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
>> linux-vdso.so.1 => (0x00007ffd951ea000)
>> libnetcdf.so.13 =>
>> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
>> (0x00002b3eb98fc000)
>> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
>> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
>> libdf.so.0 =>
>> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
>> (0x00002b3eb9e1a000)
>> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
>> libhdf5_hl.so.100 =>
>> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
>> (0x00002b3eba0e3000)
>> libhdf5.so.101 =>
>> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
>> (0x00002b3eba306000)
>> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
>> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
>> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
>> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
>> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
>> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
>> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
>> [ramos at maury2 ~]$
>>
>> Greg
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Tuesday, March 6, 2018 1:45 PM
>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>
>> Greg,
>>
>> Ah, OK, I see that you're using NetCDF version 3.6. That's likely
>> the problem. As of MET version 6.0 we switched from NetCDF version
3
>> to NetCDF version 4:
>>
>> https://dtcenter.org/met/users/support/release_notes/
>> METv6.0_release_notes.php
>>
>> Please try compiling against NetCDF4 instead of 3:
>>
>> https://dtcenter.org/met/users/support/online_tutorial/
>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>
>> I'm surprised MET compiled without error and that the problem
showed
>> up at runtime.
>>
>> John
>>
>> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
Code
>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>
>>> BTW, here are my build settings:
>>>
>>> [ramos at maury2 met-7.0]$ env |grep MET
>>> MET_GSL=/software/depot/gsl-1.16
>>> MET_ROOT=/software/depot/met-6.1_bugfix
>>> MET_NETCDF=/software/depot/netcdf-3.6.2018
>>> MET_NETCDFINC=
>>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
>>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
>>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
>>> [ramos at maury2 met-7.0]$
>>>
>>> Greg
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Tuesday, March 6, 2018 1:18 PM
>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>
>>> Greg,
>>>
>>> Another thing to test would be the runtime linker. You set the
>>> MET_NETCDF environment variable to tell MET where to find the
header
>>> and library files for compilation. By default, MET is compiled
>>> using dynamic linking. I suppose its possible that at runtime the
>>> MET executables are linking to some other version of NetCDF. You
>>> could set the LD_LIBRARY_PATH environment variable to define the
>>> search path for the linker. Let's try
>>> this:
>>>
>>> # Using c-shell syntax here. And MET_NETCDF is the top-level
>>> directory for your NetCDF compilations
>>>
>>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
>>> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
>>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
>>>
>>> Any difference in results?
>>>
>>> Thanks,
>>> John
>>>
>>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>
>>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the same
>> error:
>>>>
>>>>
>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>
>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>> NetCDF
>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>> ../out/ascii2nc/sample_ascii.nc \
>>>> -v 2
>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
>>>> off the masking grid.
>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>> DEBUG 2: Processing observations for 5 headers.
>>>> DEBUG 1: Creating NetCDF Observation file:
>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
>>>> make[1]: *** [ascii2nc] Error 1
>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>> make: *** [test] Error 2
>>>> [ramos at maury2 met-7.0]$
>>>>
>>>> -----Original Message-----
>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>> Sent: Tuesday, March 6, 2018 10:14 AM
>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>
>>>> Greg,
>>>>
>>>> To clarify, those backslashes '\' are line continuation marks...
>>>> telling the shell that the command is continued on the next line.
>>>> Including them on one line like that causes the odd behavior of
>>>> creating an output file named "2". Removing them and running:
>>>> /software/depot/met-7.0/bin/ascii2nc
>>>> /users/ramos/met/met-7.0/data/
>>>> sample_obs/ascii/sample_ascii_obs.txt
>>>> ~/sample_ascii.nc -v 2 Will still result in an error... but the
>>>> requested output file will be what
>>> you expect.
>>>> So I'd try recompiling NetCDF using the same family/version
number
>>>> of the compilers you're using to compile MET. Sorry I don't have
>>>> an easier solution for you.
>>>>
>>>> Thanks,
>>>> John
>>>>
>>>>
>>>>
>>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory, Contractor,
>>>> Code
>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>
>>>>> I had to modify your example slightly, but it didn't work:
>>>>>
>>>>>
>>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
>>>>> /software/depot/met-7.0/share/ met/config/Ascii2NcConfig_default
>>>>> ERROR
>>>>> :
>>>>> ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
>>>>> ~/ sample_ascii.nc" for reading ERROR :
>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>> file
>>> "2"
>>>>> for writing
>>>>> ERROR :
>>>>> [ramos at maury2 met-7.0]$
>>>>>
>>>>> Greg
>>>>>
>>>>> -----Original Message-----
>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>> Sent: Tuesday, March 6, 2018 9:58 AM
>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>
>>>>> Greg,
>>>>>
>>>>> The number of executables built depends on the way the build was
>>>>> configured. Options which require "extra" dependencies are
>>>>> disabled by default. Here's an excerpt from the top-level
README
>> file:
>>>>> To configure MET to install all of the available tools in the
"bin"
>>>>> subdirectory of your current directory, you would use the
>>>>> following
>>>>> command:
>>>>>
>>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
>>>>> --enable-mode_graphics --enable-lidar2nc
>>>>>
>>>>> So it's fine that you don't have all possible executables built.
>>>>> That's unrelated to the test script issue.
>>>>>
>>>>> My next guess would be an issue with the NetCDF library to which
>>>>> MET is linking. Let's replicate the error by trying to write to
>>>>> some other output directory, like your home directory:
>>>>>
>>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
>>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
>>>>> t
>>>>> \ ~/ sample_ascii.nc \ -v 2
>>>>>
>>>>> I expect that this will fail with the same error message. If it
>>>>> does, you could try recompiling NetCDF... and then recompiling
MET.
>>>>> Perhaps there's some incompatibility in how NetCDF and MET are
built?
>>>>>
>>>>> Here's some info about that:
>>>>> https://dtcenter.org/met/users/support/online_tutorial/
>>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
>>>>>
>>>>> Thanks,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory, Contractor,
>>>>> Code
>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>
>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>>>>>>
>>>>>> These are the binaries that were built. It did not build all of
>>>>>> the binaries listed in the README:
>>>>>>
>>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-7.0/bin/
>>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
>>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
>>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
>>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6 16:44
>>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
>>>>>> -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
>>>>>> -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-
xr-x
>>>>>> 1 ramos ramos 9.8M Mar 6
>>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
>>>>>> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
>>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
>>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
>>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
>>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
>>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
>>>>>> ramos 8.9M Mar 6
>>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44 mtd
>>>>>> -rwxr-xr-x
>>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
>>>>>> 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
>>>>>> 6
>>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
>>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
>>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
>>>>>> tc_pairs drwxrwxr-x 2 ramos
>>>> ramos 4.0K Mar 6 16:44 .
>>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
>>>>>> [ramos at maury2 met-7.0]$
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>
>>>>>> Hi Greg,
>>>>>>
>>>>>> This is John, I work with Howard. Well I'm glad to hear that
the
>>>>>> behavior is consistent across versions. If not, I would have
>>>>>> been
>>>>> confused.
>>>>>> So the very first command that's being run by the test script
is
>>>>>> failing because it isn't able to create an output file.
>>>>>>
>>>>>> As Howard mentioned, the most likely cause for this type of
error
>>>>>> is a permissions issue. So let's rule that out first.
>>>>>> Please try running the following commands:
>>>>>> cd /users/ramos/met/met-7.0/scripts
>>>>>> touch ../out/ascii2nc/test_output_file
>>>>>>
>>>>>> The touch command just creates an empty file. If you get an
>>>>>> error message about permissions, then we've found the problem.
>>>>>> I tried running the exact same ascii2nc command but writing my
>>>>>> output to "/usr/bin"... where of course I don't have permission
>>>>>> to
>> write.
>>>>>> Here's
>>>>> the error message I got:
>>>>>> cd met-7.0/scripts
>>>>>> ../src/tools/other/ascii2nc/ascii2nc
>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
>>>>>> /usr/bin/sample_ascii_obs.txt -v 2
>>>>>>
>>>>>> ERROR :
>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
>>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
>>>>>>
>>>>>> Do a long listing of that directory to see the permissions:
>>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
>>>>>>
>>>>>> What do you see?
>>>>>>
>>>>>> Thanks,
>>>>>> John
>>>>>>
>>>>>>
>>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory, Contractor,
>>>>>> Code
>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>
>>>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>> This morning I downloaded the latest version of MET. The tests
>>>>>>> still don't
>>>>>>> run:
>>>>>>>
>>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
>>>>>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
>>>>>>>
>>>>>>> *** Running ASCII2NC to reformat ASCII point observations into
>>>>>>> NetCDF
>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>> -v 2
>>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
>>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
>>>>>>> observations off the masking grid.
>>>>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
>>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
>> ERROR :
>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
>>>>>>> make: *** [test] Error 2
>>>>>>> [ramos at maury2 met-7.0]$
>>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
>>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
>>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
>>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>>>>>>>
>>>>>>> Would you check if ../out/ascii2nc exists?
>>>>>>>
>>>>>>> The common cause are 1) when the output directory does not
exist
>>>>>>> or
>>>>>>> 2) the user does not have the proper privileges (write
>> privilege).
>>>>>>> Would you check these?
>>>>>>>
>>>>>>> ls ../out/ascii2nc/
>>>>>>> touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Howard
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
>>>>>>> Code
>>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
>>>>>>>
>>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
>>>>>>>> Transaction: Ticket created by Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Queue: met_help
>>>>>>>> Subject: MET 6.1 bugfix
>>>>>>>> Owner: Nobody
>>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
>>>>>>>> Status: new
>>>>>>>> Ticket <URL:
>>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>>>>>>>>
>>>>>>>> I built the MET 6.1, but the tests won't run. My user is
>>>>>>>> getting the same error in trying to open the file for writing
>>>>>>>> using the
>>>>> tutorial:
>>>>>>>>
>>>>>>>> cd scripts; make
>>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
>>>>> bugfix/scripts'
>>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
>>>>>>>> NetCDF
>>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
>>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
>>>>>>>> ../out/ascii2nc/sample_ascii.nc \
>>>>>>>> -v 2
>>>>>>>> DEBUG 1: Config File:
>>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
>>>>>>>> Ascii2NcConfig_default
>>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
>>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
>>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
>>>>>>>> DEBUG 2: Processing observations for 5 headers.
>>>>>>>> DEBUG 1: Creating NetCDF Observation file:
>>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
>>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
>>>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
>>> ERROR :
>>>>>>>> make[1]: *** [ascii2nc] Error 1
>>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
>>>> bugfix/scripts'
>>>>>>>> make: *** [test] Error 2
>>>>>>>> [ramos at maury2 met-6.1_bugfix]$
>>>>>>>>
>>>>>>>> I am guessing this is a common error.
>>>>>>>>
>>>>>>>> Greg Ramos
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Wed Mar 07 13:58:16 2018
Greg,
Sorry, I'm not able to find an obvious solution for you.
In the last year or so, we began building MET within a docker
container.
Basically a single ascii file defines all the steps used to build MET,
along with all the dependent libraries. Here's that Dockerfile:
https://github.com/NCAR/container-dtc-
met/blob/master/MET/Dockerfile
Listed below is an excerpt where we download/compile NetCDF4. Looks
like
we're using netcdf-c-4.4.1.1 and netcdf-cxx4.4.3.0. Not sure if
that'll
help or not.
John
#
# NetCDF4 (c and c++)
#
RUN mkdir -p /met/external_libs \
&& cd /met/external_libs \
&& echo "Downloading netcdf-c-4.4.1.1 from ${NETCDF4C_URL}" \
&& wget ${NETCDF4C_URL} \
&& unzip v4.4.1.1.zip \
&& cd netcdf-c-4.4.1.1 \
&& LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/configure.log \
&& echo "Configuring netcdf-c-4.4.1.1 and writing log file
${LOG_FILE}" \
&& ./configure > ${LOG_FILE} \
&& LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/make_install.log \
&& echo "Compiling netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}"
\
&& make install > ${LOG_FILE} \
&& echo "Downloading from ${NETCDF4CXX_URL}" \
&& cd /met/external_libs \
&& wget ${NETCDF4CXX_URL} \
&& tar -xzf v4.3.0.tar.gz \
&& cd netcdf-cxx4-4.3.0 \
&& LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/configure.log \
&& echo "Configuring netcdf-cxx4-4.3.0 and writing log file
${LOG_FILE}" \
&& ./configure > ${LOG_FILE} \
&& LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/make_install.log \
&& echo "Compiling netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}"
\
&& make install > ${LOG_FILE}
On Tue, Mar 6, 2018 at 9:17 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This is the output of nc-config --all
>
> [ramos at maury2 ncview-2.1.7]$ /software/depot/netcdf-3.6.2018/bin/nc-
config
> --all
>
> This netCDF 4.6.0 has been built with the following features:
>
> --cc -> gcc
> --cflags -> -I/software/depot/netcdf-3.6.2018/include
> -I/software/depot/hdf5-1.10.1-gcc/include -I/software/depot/hdf-
4.2.10-
> rhel6-aerosol/include
> --libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
>
> --has-c++ -> no
> --cxx ->
>
> --has-c++4 -> yes
> --cxx4 -> g++
> --cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
> --cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib
-lnetcdf_c++4
> -lnetcdf
>
> --has-fortran-> no
> --has-dap -> no
> --has-dap4 -> no
> --has-nc2 -> yes
> --has-nc4 -> yes
> --has-hdf5 -> yes
> --has-hdf4 -> yes
> --has-logging-> no
> --has-pnetcdf-> no
> --has-szlib -> no
> --has-parallel -> no
> --has-cdf5 -> no
>
> --prefix -> /software/depot/netcdf-3.6.2018
> --includedir-> /software/depot/netcdf-3.6.2018/include
> --libdir -> /software/depot/netcdf-3.6.2018/lib
> --version -> netCDF 4.6.0
>
> [ramos at maury2 ncview-2.1.7]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 3:09 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> John, ldd ascii2nc will show the NetCDF library.
>
> NetCDF4 can be built without NetCDF4 features. This can be another
case
> that NetCDF file can not be opened (created ) because NetCDF was
created
> with NetCDF4 format.
>
> Other case will be different configuration on building MET and
executing
> MET.
>
> Building MET with NetCDF3 puzzles me. It should be detected by:
> Step 1. compile time with header files
> Step 2. link time with the library files.
>
> Cheers,
> Howard
>
>
> On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > Greg,
> >
> > No, the MET tools do not use ncdump. They use the C++ NetCDF
> > interface for reading/writing NetCDF files.
> >
> > John
> >
> > On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>
> >> Does ascii2nc use ncdump? Here is my NetCDF buld binary
directory:
> >>
> >> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
> >> total 576K
> >> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> >> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
ramos
> >> ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos 109K Mar
6
> >> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> >> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x 2
> >> ramos ramos 4.0K Mar 6 20:15 .
> >> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> >> [ramos at maury2 ~]$ ls -larth
> >> /software/depot/netcdf-3.6.2018/bin/ncdump
> >> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58
/software/depot/netcdf-3.6.
> >> 2018/bin/ncdump
> >> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
> >> linux-vdso.so.1 => (0x00007ffd951ea000)
> >> libnetcdf.so.13 =>
> >> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> >> (0x00002b3eb98fc000)
> >> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> >> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> >> libdf.so.0 =>
> >> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> >> (0x00002b3eb9e1a000)
> >> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
> >> libhdf5_hl.so.100 =>
> >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> >> (0x00002b3eba0e3000)
> >> libhdf5.so.101 =>
> >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> >> (0x00002b3eba306000)
> >> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> >> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> >> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> >> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> >> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> >> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
> >> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> >> [ramos at maury2 ~]$
> >>
> >> Greg
> >>
> >> -----Original Message-----
> >> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >> Sent: Tuesday, March 6, 2018 1:45 PM
> >> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>
> >> Greg,
> >>
> >> Ah, OK, I see that you're using NetCDF version 3.6. That's
likely
> >> the problem. As of MET version 6.0 we switched from NetCDF
version 3
> >> to NetCDF version 4:
> >>
> >> https://dtcenter.org/met/users/support/release_notes/
> >> METv6.0_release_notes.php
> >>
> >> Please try compiling against NetCDF4 instead of 3:
> >>
> >> https://dtcenter.org/met/users/support/online_tutorial/
> >> METv6.1/tutorial.php?name=compilation&category=req_libs
> >>
> >> I'm surprised MET compiled without error and that the problem
showed
> >> up at runtime.
> >>
> >> John
> >>
> >> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
Code
> >> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>
> >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>>
> >>> BTW, here are my build settings:
> >>>
> >>> [ramos at maury2 met-7.0]$ env |grep MET
> >>> MET_GSL=/software/depot/gsl-1.16
> >>> MET_ROOT=/software/depot/met-6.1_bugfix
> >>> MET_NETCDF=/software/depot/netcdf-3.6.2018
> >>> MET_NETCDFINC=
> >>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> >>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> >>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> >>> [ramos at maury2 met-7.0]$
> >>>
> >>> Greg
> >>>
> >>> -----Original Message-----
> >>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>> Sent: Tuesday, March 6, 2018 1:18 PM
> >>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>
> >>> Greg,
> >>>
> >>> Another thing to test would be the runtime linker. You set the
> >>> MET_NETCDF environment variable to tell MET where to find the
header
> >>> and library files for compilation. By default, MET is compiled
> >>> using dynamic linking. I suppose its possible that at runtime
the
> >>> MET executables are linking to some other version of NetCDF.
You
> >>> could set the LD_LIBRARY_PATH environment variable to define the
> >>> search path for the linker. Let's try
> >>> this:
> >>>
> >>> # Using c-shell syntax here. And MET_NETCDF is the top-level
> >>> directory for your NetCDF compilations
> >>>
> >>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> >>> /software/depot/met-7.0/bin/ascii2nc /users/ramos/met/met-
7.0/data/
> >>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> >>>
> >>> Any difference in results?
> >>>
> >>> Thanks,
> >>> John
> >>>
> >>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
Code
> >>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>
> >>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>>>
> >>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the
same
> >> error:
> >>>>
> >>>>
> >>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >>>>
> >>>> *** Running ASCII2NC to reformat ASCII point observations into
> >>>> NetCDF
> >>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>> ../out/ascii2nc/sample_ascii.nc \
> >>>> -v 2
> >>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> >>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
observations
> >>>> off the masking grid.
> >>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
> >>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
> >>>> DEBUG 2: Processing observations for 5 headers.
> >>>> DEBUG 1: Creating NetCDF Observation file:
> >>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> >>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> >>>> make[1]: *** [ascii2nc] Error 1
> >>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> >>>> make: *** [test] Error 2
> >>>> [ramos at maury2 met-7.0]$
> >>>>
> >>>> -----Original Message-----
> >>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>> Sent: Tuesday, March 6, 2018 10:14 AM
> >>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>
> >>>> Greg,
> >>>>
> >>>> To clarify, those backslashes '\' are line continuation
marks...
> >>>> telling the shell that the command is continued on the next
line.
> >>>> Including them on one line like that causes the odd behavior of
> >>>> creating an output file named "2". Removing them and running:
> >>>> /software/depot/met-7.0/bin/ascii2nc
> >>>> /users/ramos/met/met-7.0/data/
> >>>> sample_obs/ascii/sample_ascii_obs.txt
> >>>> ~/sample_ascii.nc -v 2 Will still result in an error... but
the
> >>>> requested output file will be what
> >>> you expect.
> >>>> So I'd try recompiling NetCDF using the same family/version
number
> >>>> of the compilers you're using to compile MET. Sorry I don't
have
> >>>> an easier solution for you.
> >>>>
> >>>> Thanks,
> >>>> John
> >>>>
> >>>>
> >>>>
> >>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory,
Contractor,
> >>>> Code
> >>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>
> >>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> >>>>>
> >>>>> I had to modify your example slightly, but it didn't work:
> >>>>>
> >>>>>
> >>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> >>>>> t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> >>>>> /software/depot/met-7.0/share/
met/config/Ascii2NcConfig_default
> >>>>> ERROR
> >>>>> :
> >>>>> ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> >>>>> ~/ sample_ascii.nc" for reading ERROR :
> >>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> >>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
> >>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> >>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>> file
> >>> "2"
> >>>>> for writing
> >>>>> ERROR :
> >>>>> [ramos at maury2 met-7.0]$
> >>>>>
> >>>>> Greg
> >>>>>
> >>>>> -----Original Message-----
> >>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>>> Sent: Tuesday, March 6, 2018 9:58 AM
> >>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>
> >>>>> Greg,
> >>>>>
> >>>>> The number of executables built depends on the way the build
was
> >>>>> configured. Options which require "extra" dependencies are
> >>>>> disabled by default. Here's an excerpt from the top-level
README
> >> file:
> >>>>> To configure MET to install all of the available tools in
the
> "bin"
> >>>>> subdirectory of your current directory, you would use the
> >>>>> following
> >>>>> command:
> >>>>>
> >>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> >>>>> --enable-mode_graphics --enable-lidar2nc
> >>>>>
> >>>>> So it's fine that you don't have all possible executables
built.
> >>>>> That's unrelated to the test script issue.
> >>>>>
> >>>>> My next guess would be an issue with the NetCDF library to
which
> >>>>> MET is linking. Let's replicate the error by trying to write
to
> >>>>> some other output directory, like your home directory:
> >>>>>
> >>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
> >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.tx
> >>>>> t
> >>>>> \ ~/ sample_ascii.nc \ -v 2
> >>>>>
> >>>>> I expect that this will fail with the same error message. If
it
> >>>>> does, you could try recompiling NetCDF... and then recompiling
MET.
> >>>>> Perhaps there's some incompatibility in how NetCDF and MET are
built?
> >>>>>
> >>>>> Here's some info about that:
> >>>>> https://dtcenter.org/met/users/support/online_tutorial/
> >>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
> >>>>>
> >>>>> Thanks,
> >>>>> John
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
Contractor,
> >>>>> Code
> >>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>
> >>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> >>>>>>
> >>>>>> These are the binaries that were built. It did not build all
of
> >>>>>> the binaries listed in the README:
> >>>>>>
> >>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> >>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> >>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> >>>>>> -rwxr-xr-x
> >>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> >>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
6
> >>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
16:44
> >>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
point_stat
> >>>>>> -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44 stat_analysis
> >>>>>> -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44 wavelet_stat -rwxr-
xr-x
> >>>>>> 1 ramos ramos 9.8M Mar 6
> >>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
16:44
> >>>>>> gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> >>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> >>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> >>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> >>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
> >>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
> >>>>>> ramos 8.9M Mar 6
> >>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> >>>>>> -rwxr-xr-x
> >>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
ramos
> >>>>>> 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos ramos 7.6M
Mar
> >>>>>> 6
> >>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> >>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> >>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> >>>>>> tc_pairs drwxrwxr-x 2 ramos
> >>>> ramos 4.0K Mar 6 16:44 .
> >>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> >>>>>> [ramos at maury2 met-7.0]$
> >>>>>>
> >>>>>> -----Original Message-----
> >>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
> >>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>>
> >>>>>> Hi Greg,
> >>>>>>
> >>>>>> This is John, I work with Howard. Well I'm glad to hear that
the
> >>>>>> behavior is consistent across versions. If not, I would have
> >>>>>> been
> >>>>> confused.
> >>>>>> So the very first command that's being run by the test script
is
> >>>>>> failing because it isn't able to create an output file.
> >>>>>>
> >>>>>> As Howard mentioned, the most likely cause for this type of
error
> >>>>>> is a permissions issue. So let's rule that out first.
> >>>>>> Please try running the following commands:
> >>>>>> cd /users/ramos/met/met-7.0/scripts
> >>>>>> touch ../out/ascii2nc/test_output_file
> >>>>>>
> >>>>>> The touch command just creates an empty file. If you get an
> >>>>>> error message about permissions, then we've found the
problem.
> >>>>>> I tried running the exact same ascii2nc command but writing
my
> >>>>>> output to "/usr/bin"... where of course I don't have
permission
> >>>>>> to
> >> write.
> >>>>>> Here's
> >>>>> the error message I got:
> >>>>>> cd met-7.0/scripts
> >>>>>> ../src/tools/other/ascii2nc/ascii2nc
> >>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
> >>>>>> /usr/bin/sample_ascii_obs.txt -v 2
> >>>>>>
> >>>>>> ERROR :
> >>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> >>>>>>
> >>>>>> Do a long listing of that directory to see the permissions:
> >>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
> >>>>>>
> >>>>>> What do you see?
> >>>>>>
> >>>>>> Thanks,
> >>>>>> John
> >>>>>>
> >>>>>>
> >>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> >>>>>> Code
> >>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>>
> >>>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >>>>>>> This morning I downloaded the latest version of MET. The
tests
> >>>>>>> still don't
> >>>>>>> run:
> >>>>>>>
> >>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
> >>>>>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> >>>>>>>
> >>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
> >>>>>>> NetCDF
> >>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>>>>> ../out/ascii2nc/sample_ascii.nc \
> >>>>>>> -v 2
> >>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> >>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> >>>>>>> observations off the masking grid.
> >>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> >>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>>>> DEBUG 1: Creating NetCDF Observation file:
> >>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
> >> ERROR :
> >>>>>>> make[1]: *** [ascii2nc] Error 1
> >>>>>>> make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> >>>>>>> make: *** [test] Error 2
> >>>>>>> [ramos at maury2 met-7.0]$
> >>>>>>>
> >>>>>>> -----Original Message-----
> >>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> >>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
> >>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>>>
> >>>>>>> Would you check if ../out/ascii2nc exists?
> >>>>>>>
> >>>>>>> The common cause are 1) when the output directory does not
exist
> >>>>>>> or
> >>>>>>> 2) the user does not have the proper privileges (write
> >> privilege).
> >>>>>>> Would you check these?
> >>>>>>>
> >>>>>>> ls ../out/ascii2nc/
> >>>>>>> touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
> >>>>>>>
> >>>>>>> Cheers,
> >>>>>>> Howard
> >>>>>>>
> >>>>>>>
> >>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
Contractor,
> >>>>>>> Code
> >>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>>>
> >>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> >>>>>>>> Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> >>>>>>>> Queue: met_help
> >>>>>>>> Subject: MET 6.1 bugfix
> >>>>>>>> Owner: Nobody
> >>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> >>>>>>>> Status: new
> >>>>>>>> Ticket <URL:
> >>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >>>>>>>>
> >>>>>>>> I built the MET 6.1, but the tests won't run. My user is
> >>>>>>>> getting the same error in trying to open the file for
writing
> >>>>>>>> using the
> >>>>> tutorial:
> >>>>>>>>
> >>>>>>>> cd scripts; make
> >>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
> >>>>> bugfix/scripts'
> >>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
> >>>>>>>> NetCDF
> >>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>>>>>> ../out/ascii2nc/sample_ascii.nc \
> >>>>>>>> -v 2
> >>>>>>>> DEBUG 1: Config File:
> >>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
> >>>>>>>> Ascii2NcConfig_default
> >>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> >>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> >>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>>>>> DEBUG 1: Creating NetCDF Observation file:
> >>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>>>>> file "../out/ascii2nc/sample_ascii.nc" for writing
> >>> ERROR :
> >>>>>>>> make[1]: *** [ascii2nc] Error 1
> >>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
> >>>> bugfix/scripts'
> >>>>>>>> make: *** [test] Error 2
> >>>>>>>> [ramos at maury2 met-6.1_bugfix]$
> >>>>>>>>
> >>>>>>>> I am guessing this is a common error.
> >>>>>>>>
> >>>>>>>> Greg Ramos
> >>>>>>>>
> >>>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>
> >>
> >>
>
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Wed Mar 07 15:43:28 2018
Attached are the results using the exact versions of NetCDF "C" and
"C++" specified in the instructions.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Wednesday, March 7, 2018 12:58 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Greg,
Sorry, I'm not able to find an obvious solution for you.
In the last year or so, we began building MET within a docker
container.
Basically a single ascii file defines all the steps used to build MET,
along with all the dependent libraries. Here's that Dockerfile:
https://github.com/NCAR/container-dtc-
met/blob/master/MET/Dockerfile
Listed below is an excerpt where we download/compile NetCDF4. Looks
like we're using netcdf-c-4.4.1.1 and netcdf-cxx4.4.3.0. Not sure if
that'll help or not.
John
#
# NetCDF4 (c and c++)
#
RUN mkdir -p /met/external_libs \
&& cd /met/external_libs \
&& echo "Downloading netcdf-c-4.4.1.1 from ${NETCDF4C_URL}" \ && wget
${NETCDF4C_URL} \ && unzip v4.4.1.1.zip \ && cd netcdf-c-4.4.1.1 \ &&
LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/configure.log \ && echo
"Configuring netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ &&
./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/netcdf-c-
4.4.1.1/make_install.log \ && echo "Compiling netcdf-c-4.4.1.1 and
writing log file ${LOG_FILE}" \ && make install > ${LOG_FILE} \ &&
echo "Downloading from ${NETCDF4CXX_URL}" \ && cd /met/external_libs \
&& wget ${NETCDF4CXX_URL} \ && tar -xzf v4.3.0.tar.gz \ && cd netcdf-
cxx4-4.3.0 \ && LOG_FILE=/met/external_libs/netcdf-cxx4-
4.3.0/configure.log \ && echo "Configuring netcdf-cxx4-4.3.0 and
writing log file ${LOG_FILE}" \ && ./configure > ${LOG_FILE} \ &&
LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/make_install.log \ &&
echo "Compiling netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \
&& make install > ${LOG_FILE}
On Tue, Mar 6, 2018 at 9:17 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> This is the output of nc-config --all
>
> [ramos at maury2 ncview-2.1.7]$
> /software/depot/netcdf-3.6.2018/bin/nc-config
> --all
>
> This netCDF 4.6.0 has been built with the following features:
>
> --cc -> gcc
> --cflags -> -I/software/depot/netcdf-3.6.2018/include
> -I/software/depot/hdf5-1.10.1-gcc/include
> -I/software/depot/hdf-4.2.10- rhel6-aerosol/include
> --libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
>
> --has-c++ -> no
> --cxx ->
>
> --has-c++4 -> yes
> --cxx4 -> g++
> --cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
> --cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib
-lnetcdf_c++4
> -lnetcdf
>
> --has-fortran-> no
> --has-dap -> no
> --has-dap4 -> no
> --has-nc2 -> yes
> --has-nc4 -> yes
> --has-hdf5 -> yes
> --has-hdf4 -> yes
> --has-logging-> no
> --has-pnetcdf-> no
> --has-szlib -> no
> --has-parallel -> no
> --has-cdf5 -> no
>
> --prefix -> /software/depot/netcdf-3.6.2018
> --includedir-> /software/depot/netcdf-3.6.2018/include
> --libdir -> /software/depot/netcdf-3.6.2018/lib
> --version -> netCDF 4.6.0
>
> [ramos at maury2 ncview-2.1.7]$
>
> -----Original Message-----
> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, March 6, 2018 3:09 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> John, ldd ascii2nc will show the NetCDF library.
>
> NetCDF4 can be built without NetCDF4 features. This can be another
> case that NetCDF file can not be opened (created ) because NetCDF
was
> created with NetCDF4 format.
>
> Other case will be different configuration on building MET and
> executing MET.
>
> Building MET with NetCDF3 puzzles me. It should be detected by:
> Step 1. compile time with header files Step 2. link time with the
> library files.
>
> Cheers,
> Howard
>
>
> On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > Greg,
> >
> > No, the MET tools do not use ncdump. They use the C++ NetCDF
> > interface for reading/writing NetCDF files.
> >
> > John
> >
> > On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor,
Code
> > 7501.1 via RT <met_help at ucar.edu> wrote:
> >
> >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>
> >> Does ascii2nc use ncdump? Here is my NetCDF buld binary
directory:
> >>
> >> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-3.6.2018/bin
> >> total 576K
> >> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> >> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
> >> ramos ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos 109K
> >> Mar 6
> >> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> >> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x 2
> >> ramos ramos 4.0K Mar 6 20:15 .
> >> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> >> [ramos at maury2 ~]$ ls -larth
> >> /software/depot/netcdf-3.6.2018/bin/ncdump
> >> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58
/software/depot/netcdf-3.6.
> >> 2018/bin/ncdump
> >> [ramos at maury2 ~]$ ldd /software/depot/netcdf-3.6.2018/bin/ncdump
> >> linux-vdso.so.1 => (0x00007ffd951ea000)
> >> libnetcdf.so.13 =>
> >> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> >> (0x00002b3eb98fc000)
> >> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> >> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> >> libdf.so.0 =>
> >> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> >> (0x00002b3eb9e1a000)
> >> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
(0x00000032a6200000)
> >> libhdf5_hl.so.100 =>
> >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> >> (0x00002b3eba0e3000)
> >> libhdf5.so.101 =>
> >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> >> (0x00002b3eba306000)
> >> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> >> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> >> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> >> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> >> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> >> libpthread.so.0 => /lib64/libpthread.so.0
(0x0000003299600000)
> >> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> >> [ramos at maury2 ~]$
> >>
> >> Greg
> >>
> >> -----Original Message-----
> >> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >> Sent: Tuesday, March 6, 2018 1:45 PM
> >> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>
> >> Greg,
> >>
> >> Ah, OK, I see that you're using NetCDF version 3.6. That's
likely
> >> the problem. As of MET version 6.0 we switched from NetCDF
version
> >> 3 to NetCDF version 4:
> >>
> >> https://dtcenter.org/met/users/support/release_notes/
> >> METv6.0_release_notes.php
> >>
> >> Please try compiling against NetCDF4 instead of 3:
> >>
> >> https://dtcenter.org/met/users/support/online_tutorial/
> >> METv6.1/tutorial.php?name=compilation&category=req_libs
> >>
> >> I'm surprised MET compiled without error and that the problem
> >> showed up at runtime.
> >>
> >> John
> >>
> >> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
> >> Code
> >> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>
> >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>>
> >>> BTW, here are my build settings:
> >>>
> >>> [ramos at maury2 met-7.0]$ env |grep MET
> >>> MET_GSL=/software/depot/gsl-1.16
> >>> MET_ROOT=/software/depot/met-6.1_bugfix
> >>> MET_NETCDF=/software/depot/netcdf-3.6.2018
> >>> MET_NETCDFINC=
> >>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> >>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> >>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> >>> [ramos at maury2 met-7.0]$
> >>>
> >>> Greg
> >>>
> >>> -----Original Message-----
> >>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>> Sent: Tuesday, March 6, 2018 1:18 PM
> >>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>
> >>> Greg,
> >>>
> >>> Another thing to test would be the runtime linker. You set the
> >>> MET_NETCDF environment variable to tell MET where to find the
> >>> header and library files for compilation. By default, MET is
> >>> compiled using dynamic linking. I suppose its possible that at
> >>> runtime the MET executables are linking to some other version of
> >>> NetCDF. You could set the LD_LIBRARY_PATH environment variable
to
> >>> define the search path for the linker. Let's try
> >>> this:
> >>>
> >>> # Using c-shell syntax here. And MET_NETCDF is the top-level
> >>> directory for your NetCDF compilations
> >>>
> >>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> >>> /software/depot/met-7.0/bin/ascii2nc
> >>> /users/ramos/met/met-7.0/data/
> >>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> >>>
> >>> Any difference in results?
> >>>
> >>> Thanks,
> >>> John
> >>>
> >>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory, Contractor,
> >>> Code
> >>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>
> >>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >>>>
> >>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the
same
> >> error:
> >>>>
> >>>>
> >>>> make[1]: Entering directory `/users/ramos/met/met-7.0/scripts'
> >>>>
> >>>> *** Running ASCII2NC to reformat ASCII point observations into
> >>>> NetCDF
> >>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>> ../out/ascii2nc/sample_ascii.nc \
> >>>> -v 2
> >>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> >>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> >>>> observations off the masking grid.
> >>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
> >>>> DEBUG 2: Rejected 0 observations not matched with station ID's.
> >>>> DEBUG 2: Processing observations for 5 headers.
> >>>> DEBUG 1: Creating NetCDF Observation file:
> >>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>> ERROR : FileHandler::_openNetcdf() -> can't open output NetCDF
> >>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> >>>> make[1]: *** [ascii2nc] Error 1
> >>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> >>>> make: *** [test] Error 2
> >>>> [ramos at maury2 met-7.0]$
> >>>>
> >>>> -----Original Message-----
> >>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>> Sent: Tuesday, March 6, 2018 10:14 AM
> >>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>
> >>>> Greg,
> >>>>
> >>>> To clarify, those backslashes '\' are line continuation
marks...
> >>>> telling the shell that the command is continued on the next
line.
> >>>> Including them on one line like that causes the odd behavior of
> >>>> creating an output file named "2". Removing them and running:
> >>>> /software/depot/met-7.0/bin/ascii2nc
> >>>> /users/ramos/met/met-7.0/data/
> >>>> sample_obs/ascii/sample_ascii_obs.txt
> >>>> ~/sample_ascii.nc -v 2 Will still result in an error... but
the
> >>>> requested output file will be what
> >>> you expect.
> >>>> So I'd try recompiling NetCDF using the same family/version
> >>>> number of the compilers you're using to compile MET. Sorry I
> >>>> don't have an easier solution for you.
> >>>>
> >>>> Thanks,
> >>>> John
> >>>>
> >>>>
> >>>>
> >>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory,
Contractor,
> >>>> Code
> >>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>
> >>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> >>>>>
> >>>>> I had to modify your example slightly, but it didn't work:
> >>>>>
> >>>>>
> >>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> >>>>> tx t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> >>>>> /software/depot/met-7.0/share/
met/config/Ascii2NcConfig_default
> >>>>> ERROR
> >>>>> :
> >>>>> ERROR : FileHandler::processFiles() -> can't open input ASCII
file "
> >>>>> ~/ sample_ascii.nc" for reading ERROR :
> >>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> >>>>> DEBUG 2: Rejected 0 observations outside the masking polyline.
> >>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> >>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>> file
> >>> "2"
> >>>>> for writing
> >>>>> ERROR :
> >>>>> [ramos at maury2 met-7.0]$
> >>>>>
> >>>>> Greg
> >>>>>
> >>>>> -----Original Message-----
> >>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>>> Sent: Tuesday, March 6, 2018 9:58 AM
> >>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>
> >>>>> Greg,
> >>>>>
> >>>>> The number of executables built depends on the way the build
was
> >>>>> configured. Options which require "extra" dependencies are
> >>>>> disabled by default. Here's an excerpt from the top-level
> >>>>> README
> >> file:
> >>>>> To configure MET to install all of the available tools in
the
> "bin"
> >>>>> subdirectory of your current directory, you would use the
> >>>>> following
> >>>>> command:
> >>>>>
> >>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis \
> >>>>> --enable-mode_graphics --enable-lidar2nc
> >>>>>
> >>>>> So it's fine that you don't have all possible executables
built.
> >>>>> That's unrelated to the test script issue.
> >>>>>
> >>>>> My next guess would be an issue with the NetCDF library to
which
> >>>>> MET is linking. Let's replicate the error by trying to write
to
> >>>>> some other output directory, like your home directory:
> >>>>>
> >>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
> >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> >>>>> tx
> >>>>> t
> >>>>> \ ~/ sample_ascii.nc \ -v 2
> >>>>>
> >>>>> I expect that this will fail with the same error message. If
it
> >>>>> does, you could try recompiling NetCDF... and then recompiling
MET.
> >>>>> Perhaps there's some incompatibility in how NetCDF and MET are
built?
> >>>>>
> >>>>> Here's some info about that:
> >>>>> https://dtcenter.org/met/users/support/online_tutorial/
> >>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
> >>>>>
> >>>>> Thanks,
> >>>>> John
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
Contractor,
> >>>>> Code
> >>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>
> >>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> >>>>>>
> >>>>>> These are the binaries that were built. It did not build all
of
> >>>>>> the binaries listed in the README:
> >>>>>>
> >>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> >>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> >>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> >>>>>> -rwxr-xr-x
> >>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1 ramos
> >>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M Mar
> >>>>>> 6
> >>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
16:44
> >>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> >>>>>> point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> >>>>>> stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> >>>>>> wavelet_stat -rwxr-xr-x
> >>>>>> 1 ramos ramos 9.8M Mar 6
> >>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
> >>>>>> 16:44 gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6 16:44
> >>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> >>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> >>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44 madis2nc
> >>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x 1
> >>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1 ramos
> >>>>>> ramos 8.9M Mar 6
> >>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> >>>>>> -rwxr-xr-x
> >>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> >>>>>> ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
ramos
> >>>>>> 7.6M Mar
> >>>>>> 6
> >>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar 6
> >>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> >>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> >>>>>> tc_pairs drwxrwxr-x 2 ramos
> >>>> ramos 4.0K Mar 6 16:44 .
> >>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> >>>>>> [ramos at maury2 met-7.0]$
> >>>>>>
> >>>>>> -----Original Message-----
> >>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> >>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
> >>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>>
> >>>>>> Hi Greg,
> >>>>>>
> >>>>>> This is John, I work with Howard. Well I'm glad to hear that
> >>>>>> the behavior is consistent across versions. If not, I would
> >>>>>> have been
> >>>>> confused.
> >>>>>> So the very first command that's being run by the test script
> >>>>>> is failing because it isn't able to create an output file.
> >>>>>>
> >>>>>> As Howard mentioned, the most likely cause for this type of
> >>>>>> error is a permissions issue. So let's rule that out first.
> >>>>>> Please try running the following commands:
> >>>>>> cd /users/ramos/met/met-7.0/scripts
> >>>>>> touch ../out/ascii2nc/test_output_file
> >>>>>>
> >>>>>> The touch command just creates an empty file. If you get an
> >>>>>> error message about permissions, then we've found the
problem.
> >>>>>> I tried running the exact same ascii2nc command but writing
my
> >>>>>> output to "/usr/bin"... where of course I don't have
permission
> >>>>>> to
> >> write.
> >>>>>> Here's
> >>>>> the error message I got:
> >>>>>> cd met-7.0/scripts
> >>>>>> ../src/tools/other/ascii2nc/ascii2nc
> >>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
> >>>>>> /usr/bin/sample_ascii_obs.txt -v 2
> >>>>>>
> >>>>>> ERROR :
> >>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> >>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> >>>>>>
> >>>>>> Do a long listing of that directory to see the permissions:
> >>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
> >>>>>>
> >>>>>> What do you see?
> >>>>>>
> >>>>>> Thanks,
> >>>>>> John
> >>>>>>
> >>>>>>
> >>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> >>>>>> Code
> >>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>>
> >>>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >>>>>>> This morning I downloaded the latest version of MET. The
tests
> >>>>>>> still don't
> >>>>>>> run:
> >>>>>>>
> >>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
> >>>>>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> >>>>>>>
> >>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
> >>>>>>> NetCDF
> >>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>>>>> ../out/ascii2nc/sample_ascii.nc \
> >>>>>>> -v 2
> >>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> >>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> >>>>>>> observations off the masking grid.
> >>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> >>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>>>> DEBUG 1: Creating NetCDF Observation file:
> >>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> >>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> >> ERROR :
> >>>>>>> make[1]: *** [ascii2nc] Error 1
> >>>>>>> make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> >>>>>>> make: *** [test] Error 2
> >>>>>>> [ramos at maury2 met-7.0]$
> >>>>>>>
> >>>>>>> -----Original Message-----
> >>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> >>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
> >>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> >>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >>>>>>>
> >>>>>>> Would you check if ../out/ascii2nc exists?
> >>>>>>>
> >>>>>>> The common cause are 1) when the output directory does not
> >>>>>>> exist or
> >>>>>>> 2) the user does not have the proper privileges (write
> >> privilege).
> >>>>>>> Would you check these?
> >>>>>>>
> >>>>>>> ls ../out/ascii2nc/
> >>>>>>> touch ../out/ascii2nc/test.txt rm ../out/ascii2nc/test.txt
> >>>>>>>
> >>>>>>> Cheers,
> >>>>>>> Howard
> >>>>>>>
> >>>>>>>
> >>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> >>>>>>> Contractor, Code
> >>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> >>>>>>>
> >>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> >>>>>>>> Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> >>>>>>>> Queue: met_help
> >>>>>>>> Subject: MET 6.1 bugfix
> >>>>>>>> Owner: Nobody
> >>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> >>>>>>>> Status: new
> >>>>>>>> Ticket <URL:
> >>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> >>>>>>>>
> >>>>>>>> I built the MET 6.1, but the tests won't run. My user is
> >>>>>>>> getting the same error in trying to open the file for
writing
> >>>>>>>> using the
> >>>>> tutorial:
> >>>>>>>>
> >>>>>>>> cd scripts; make
> >>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
> >>>>> bugfix/scripts'
> >>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> >>>>>>>> into NetCDF
> >>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> >>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> >>>>>>>> ../out/ascii2nc/sample_ascii.nc \
> >>>>>>>> -v 2
> >>>>>>>> DEBUG 1: Config File:
> >>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
> >>>>>>>> Ascii2NcConfig_default
> >>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> >>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> >>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> >>>>>>>> DEBUG 2: Processing observations for 5 headers.
> >>>>>>>> DEBUG 1: Creating NetCDF Observation file:
> >>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> >>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> >>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> >>> ERROR :
> >>>>>>>> make[1]: *** [ascii2nc] Error 1
> >>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
> >>>> bugfix/scripts'
> >>>>>>>> make: *** [test] Error 2
> >>>>>>>> [ramos at maury2 met-6.1_bugfix]$
> >>>>>>>>
> >>>>>>>> I am guessing this is a common error.
> >>>>>>>>
> >>>>>>>> Greg Ramos
> >>>>>>>>
> >>>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>
> >>
> >>
>
>
>
>
>
------------------------------------------------
Subject: MET 6.1 bugfix
From: John Halley Gotway
Time: Wed Mar 07 16:03:54 2018
Looks like that works well! No more runtime errors.
Sounds like on our end we should test building with NetCDF 4.6.0 to
see if
we can replicate the issues you ran into.
John
On Wed, Mar 7, 2018 at 3:43 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1
via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Attached are the results using the exact versions of NetCDF "C" and
"C++"
> specified in the instructions.
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Wednesday, March 7, 2018 12:58 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Sorry, I'm not able to find an obvious solution for you.
>
> In the last year or so, we began building MET within a docker
container.
> Basically a single ascii file defines all the steps used to build
MET,
> along with all the dependent libraries. Here's that Dockerfile:
> https://github.com/NCAR/container-dtc-
met/blob/master/MET/Dockerfile
>
> Listed below is an excerpt where we download/compile NetCDF4. Looks
like
> we're using netcdf-c-4.4.1.1 and netcdf-cxx4.4.3.0. Not sure if
that'll
> help or not.
>
> John
>
> #
> # NetCDF4 (c and c++)
> #
> RUN mkdir -p /met/external_libs \
> && cd /met/external_libs \
> && echo "Downloading netcdf-c-4.4.1.1 from ${NETCDF4C_URL}" \ &&
wget
> ${NETCDF4C_URL} \ && unzip v4.4.1.1.zip \ && cd netcdf-c-4.4.1.1 \
&&
> LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/configure.log \ && echo
> "Configuring netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ &&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-c-4.4.1.1/make_install.log \ && echo "Compiling netcdf-c-
4.4.1.1
> and writing log file ${LOG_FILE}" \ && make install > ${LOG_FILE} \
&& echo
> "Downloading from ${NETCDF4CXX_URL}" \ && cd /met/external_libs \ &&
wget
> ${NETCDF4CXX_URL} \ && tar -xzf v4.3.0.tar.gz \ && cd netcdf-cxx4-
4.3.0 \
> && LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/configure.log \ &&
echo
> "Configuring netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \
&&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-cxx4-4.3.0/make_install.log \ && echo "Compiling netcdf-cxx4-
4.3.0
> and writing log file ${LOG_FILE}" \ && make install > ${LOG_FILE}
>
> On Tue, Mar 6, 2018 at 9:17 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This is the output of nc-config --all
> >
> > [ramos at maury2 ncview-2.1.7]$
> > /software/depot/netcdf-3.6.2018/bin/nc-config
> > --all
> >
> > This netCDF 4.6.0 has been built with the following features:
> >
> > --cc -> gcc
> > --cflags -> -I/software/depot/netcdf-3.6.2018/include
> > -I/software/depot/hdf5-1.10.1-gcc/include
> > -I/software/depot/hdf-4.2.10- rhel6-aerosol/include
> > --libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
> >
> > --has-c++ -> no
> > --cxx ->
> >
> > --has-c++4 -> yes
> > --cxx4 -> g++
> > --cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
> > --cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib
-lnetcdf_c++4
> > -lnetcdf
> >
> > --has-fortran-> no
> > --has-dap -> no
> > --has-dap4 -> no
> > --has-nc2 -> yes
> > --has-nc4 -> yes
> > --has-hdf5 -> yes
> > --has-hdf4 -> yes
> > --has-logging-> no
> > --has-pnetcdf-> no
> > --has-szlib -> no
> > --has-parallel -> no
> > --has-cdf5 -> no
> >
> > --prefix -> /software/depot/netcdf-3.6.2018
> > --includedir-> /software/depot/netcdf-3.6.2018/include
> > --libdir -> /software/depot/netcdf-3.6.2018/lib
> > --version -> netCDF 4.6.0
> >
> > [ramos at maury2 ncview-2.1.7]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 3:09 PM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > John, ldd ascii2nc will show the NetCDF library.
> >
> > NetCDF4 can be built without NetCDF4 features. This can be another
> > case that NetCDF file can not be opened (created ) because NetCDF
was
> > created with NetCDF4 format.
> >
> > Other case will be different configuration on building MET and
> > executing MET.
> >
> > Building MET with NetCDF3 puzzles me. It should be detected by:
> > Step 1. compile time with header files Step 2. link time with the
> > library files.
> >
> > Cheers,
> > Howard
> >
> >
> > On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > Greg,
> > >
> > > No, the MET tools do not use ncdump. They use the C++ NetCDF
> > > interface for reading/writing NetCDF files.
> > >
> > > John
> > >
> > > On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor,
Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >>
> > >> Does ascii2nc use ncdump? Here is my NetCDF buld binary
directory:
> > >>
> > >> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin
> > >> total 576K
> > >> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> > >> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
> > >> ramos ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos
109K
> > >> Mar 6
> > >> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> > >> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x
2
> > >> ramos ramos 4.0K Mar 6 20:15 .
> > >> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> > >> [ramos at maury2 ~]$ ls -larth
> > >> /software/depot/netcdf-3.6.2018/bin/ncdump
> > >> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58
/software/depot/netcdf-3.6.
> > >> 2018/bin/ncdump
> > >> [ramos at maury2 ~]$ ldd /software/depot/netcdf-
3.6.2018/bin/ncdump
> > >> linux-vdso.so.1 => (0x00007ffd951ea000)
> > >> libnetcdf.so.13 =>
> > >> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> > >> (0x00002b3eb98fc000)
> > >> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> > >> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> > >> libdf.so.0 =>
> > >> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> > >> (0x00002b3eb9e1a000)
> > >> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
> (0x00000032a6200000)
> > >> libhdf5_hl.so.100 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> > >> (0x00002b3eba0e3000)
> > >> libhdf5.so.101 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> > >> (0x00002b3eba306000)
> > >> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> > >> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> > >> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> > >> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> > >> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> > >> libpthread.so.0 => /lib64/libpthread.so.0
> (0x0000003299600000)
> > >> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> > >> [ramos at maury2 ~]$
> > >>
> > >> Greg
> > >>
> > >> -----Original Message-----
> > >> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >> Sent: Tuesday, March 6, 2018 1:45 PM
> > >> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>
> > >> Greg,
> > >>
> > >> Ah, OK, I see that you're using NetCDF version 3.6. That's
likely
> > >> the problem. As of MET version 6.0 we switched from NetCDF
version
> > >> 3 to NetCDF version 4:
> > >>
> > >> https://dtcenter.org/met/users/support/release_notes/
> > >> METv6.0_release_notes.php
> > >>
> > >> Please try compiling against NetCDF4 instead of 3:
> > >>
> > >> https://dtcenter.org/met/users/support/online_tutorial/
> > >> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>
> > >> I'm surprised MET compiled without error and that the problem
> > >> showed up at runtime.
> > >>
> > >> John
> > >>
> > >> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
> > >> Code
> > >> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>
> > >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>
> > >>> BTW, here are my build settings:
> > >>>
> > >>> [ramos at maury2 met-7.0]$ env |grep MET
> > >>> MET_GSL=/software/depot/gsl-1.16
> > >>> MET_ROOT=/software/depot/met-6.1_bugfix
> > >>> MET_NETCDF=/software/depot/netcdf-3.6.2018
> > >>> MET_NETCDFINC=
> > >>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> > >>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> > >>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> > >>> [ramos at maury2 met-7.0]$
> > >>>
> > >>> Greg
> > >>>
> > >>> -----Original Message-----
> > >>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>> Sent: Tuesday, March 6, 2018 1:18 PM
> > >>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>
> > >>> Greg,
> > >>>
> > >>> Another thing to test would be the runtime linker. You set
the
> > >>> MET_NETCDF environment variable to tell MET where to find the
> > >>> header and library files for compilation. By default, MET is
> > >>> compiled using dynamic linking. I suppose its possible that
at
> > >>> runtime the MET executables are linking to some other version
of
> > >>> NetCDF. You could set the LD_LIBRARY_PATH environment
variable to
> > >>> define the search path for the linker. Let's try
> > >>> this:
> > >>>
> > >>> # Using c-shell syntax here. And MET_NETCDF is the top-level
> > >>> directory for your NetCDF compilations
> > >>>
> > >>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> > >>> /software/depot/met-7.0/bin/ascii2nc
> > >>> /users/ramos/met/met-7.0/data/
> > >>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> > >>>
> > >>> Any difference in results?
> > >>>
> > >>> Thanks,
> > >>> John
> > >>>
> > >>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory,
Contractor,
> > >>> Code
> > >>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>
> > >>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>>
> > >>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the
same
> > >> error:
> > >>>>
> > >>>>
> > >>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>
> > >>>> *** Running ASCII2NC to reformat ASCII point observations
into
> > >>>> NetCDF
> > >>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>> -v 2
> > >>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>> observations off the masking grid.
> > >>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>> DEBUG 2: Processing observations for 5 headers.
> > >>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > >>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > >>>> make[1]: *** [ascii2nc] Error 1
> > >>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > >>>> make: *** [test] Error 2
> > >>>> [ramos at maury2 met-7.0]$
> > >>>>
> > >>>> -----Original Message-----
> > >>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>> Sent: Tuesday, March 6, 2018 10:14 AM
> > >>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>
> > >>>> Greg,
> > >>>>
> > >>>> To clarify, those backslashes '\' are line continuation
marks...
> > >>>> telling the shell that the command is continued on the next
line.
> > >>>> Including them on one line like that causes the odd behavior
of
> > >>>> creating an output file named "2". Removing them and
running:
> > >>>> /software/depot/met-7.0/bin/ascii2nc
> > >>>> /users/ramos/met/met-7.0/data/
> > >>>> sample_obs/ascii/sample_ascii_obs.txt
> > >>>> ~/sample_ascii.nc -v 2 Will still result in an error... but
the
> > >>>> requested output file will be what
> > >>> you expect.
> > >>>> So I'd try recompiling NetCDF using the same family/version
> > >>>> number of the compilers you're using to compile MET. Sorry I
> > >>>> don't have an easier solution for you.
> > >>>>
> > >>>> Thanks,
> > >>>> John
> > >>>>
> > >>>>
> > >>>>
> > >>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory,
Contractor,
> > >>>> Code
> > >>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>
> > >>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >>>>>
> > >>>>> I had to modify your example slightly, but it didn't work:
> > >>>>>
> > >>>>>
> > >>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > >>>>> /software/depot/met-7.0/share/
met/config/Ascii2NcConfig_default
> > >>>>> ERROR
> > >>>>> :
> > >>>>> ERROR : FileHandler::processFiles() -> can't open input
ASCII
> file "
> > >>>>> ~/ sample_ascii.nc" for reading ERROR :
> > >>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > >>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > >>>>> file
> > >>> "2"
> > >>>>> for writing
> > >>>>> ERROR :
> > >>>>> [ramos at maury2 met-7.0]$
> > >>>>>
> > >>>>> Greg
> > >>>>>
> > >>>>> -----Original Message-----
> > >>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>> Sent: Tuesday, March 6, 2018 9:58 AM
> > >>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>
> > >>>>> Greg,
> > >>>>>
> > >>>>> The number of executables built depends on the way the build
was
> > >>>>> configured. Options which require "extra" dependencies are
> > >>>>> disabled by default. Here's an excerpt from the top-level
> > >>>>> README
> > >> file:
> > >>>>> To configure MET to install all of the available tools in
the
> > "bin"
> > >>>>> subdirectory of your current directory, you would use the
> > >>>>> following
> > >>>>> command:
> > >>>>>
> > >>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis
\
> > >>>>> --enable-mode_graphics --enable-lidar2nc
> > >>>>>
> > >>>>> So it's fine that you don't have all possible executables
built.
> > >>>>> That's unrelated to the test script issue.
> > >>>>>
> > >>>>> My next guess would be an issue with the NetCDF library to
which
> > >>>>> MET is linking. Let's replicate the error by trying to
write to
> > >>>>> some other output directory, like your home directory:
> > >>>>>
> > >>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx
> > >>>>> t
> > >>>>> \ ~/ sample_ascii.nc \ -v 2
> > >>>>>
> > >>>>> I expect that this will fail with the same error message.
If it
> > >>>>> does, you could try recompiling NetCDF... and then
recompiling MET.
> > >>>>> Perhaps there's some incompatibility in how NetCDF and MET
are
> built?
> > >>>>>
> > >>>>> Here's some info about that:
> > >>>>> https://dtcenter.org/met/users/support/online_tutorial/
> > >>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>>>>
> > >>>>> Thanks,
> > >>>>> John
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
Contractor,
> > >>>>> Code
> > >>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>
> > >>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >>>>>>
> > >>>>>> These are the binaries that were built. It did not build
all of
> > >>>>>> the binaries listed in the README:
> > >>>>>>
> > >>>>>> [ramos at maury2 met-7.0]$ ls -larth /software/depot/met-
7.0/bin/
> > >>>>>> total 208M drwxr-xr-x 4 ramos ramos 4.0K Mar 6 16:44 ..
> > >>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > >>>>>> -rwxr-xr-x
> > >>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1
ramos
> > >>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M
Mar
> > >>>>>> 6
> > >>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
16:44
> > >>>>>> pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44
> > >>>>>> point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > >>>>>> stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > >>>>>> wavelet_stat -rwxr-xr-x
> > >>>>>> 1 ramos ramos 9.8M Mar 6
> > >>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
> > >>>>>> 16:44 gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6
16:44
> > >>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > >>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > >>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44
madis2nc
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x
1
> > >>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
ramos
> > >>>>>> ramos 8.9M Mar 6
> > >>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
mtd
> > >>>>>> -rwxr-xr-x
> > >>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > >>>>>> ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
ramos
> > >>>>>> 7.6M Mar
> > >>>>>> 6
> > >>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar
6
> > >>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > >>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > >>>>>> tc_pairs drwxrwxr-x 2 ramos
> > >>>> ramos 4.0K Mar 6 16:44 .
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > >>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>
> > >>>>>> -----Original Message-----
> > >>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
> > >>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>
> > >>>>>> Hi Greg,
> > >>>>>>
> > >>>>>> This is John, I work with Howard. Well I'm glad to hear
that
> > >>>>>> the behavior is consistent across versions. If not, I
would
> > >>>>>> have been
> > >>>>> confused.
> > >>>>>> So the very first command that's being run by the test
script
> > >>>>>> is failing because it isn't able to create an output file.
> > >>>>>>
> > >>>>>> As Howard mentioned, the most likely cause for this type of
> > >>>>>> error is a permissions issue. So let's rule that out
first.
> > >>>>>> Please try running the following commands:
> > >>>>>> cd /users/ramos/met/met-7.0/scripts
> > >>>>>> touch ../out/ascii2nc/test_output_file
> > >>>>>>
> > >>>>>> The touch command just creates an empty file. If you get
an
> > >>>>>> error message about permissions, then we've found the
problem.
> > >>>>>> I tried running the exact same ascii2nc command but writing
my
> > >>>>>> output to "/usr/bin"... where of course I don't have
permission
> > >>>>>> to
> > >> write.
> > >>>>>> Here's
> > >>>>> the error message I got:
> > >>>>>> cd met-7.0/scripts
> > >>>>>> ../src/tools/other/ascii2nc/ascii2nc
> > >>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
> > >>>>>> /usr/bin/sample_ascii_obs.txt -v 2
> > >>>>>>
> > >>>>>> ERROR :
> > >>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > >>>>>> file "/usr/bin/sample_ascii_obs.txt" for writing ERROR :
> > >>>>>>
> > >>>>>> Do a long listing of that directory to see the permissions:
> > >>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >>>>>>
> > >>>>>> What do you see?
> > >>>>>>
> > >>>>>> Thanks,
> > >>>>>> John
> > >>>>>>
> > >>>>>>
> > >>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
Contractor,
> > >>>>>> Code
> > >>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>
> > >>>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>> This morning I downloaded the latest version of MET. The
tests
> > >>>>>>> still don't
> > >>>>>>> run:
> > >>>>>>>
> > >>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
> > >>>>>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>>
> > >>>>>>> *** Running ASCII2NC to reformat ASCII point observations
into
> > >>>>>>> NetCDF
> > >>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>> -v 2
> > >>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>>>>> observations off the masking grid.
> > >>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >> ERROR :
> > >>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>> make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>> make: *** [test] Error 2
> > >>>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>>
> > >>>>>>> -----Original Message-----
> > >>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > >>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
> > >>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>>
> > >>>>>>> Would you check if ../out/ascii2nc exists?
> > >>>>>>>
> > >>>>>>> The common cause are 1) when the output directory does not
> > >>>>>>> exist or
> > >>>>>>> 2) the user does not have the proper privileges (write
> > >> privilege).
> > >>>>>>> Would you check these?
> > >>>>>>>
> > >>>>>>> ls ../out/ascii2nc/
> > >>>>>>> touch ../out/ascii2nc/test.txt rm
../out/ascii2nc/test.txt
> > >>>>>>>
> > >>>>>>> Cheers,
> > >>>>>>> Howard
> > >>>>>>>
> > >>>>>>>
> > >>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > >>>>>>> Contractor, Code
> > >>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>>
> > >>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > >>>>>>>> Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Queue: met_help
> > >>>>>>>> Subject: MET 6.1 bugfix
> > >>>>>>>> Owner: Nobody
> > >>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Status: new
> > >>>>>>>> Ticket <URL:
> > >>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>>>
> > >>>>>>>> I built the MET 6.1, but the tests won't run. My user is
> > >>>>>>>> getting the same error in trying to open the file for
writing
> > >>>>>>>> using the
> > >>>>> tutorial:
> > >>>>>>>>
> > >>>>>>>> cd scripts; make
> > >>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
> > >>>>> bugfix/scripts'
> > >>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> > >>>>>>>> into NetCDF
> > >>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>>> -v 2
> > >>>>>>>> DEBUG 1: Config File:
> > >>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
> > >>>>>>>> Ascii2NcConfig_default
> > >>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >>> ERROR :
> > >>>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > >>>> bugfix/scripts'
> > >>>>>>>> make: *** [test] Error 2
> > >>>>>>>> [ramos at maury2 met-6.1_bugfix]$
> > >>>>>>>>
> > >>>>>>>> I am guessing this is a common error.
> > >>>>>>>>
> > >>>>>>>> Greg Ramos
> > >>>>>>>>
> > >>>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>
> > >>>
> > >>>
> > >>
> > >>
> > >>
> >
> >
> >
> >
> >
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Wed Mar 07 16:06:43 2018
I figured I could just download and compile the latest versions, then
I went back and used the exact versions listed using the wget.
My user's test using the tutorial worked as well.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Wednesday, March 7, 2018 3:04 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Looks like that works well! No more runtime errors.
Sounds like on our end we should test building with NetCDF 4.6.0 to
see if we can replicate the issues you ran into.
John
On Wed, Mar 7, 2018 at 3:43 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Attached are the results using the exact versions of NetCDF "C" and
"C++"
> specified in the instructions.
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Wednesday, March 7, 2018 12:58 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Sorry, I'm not able to find an obvious solution for you.
>
> In the last year or so, we began building MET within a docker
container.
> Basically a single ascii file defines all the steps used to build
MET,
> along with all the dependent libraries. Here's that Dockerfile:
>
> https://github.com/NCAR/container-dtc-met/blob/master/MET/Dockerfile
>
> Listed below is an excerpt where we download/compile NetCDF4. Looks
> like we're using netcdf-c-4.4.1.1 and netcdf-cxx4.4.3.0. Not sure
if
> that'll help or not.
>
> John
>
> #
> # NetCDF4 (c and c++)
> #
> RUN mkdir -p /met/external_libs \
> && cd /met/external_libs \
> && echo "Downloading netcdf-c-4.4.1.1 from ${NETCDF4C_URL}" \ &&
wget
> ${NETCDF4C_URL} \ && unzip v4.4.1.1.zip \ && cd netcdf-c-4.4.1.1 \
&&
> LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/configure.log \ && echo
> "Configuring netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ &&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-c-4.4.1.1/make_install.log \ && echo "Compiling
> netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ && make install
>
> ${LOG_FILE} \ && echo "Downloading from ${NETCDF4CXX_URL}" \ && cd
> /met/external_libs \ && wget ${NETCDF4CXX_URL} \ && tar -xzf
> v4.3.0.tar.gz \ && cd netcdf-cxx4-4.3.0 \ &&
> LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/configure.log \ &&
echo
> "Configuring netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \
&&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-cxx4-4.3.0/make_install.log \ && echo "Compiling
> netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \ && make
install
> > ${LOG_FILE}
>
> On Tue, Mar 6, 2018 at 9:17 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This is the output of nc-config --all
> >
> > [ramos at maury2 ncview-2.1.7]$
> > /software/depot/netcdf-3.6.2018/bin/nc-config
> > --all
> >
> > This netCDF 4.6.0 has been built with the following features:
> >
> > --cc -> gcc
> > --cflags -> -I/software/depot/netcdf-3.6.2018/include
> > -I/software/depot/hdf5-1.10.1-gcc/include
> > -I/software/depot/hdf-4.2.10- rhel6-aerosol/include
> > --libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
> >
> > --has-c++ -> no
> > --cxx ->
> >
> > --has-c++4 -> yes
> > --cxx4 -> g++
> > --cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
> > --cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib
> > -lnetcdf_c++4 -lnetcdf
> >
> > --has-fortran-> no
> > --has-dap -> no
> > --has-dap4 -> no
> > --has-nc2 -> yes
> > --has-nc4 -> yes
> > --has-hdf5 -> yes
> > --has-hdf4 -> yes
> > --has-logging-> no
> > --has-pnetcdf-> no
> > --has-szlib -> no
> > --has-parallel -> no
> > --has-cdf5 -> no
> >
> > --prefix -> /software/depot/netcdf-3.6.2018
> > --includedir-> /software/depot/netcdf-3.6.2018/include
> > --libdir -> /software/depot/netcdf-3.6.2018/lib
> > --version -> netCDF 4.6.0
> >
> > [ramos at maury2 ncview-2.1.7]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 3:09 PM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > John, ldd ascii2nc will show the NetCDF library.
> >
> > NetCDF4 can be built without NetCDF4 features. This can be another
> > case that NetCDF file can not be opened (created ) because NetCDF
> > was created with NetCDF4 format.
> >
> > Other case will be different configuration on building MET and
> > executing MET.
> >
> > Building MET with NetCDF3 puzzles me. It should be detected by:
> > Step 1. compile time with header files Step 2. link time with the
> > library files.
> >
> > Cheers,
> > Howard
> >
> >
> > On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > Greg,
> > >
> > > No, the MET tools do not use ncdump. They use the C++ NetCDF
> > > interface for reading/writing NetCDF files.
> > >
> > > John
> > >
> > > On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >>
> > >> Does ascii2nc use ncdump? Here is my NetCDF buld binary
directory:
> > >>
> > >> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin
> > >> total 576K
> > >> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> > >> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
> > >> ramos ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos
109K
> > >> Mar 6
> > >> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> > >> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x
2
> > >> ramos ramos 4.0K Mar 6 20:15 .
> > >> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> > >> [ramos at maury2 ~]$ ls -larth
> > >> /software/depot/netcdf-3.6.2018/bin/ncdump
> > >> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58
/software/depot/netcdf-3.6.
> > >> 2018/bin/ncdump
> > >> [ramos at maury2 ~]$ ldd /software/depot/netcdf-
3.6.2018/bin/ncdump
> > >> linux-vdso.so.1 => (0x00007ffd951ea000)
> > >> libnetcdf.so.13 =>
> > >> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> > >> (0x00002b3eb98fc000)
> > >> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> > >> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> > >> libdf.so.0 =>
> > >> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> > >> (0x00002b3eb9e1a000)
> > >> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
> (0x00000032a6200000)
> > >> libhdf5_hl.so.100 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> > >> (0x00002b3eba0e3000)
> > >> libhdf5.so.101 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> > >> (0x00002b3eba306000)
> > >> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> > >> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> > >> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> > >> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> > >> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> > >> libpthread.so.0 => /lib64/libpthread.so.0
> (0x0000003299600000)
> > >> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> > >> [ramos at maury2 ~]$
> > >>
> > >> Greg
> > >>
> > >> -----Original Message-----
> > >> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >> Sent: Tuesday, March 6, 2018 1:45 PM
> > >> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>
> > >> Greg,
> > >>
> > >> Ah, OK, I see that you're using NetCDF version 3.6. That's
> > >> likely the problem. As of MET version 6.0 we switched from
> > >> NetCDF version
> > >> 3 to NetCDF version 4:
> > >>
> > >> https://dtcenter.org/met/users/support/release_notes/
> > >> METv6.0_release_notes.php
> > >>
> > >> Please try compiling against NetCDF4 instead of 3:
> > >>
> > >> https://dtcenter.org/met/users/support/online_tutorial/
> > >> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>
> > >> I'm surprised MET compiled without error and that the problem
> > >> showed up at runtime.
> > >>
> > >> John
> > >>
> > >> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
> > >> Code
> > >> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>
> > >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>
> > >>> BTW, here are my build settings:
> > >>>
> > >>> [ramos at maury2 met-7.0]$ env |grep MET
> > >>> MET_GSL=/software/depot/gsl-1.16
> > >>> MET_ROOT=/software/depot/met-6.1_bugfix
> > >>> MET_NETCDF=/software/depot/netcdf-3.6.2018
> > >>> MET_NETCDFINC=
> > >>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> > >>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> > >>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> > >>> [ramos at maury2 met-7.0]$
> > >>>
> > >>> Greg
> > >>>
> > >>> -----Original Message-----
> > >>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>> Sent: Tuesday, March 6, 2018 1:18 PM
> > >>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>
> > >>> Greg,
> > >>>
> > >>> Another thing to test would be the runtime linker. You set
the
> > >>> MET_NETCDF environment variable to tell MET where to find the
> > >>> header and library files for compilation. By default, MET is
> > >>> compiled using dynamic linking. I suppose its possible that
at
> > >>> runtime the MET executables are linking to some other version
of
> > >>> NetCDF. You could set the LD_LIBRARY_PATH environment
variable
> > >>> to define the search path for the linker. Let's try
> > >>> this:
> > >>>
> > >>> # Using c-shell syntax here. And MET_NETCDF is the top-level
> > >>> directory for your NetCDF compilations
> > >>>
> > >>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> > >>> /software/depot/met-7.0/bin/ascii2nc
> > >>> /users/ramos/met/met-7.0/data/
> > >>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> > >>>
> > >>> Any difference in results?
> > >>>
> > >>> Thanks,
> > >>> John
> > >>>
> > >>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory,
Contractor,
> > >>> Code
> > >>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>
> > >>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>>
> > >>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the
> > >>>> same
> > >> error:
> > >>>>
> > >>>>
> > >>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>
> > >>>> *** Running ASCII2NC to reformat ASCII point observations
into
> > >>>> NetCDF
> > >>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>> -v 2
> > >>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>> observations off the masking grid.
> > >>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>> DEBUG 2: Processing observations for 5 headers.
> > >>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > >>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > >>>> make[1]: *** [ascii2nc] Error 1
> > >>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > >>>> make: *** [test] Error 2
> > >>>> [ramos at maury2 met-7.0]$
> > >>>>
> > >>>> -----Original Message-----
> > >>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>> Sent: Tuesday, March 6, 2018 10:14 AM
> > >>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>
> > >>>> Greg,
> > >>>>
> > >>>> To clarify, those backslashes '\' are line continuation
marks...
> > >>>> telling the shell that the command is continued on the next
line.
> > >>>> Including them on one line like that causes the odd behavior
of
> > >>>> creating an output file named "2". Removing them and
running:
> > >>>> /software/depot/met-7.0/bin/ascii2nc
> > >>>> /users/ramos/met/met-7.0/data/
> > >>>> sample_obs/ascii/sample_ascii_obs.txt
> > >>>> ~/sample_ascii.nc -v 2 Will still result in an error... but
> > >>>> the requested output file will be what
> > >>> you expect.
> > >>>> So I'd try recompiling NetCDF using the same family/version
> > >>>> number of the compilers you're using to compile MET. Sorry I
> > >>>> don't have an easier solution for you.
> > >>>>
> > >>>> Thanks,
> > >>>> John
> > >>>>
> > >>>>
> > >>>>
> > >>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory,
> > >>>> Contractor, Code
> > >>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>
> > >>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>> >
> > >>>>>
> > >>>>> I had to modify your example slightly, but it didn't work:
> > >>>>>
> > >>>>>
> > >>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > >>>>> /software/depot/met-7.0/share/
> > >>>>> met/config/Ascii2NcConfig_default ERROR
> > >>>>> :
> > >>>>> ERROR : FileHandler::processFiles() -> can't open input
ASCII
> file "
> > >>>>> ~/ sample_ascii.nc" for reading ERROR :
> > >>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > >>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>> NetCDF file
> > >>> "2"
> > >>>>> for writing
> > >>>>> ERROR :
> > >>>>> [ramos at maury2 met-7.0]$
> > >>>>>
> > >>>>> Greg
> > >>>>>
> > >>>>> -----Original Message-----
> > >>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>> Sent: Tuesday, March 6, 2018 9:58 AM
> > >>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>
> > >>>>> Greg,
> > >>>>>
> > >>>>> The number of executables built depends on the way the build
> > >>>>> was configured. Options which require "extra" dependencies
> > >>>>> are disabled by default. Here's an excerpt from the top-
level
> > >>>>> README
> > >> file:
> > >>>>> To configure MET to install all of the available tools in
> > >>>>> the
> > "bin"
> > >>>>> subdirectory of your current directory, you would use the
> > >>>>> following
> > >>>>> command:
> > >>>>>
> > >>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis
\
> > >>>>> --enable-mode_graphics --enable-lidar2nc
> > >>>>>
> > >>>>> So it's fine that you don't have all possible executables
built.
> > >>>>> That's unrelated to the test script issue.
> > >>>>>
> > >>>>> My next guess would be an issue with the NetCDF library to
> > >>>>> which MET is linking. Let's replicate the error by trying
to
> > >>>>> write to some other output directory, like your home
directory:
> > >>>>>
> > >>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx
> > >>>>> t
> > >>>>> \ ~/ sample_ascii.nc \ -v 2
> > >>>>>
> > >>>>> I expect that this will fail with the same error message.
If
> > >>>>> it does, you could try recompiling NetCDF... and then
recompiling MET.
> > >>>>> Perhaps there's some incompatibility in how NetCDF and MET
are
> built?
> > >>>>>
> > >>>>> Here's some info about that:
> > >>>>> https://dtcenter.org/met/users/support/online_tutorial/
> > >>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>>>>
> > >>>>> Thanks,
> > >>>>> John
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
> > >>>>> Contractor, Code
> > >>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>
> > >>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>> >
> > >>>>>>
> > >>>>>> These are the binaries that were built. It did not build
all
> > >>>>>> of the binaries listed in the README:
> > >>>>>>
> > >>>>>> [ramos at maury2 met-7.0]$ ls -larth
> > >>>>>> /software/depot/met-7.0/bin/ total 208M drwxr-xr-x 4 ramos
ramos 4.0K Mar 6 16:44 ..
> > >>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > >>>>>> -rwxr-xr-x
> > >>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1
ramos
> > >>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M
Mar
> > >>>>>> 6
> > >>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
> > >>>>>> 16:44 pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6
16:44
> > >>>>>> point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > >>>>>> stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > >>>>>> wavelet_stat -rwxr-xr-x
> > >>>>>> 1 ramos ramos 9.8M Mar 6
> > >>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
> > >>>>>> 16:44 gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6
16:44
> > >>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > >>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > >>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44
madis2nc
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x
1
> > >>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
ramos
> > >>>>>> ramos 8.9M Mar 6
> > >>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
> > >>>>>> mtd -rwxr-xr-x
> > >>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > >>>>>> ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
> > >>>>>> ramos 7.6M Mar
> > >>>>>> 6
> > >>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar
6
> > >>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > >>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > >>>>>> tc_pairs drwxrwxr-x 2 ramos
> > >>>> ramos 4.0K Mar 6 16:44 .
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > >>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>
> > >>>>>> -----Original Message-----
> > >>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
> > >>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>
> > >>>>>> Hi Greg,
> > >>>>>>
> > >>>>>> This is John, I work with Howard. Well I'm glad to hear
that
> > >>>>>> the behavior is consistent across versions. If not, I
would
> > >>>>>> have been
> > >>>>> confused.
> > >>>>>> So the very first command that's being run by the test
script
> > >>>>>> is failing because it isn't able to create an output file.
> > >>>>>>
> > >>>>>> As Howard mentioned, the most likely cause for this type of
> > >>>>>> error is a permissions issue. So let's rule that out
first.
> > >>>>>> Please try running the following commands:
> > >>>>>> cd /users/ramos/met/met-7.0/scripts
> > >>>>>> touch ../out/ascii2nc/test_output_file
> > >>>>>>
> > >>>>>> The touch command just creates an empty file. If you get
an
> > >>>>>> error message about permissions, then we've found the
problem.
> > >>>>>> I tried running the exact same ascii2nc command but writing
> > >>>>>> my output to "/usr/bin"... where of course I don't have
> > >>>>>> permission to
> > >> write.
> > >>>>>> Here's
> > >>>>> the error message I got:
> > >>>>>> cd met-7.0/scripts
> > >>>>>> ../src/tools/other/ascii2nc/ascii2nc
> > >>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
> > >>>>>> /usr/bin/sample_ascii_obs.txt -v 2
> > >>>>>>
> > >>>>>> ERROR :
> > >>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>> NetCDF file "/usr/bin/sample_ascii_obs.txt" for writing
ERROR :
> > >>>>>>
> > >>>>>> Do a long listing of that directory to see the permissions:
> > >>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >>>>>>
> > >>>>>> What do you see?
> > >>>>>>
> > >>>>>> Thanks,
> > >>>>>> John
> > >>>>>>
> > >>>>>>
> > >>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
> > >>>>>> Contractor, Code
> > >>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>
> > >>>>>>> <URL:
> > >>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>> This morning I downloaded the latest version of MET. The
> > >>>>>>> tests still don't
> > >>>>>>> run:
> > >>>>>>>
> > >>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
> > >>>>>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>>
> > >>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> > >>>>>>> into NetCDF
> > >>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>> -v 2
> > >>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>>>>> observations off the masking grid.
> > >>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >> ERROR :
> > >>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>> make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>> make: *** [test] Error 2
> > >>>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>>
> > >>>>>>> -----Original Message-----
> > >>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > >>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
> > >>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>>
> > >>>>>>> Would you check if ../out/ascii2nc exists?
> > >>>>>>>
> > >>>>>>> The common cause are 1) when the output directory does not
> > >>>>>>> exist or
> > >>>>>>> 2) the user does not have the proper privileges (write
> > >> privilege).
> > >>>>>>> Would you check these?
> > >>>>>>>
> > >>>>>>> ls ../out/ascii2nc/
> > >>>>>>> touch ../out/ascii2nc/test.txt rm
../out/ascii2nc/test.txt
> > >>>>>>>
> > >>>>>>> Cheers,
> > >>>>>>> Howard
> > >>>>>>>
> > >>>>>>>
> > >>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > >>>>>>> Contractor, Code
> > >>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>>
> > >>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > >>>>>>>> Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Queue: met_help
> > >>>>>>>> Subject: MET 6.1 bugfix
> > >>>>>>>> Owner: Nobody
> > >>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Status: new
> > >>>>>>>> Ticket <URL:
> > >>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>>>
> > >>>>>>>> I built the MET 6.1, but the tests won't run. My user is
> > >>>>>>>> getting the same error in trying to open the file for
> > >>>>>>>> writing using the
> > >>>>> tutorial:
> > >>>>>>>>
> > >>>>>>>> cd scripts; make
> > >>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
> > >>>>> bugfix/scripts'
> > >>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> > >>>>>>>> into NetCDF
> > >>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>>> -v 2
> > >>>>>>>> DEBUG 1: Config File:
> > >>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
> > >>>>>>>> Ascii2NcConfig_default
> > >>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >>> ERROR :
> > >>>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > >>>> bugfix/scripts'
> > >>>>>>>> make: *** [test] Error 2
> > >>>>>>>> [ramos at maury2 met-6.1_bugfix]$
> > >>>>>>>>
> > >>>>>>>> I am guessing this is a common error.
> > >>>>>>>>
> > >>>>>>>> Greg Ramos
> > >>>>>>>>
> > >>>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>
> > >>>
> > >>>
> > >>
> > >>
> > >>
> >
> >
> >
> >
> >
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
From: Ramos, Mr. Gregory, Contractor, Code 7501.1
Time: Wed Mar 07 16:21:51 2018
One other glitch with the test. After building MET 7.0, I tried to run
the test. It was expecting dat to be in /usr/local. After I did my
'make install', which is another location, the tests ran.
Greg
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Wednesday, March 7, 2018 3:04 PM
To: Ramos, Mr. Gregory, Contractor, Code 7501.1
Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
Looks like that works well! No more runtime errors.
Sounds like on our end we should test building with NetCDF 4.6.0 to
see if we can replicate the issues you ran into.
John
On Wed, Mar 7, 2018 at 3:43 PM, Ramos, Mr. Gregory, Contractor, Code
7501.1 via RT <met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
>
> Attached are the results using the exact versions of NetCDF "C" and
"C++"
> specified in the instructions.
>
> Greg
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Wednesday, March 7, 2018 12:58 PM
> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
>
> Greg,
>
> Sorry, I'm not able to find an obvious solution for you.
>
> In the last year or so, we began building MET within a docker
container.
> Basically a single ascii file defines all the steps used to build
MET,
> along with all the dependent libraries. Here's that Dockerfile:
>
> https://github.com/NCAR/container-dtc-met/blob/master/MET/Dockerfile
>
> Listed below is an excerpt where we download/compile NetCDF4. Looks
> like we're using netcdf-c-4.4.1.1 and netcdf-cxx4.4.3.0. Not sure
if
> that'll help or not.
>
> John
>
> #
> # NetCDF4 (c and c++)
> #
> RUN mkdir -p /met/external_libs \
> && cd /met/external_libs \
> && echo "Downloading netcdf-c-4.4.1.1 from ${NETCDF4C_URL}" \ &&
wget
> ${NETCDF4C_URL} \ && unzip v4.4.1.1.zip \ && cd netcdf-c-4.4.1.1 \
&&
> LOG_FILE=/met/external_libs/netcdf-c-4.4.1.1/configure.log \ && echo
> "Configuring netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ &&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-c-4.4.1.1/make_install.log \ && echo "Compiling
> netcdf-c-4.4.1.1 and writing log file ${LOG_FILE}" \ && make install
>
> ${LOG_FILE} \ && echo "Downloading from ${NETCDF4CXX_URL}" \ && cd
> /met/external_libs \ && wget ${NETCDF4CXX_URL} \ && tar -xzf
> v4.3.0.tar.gz \ && cd netcdf-cxx4-4.3.0 \ &&
> LOG_FILE=/met/external_libs/netcdf-cxx4-4.3.0/configure.log \ &&
echo
> "Configuring netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \
&&
> ./configure > ${LOG_FILE} \ && LOG_FILE=/met/external_libs/
> netcdf-cxx4-4.3.0/make_install.log \ && echo "Compiling
> netcdf-cxx4-4.3.0 and writing log file ${LOG_FILE}" \ && make
install
> > ${LOG_FILE}
>
> On Tue, Mar 6, 2018 at 9:17 PM, Ramos, Mr. Gregory, Contractor, Code
> 7501.1 via RT <met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> >
> > This is the output of nc-config --all
> >
> > [ramos at maury2 ncview-2.1.7]$
> > /software/depot/netcdf-3.6.2018/bin/nc-config
> > --all
> >
> > This netCDF 4.6.0 has been built with the following features:
> >
> > --cc -> gcc
> > --cflags -> -I/software/depot/netcdf-3.6.2018/include
> > -I/software/depot/hdf5-1.10.1-gcc/include
> > -I/software/depot/hdf-4.2.10- rhel6-aerosol/include
> > --libs -> -L/software/depot/netcdf-3.6.2018/lib -lnetcdf
> >
> > --has-c++ -> no
> > --cxx ->
> >
> > --has-c++4 -> yes
> > --cxx4 -> g++
> > --cxx4flags -> -I/software/depot/netcdf-cxx4-4.3.0/include
> > --cxx4libs -> -L/software/depot/netcdf-cxx4-4.3.0/lib
> > -lnetcdf_c++4 -lnetcdf
> >
> > --has-fortran-> no
> > --has-dap -> no
> > --has-dap4 -> no
> > --has-nc2 -> yes
> > --has-nc4 -> yes
> > --has-hdf5 -> yes
> > --has-hdf4 -> yes
> > --has-logging-> no
> > --has-pnetcdf-> no
> > --has-szlib -> no
> > --has-parallel -> no
> > --has-cdf5 -> no
> >
> > --prefix -> /software/depot/netcdf-3.6.2018
> > --includedir-> /software/depot/netcdf-3.6.2018/include
> > --libdir -> /software/depot/netcdf-3.6.2018/lib
> > --version -> netCDF 4.6.0
> >
> > [ramos at maury2 ncview-2.1.7]$
> >
> > -----Original Message-----
> > From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > Sent: Tuesday, March 6, 2018 3:09 PM
> > To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> >
> > John, ldd ascii2nc will show the NetCDF library.
> >
> > NetCDF4 can be built without NetCDF4 features. This can be another
> > case that NetCDF file can not be opened (created ) because NetCDF
> > was created with NetCDF4 format.
> >
> > Other case will be different configuration on building MET and
> > executing MET.
> >
> > Building MET with NetCDF3 puzzles me. It should be detected by:
> > Step 1. compile time with header files Step 2. link time with the
> > library files.
> >
> > Cheers,
> > Howard
> >
> >
> > On 3/6/2018 3:02 PM, John Halley Gotway via RT wrote:
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >
> > > Greg,
> > >
> > > No, the MET tools do not use ncdump. They use the C++ NetCDF
> > > interface for reading/writing NetCDF files.
> > >
> > > John
> > >
> > > On Tue, Mar 6, 2018 at 2:57 PM, Ramos, Mr. Gregory, Contractor,
> > > Code
> > > 7501.1 via RT <met_help at ucar.edu> wrote:
> > >
> > >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316 >
> > >>
> > >> Does ascii2nc use ncdump? Here is my NetCDF buld binary
directory:
> > >>
> > >> [ramos at maury2 ~]$ ls -larth /software/depot/netcdf-
3.6.2018/bin
> > >> total 576K
> > >> drwxr-xr-x 6 ramos ramos 4.0K Mar 6 19:57 ..
> > >> -rwxr-xr-x 1 ramos ramos 98K Mar 6 19:57 ncgen3 -rwxr-xr-x 1
> > >> ramos ramos 260K Mar 6 19:57 ncgen -rwxr-xr-x 1 ramos ramos
109K
> > >> Mar 6
> > >> 19:58 ncdump -rwxr-xr-x 1 ramos ramos 66K Mar 6 19:58 nccopy
> > >> -rwxr-xr-x 1 ramos ramos 6.7K Mar 6 19:58 nc-config drwxr-xr-x
2
> > >> ramos ramos 4.0K Mar 6 20:15 .
> > >> -rwxr-xr-x 1 ramos ramos 2.8K Mar 6 20:15 ncxx4-config
> > >> [ramos at maury2 ~]$ ls -larth
> > >> /software/depot/netcdf-3.6.2018/bin/ncdump
> > >> -rwxr-xr-x 1 ramos ramos 109K Mar 6 19:58
/software/depot/netcdf-3.6.
> > >> 2018/bin/ncdump
> > >> [ramos at maury2 ~]$ ldd /software/depot/netcdf-
3.6.2018/bin/ncdump
> > >> linux-vdso.so.1 => (0x00007ffd951ea000)
> > >> libnetcdf.so.13 =>
> > >> /software/depot/netcdf-3.6.2018/lib/libnetcdf.so.13
> > >> (0x00002b3eb98fc000)
> > >> libmfhdf.so.0 => /software/depot/hdf-4.2.10-
> > >> rhel6-aerosol/lib/libmfhdf.so.0 (0x00002b3eb9bf3000)
> > >> libdf.so.0 =>
> > >> /software/depot/hdf-4.2.10-rhel6-aerosol/lib/libdf.so.0
> > >> (0x00002b3eb9e1a000)
> > >> libjpeg.so.62 => /usr/lib64/libjpeg.so.62
> (0x00000032a6200000)
> > >> libhdf5_hl.so.100 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5_hl.so.100
> > >> (0x00002b3eba0e3000)
> > >> libhdf5.so.101 =>
> > >> /software/depot/hdf5-1.10.1-gcc/lib/libhdf5.so.101
> > >> (0x00002b3eba306000)
> > >> librt.so.1 => /lib64/librt.so.1 (0x000000329a200000)
> > >> libdl.so.2 => /lib64/libdl.so.2 (0x0000003299a00000)
> > >> libm.so.6 => /lib64/libm.so.6 (0x0000003299200000)
> > >> libz.so.1 => /lib64/libz.so.1 (0x0000003299e00000)
> > >> libc.so.6 => /lib64/libc.so.6 (0x0000003298e00000)
> > >> libpthread.so.0 => /lib64/libpthread.so.0
> (0x0000003299600000)
> > >> /lib64/ld-linux-x86-64.so.2 (0x0000003298a00000)
> > >> [ramos at maury2 ~]$
> > >>
> > >> Greg
> > >>
> > >> -----Original Message-----
> > >> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >> Sent: Tuesday, March 6, 2018 1:45 PM
> > >> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>
> > >> Greg,
> > >>
> > >> Ah, OK, I see that you're using NetCDF version 3.6. That's
> > >> likely the problem. As of MET version 6.0 we switched from
> > >> NetCDF version
> > >> 3 to NetCDF version 4:
> > >>
> > >> https://dtcenter.org/met/users/support/release_notes/
> > >> METv6.0_release_notes.php
> > >>
> > >> Please try compiling against NetCDF4 instead of 3:
> > >>
> > >> https://dtcenter.org/met/users/support/online_tutorial/
> > >> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>
> > >> I'm surprised MET compiled without error and that the problem
> > >> showed up at runtime.
> > >>
> > >> John
> > >>
> > >> On Tue, Mar 6, 2018 at 2:36 PM, Ramos, Mr. Gregory, Contractor,
> > >> Code
> > >> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>
> > >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>
> > >>> BTW, here are my build settings:
> > >>>
> > >>> [ramos at maury2 met-7.0]$ env |grep MET
> > >>> MET_GSL=/software/depot/gsl-1.16
> > >>> MET_ROOT=/software/depot/met-6.1_bugfix
> > >>> MET_NETCDF=/software/depot/netcdf-3.6.2018
> > >>> MET_NETCDFINC=
> > >>> MET_GRIB2C=/software/depot/grib_api-1.14.0-gcc
> > >>> MET_HDF5=/software/depot/hdf5-1.10.1-gcc/
> > >>> MET_BUFRLIB=/software/depot/bufrlib_v10-2-3
> > >>> [ramos at maury2 met-7.0]$
> > >>>
> > >>> Greg
> > >>>
> > >>> -----Original Message-----
> > >>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>> Sent: Tuesday, March 6, 2018 1:18 PM
> > >>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>
> > >>> Greg,
> > >>>
> > >>> Another thing to test would be the runtime linker. You set
the
> > >>> MET_NETCDF environment variable to tell MET where to find the
> > >>> header and library files for compilation. By default, MET is
> > >>> compiled using dynamic linking. I suppose its possible that
at
> > >>> runtime the MET executables are linking to some other version
of
> > >>> NetCDF. You could set the LD_LIBRARY_PATH environment
variable
> > >>> to define the search path for the linker. Let's try
> > >>> this:
> > >>>
> > >>> # Using c-shell syntax here. And MET_NETCDF is the top-level
> > >>> directory for your NetCDF compilations
> > >>>
> > >>> setenv LD_LIBRARY_PATH ${MET_NETCDF}/lib:${LD_LIBRARY_PATH}
> > >>> /software/depot/met-7.0/bin/ascii2nc
> > >>> /users/ramos/met/met-7.0/data/
> > >>> sample_obs/ascii/sample_ascii_obs.txt ~/sample_ascii.nc -v 2
> > >>>
> > >>> Any difference in results?
> > >>>
> > >>> Thanks,
> > >>> John
> > >>>
> > >>> On Tue, Mar 6, 2018 at 1:55 PM, Ramos, Mr. Gregory,
Contractor,
> > >>> Code
> > >>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>
> > >>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
>
> > >>>>
> > >>>> I rebuilt NetCDF C and NetCDF C++ and I am still getting the
> > >>>> same
> > >> error:
> > >>>>
> > >>>>
> > >>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>
> > >>>> *** Running ASCII2NC to reformat ASCII point observations
into
> > >>>> NetCDF
> > >>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>> -v 2
> > >>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>> observations off the masking grid.
> > >>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>> DEBUG 2: Processing observations for 5 headers.
> > >>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>> ERROR : FileHandler::_openNetcdf() -> can't open output
NetCDF
> > >>>> file "../out/ascii2nc/sample_ascii.nc" for writing ERROR :
> > >>>> make[1]: *** [ascii2nc] Error 1
> > >>>> make[1]: Leaving directory `/users/ramos/met/met-7.0/scripts'
> > >>>> make: *** [test] Error 2
> > >>>> [ramos at maury2 met-7.0]$
> > >>>>
> > >>>> -----Original Message-----
> > >>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>> Sent: Tuesday, March 6, 2018 10:14 AM
> > >>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>
> > >>>> Greg,
> > >>>>
> > >>>> To clarify, those backslashes '\' are line continuation
marks...
> > >>>> telling the shell that the command is continued on the next
line.
> > >>>> Including them on one line like that causes the odd behavior
of
> > >>>> creating an output file named "2". Removing them and
running:
> > >>>> /software/depot/met-7.0/bin/ascii2nc
> > >>>> /users/ramos/met/met-7.0/data/
> > >>>> sample_obs/ascii/sample_ascii_obs.txt
> > >>>> ~/sample_ascii.nc -v 2 Will still result in an error... but
> > >>>> the requested output file will be what
> > >>> you expect.
> > >>>> So I'd try recompiling NetCDF using the same family/version
> > >>>> number of the compilers you're using to compile MET. Sorry I
> > >>>> don't have an easier solution for you.
> > >>>>
> > >>>> Thanks,
> > >>>> John
> > >>>>
> > >>>>
> > >>>>
> > >>>> On Tue, Mar 6, 2018 at 11:06 AM, Ramos, Mr. Gregory,
> > >>>> Contractor, Code
> > >>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>
> > >>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>> >
> > >>>>>
> > >>>>> I had to modify your example slightly, but it didn't work:
> > >>>>>
> > >>>>>
> > >>>>> [ramos at maury2 met-7.0]$ /software/depot/met-7.0/bin/ascii2nc
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx t \ ~/ sample_ascii.nc \ -v 2 DEBUG 1: Config File:
> > >>>>> /software/depot/met-7.0/share/
> > >>>>> met/config/Ascii2NcConfig_default ERROR
> > >>>>> :
> > >>>>> ERROR : FileHandler::processFiles() -> can't open input
ASCII
> file "
> > >>>>> ~/ sample_ascii.nc" for reading ERROR :
> > >>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>> DEBUG 1: Creating NetCDF Observation file: 2 ERROR :
> > >>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>> NetCDF file
> > >>> "2"
> > >>>>> for writing
> > >>>>> ERROR :
> > >>>>> [ramos at maury2 met-7.0]$
> > >>>>>
> > >>>>> Greg
> > >>>>>
> > >>>>> -----Original Message-----
> > >>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>> Sent: Tuesday, March 6, 2018 9:58 AM
> > >>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>
> > >>>>> Greg,
> > >>>>>
> > >>>>> The number of executables built depends on the way the build
> > >>>>> was configured. Options which require "extra" dependencies
> > >>>>> are disabled by default. Here's an excerpt from the top-
level
> > >>>>> README
> > >> file:
> > >>>>> To configure MET to install all of the available tools in
> > >>>>> the
> > "bin"
> > >>>>> subdirectory of your current directory, you would use the
> > >>>>> following
> > >>>>> command:
> > >>>>>
> > >>>>> ./configure --prefix=`pwd` --enable-grib2 --enable-modis
\
> > >>>>> --enable-mode_graphics --enable-lidar2nc
> > >>>>>
> > >>>>> So it's fine that you don't have all possible executables
built.
> > >>>>> That's unrelated to the test script issue.
> > >>>>>
> > >>>>> My next guess would be an issue with the NetCDF library to
> > >>>>> which MET is linking. Let's replicate the error by trying
to
> > >>>>> write to some other output directory, like your home
directory:
> > >>>>>
> > >>>>> /users/ramos/met/met-7.0/bin/ascii2nc \
> > >>>>> /users/ramos/met/met-
7.0/data/sample_obs/ascii/sample_ascii_obs.
> > >>>>> tx
> > >>>>> t
> > >>>>> \ ~/ sample_ascii.nc \ -v 2
> > >>>>>
> > >>>>> I expect that this will fail with the same error message.
If
> > >>>>> it does, you could try recompiling NetCDF... and then
recompiling MET.
> > >>>>> Perhaps there's some incompatibility in how NetCDF and MET
are
> built?
> > >>>>>
> > >>>>> Here's some info about that:
> > >>>>> https://dtcenter.org/met/users/support/online_tutorial/
> > >>>>> METv6.1/tutorial.php?name=compilation&category=req_libs
> > >>>>>
> > >>>>> Thanks,
> > >>>>> John
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> On Tue, Mar 6, 2018 at 10:38 AM, Ramos, Mr. Gregory,
> > >>>>> Contractor, Code
> > >>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>
> > >>>>>> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>> >
> > >>>>>>
> > >>>>>> These are the binaries that were built. It did not build
all
> > >>>>>> of the binaries listed in the README:
> > >>>>>>
> > >>>>>> [ramos at maury2 met-7.0]$ ls -larth
> > >>>>>> /software/depot/met-7.0/bin/ total 208M drwxr-xr-x 4 ramos
ramos 4.0K Mar 6 16:44 ..
> > >>>>>> -rwxr-xr-x 1 ramos ramos 11M Mar 6 16:44 ensemble_stat
> > >>>>>> -rwxr-xr-x
> > >>>>>> 1 ramos ramos 11M Mar 6 16:44 grid_stat -rwxr-xr-x 1
ramos
> > >>>>>> ramos 13M Mar 6 16:44 mode -rwxr-xr-x 1 ramos ramos 6.1M
Mar
> > >>>>>> 6
> > >>>>>> 16:44 mode_analysis -rwxr-xr-x 1 ramos ramos 7.0M Mar 6
> > >>>>>> 16:44 pcp_combine -rwxr-xr-x 1 ramos ramos 11M Mar 6
16:44
> > >>>>>> point_stat -rwxr-xr-x 1 ramos ramos 9.7M Mar 6 16:44
> > >>>>>> stat_analysis -rwxr-xr-x 1 ramos ramos 12M Mar 6 16:44
> > >>>>>> wavelet_stat -rwxr-xr-x
> > >>>>>> 1 ramos ramos 9.8M Mar 6
> > >>>>>> 16:44 series_analysis -rwxr-xr-x 1 ramos ramos 7.3M Mar 6
> > >>>>>> 16:44 gen_vx_mask -rwxr-xr-x 1 ramos ramos 8.0M Mar 6
16:44
> > >>>>>> plot_data_plane -rwxr-xr-x 1 ramos ramos 7.3M Mar 6 16:44
> > >>>>>> plot_point_obs -rwxr-xr-x 1 ramos ramos 8.3M Mar 6 16:44
> > >>>>>> ascii2nc -rwxr-xr-x 1 ramos ramos 6.6M Mar 6 16:44
madis2nc
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.2M Mar 6 16:44 pb2nc -rwxr-xr-x
1
> > >>>>>> ramos ramos 8.4M Mar 6 16:44 wwmca_regrid -rwxr-xr-x 1
ramos
> > >>>>>> ramos 8.9M Mar 6
> > >>>>>> 16:44 wwmca_plot -rwxr-xr-x 1 ramos ramos 10M Mar 6 16:44
> > >>>>>> mtd -rwxr-xr-x
> > >>>>>> 1 ramos ramos 6.9M Mar 6 16:44 gsid2mpr -rwxr-xr-x 1 ramos
> > >>>>>> ramos 7.5M Mar 6 16:44 gsidens2orank -rwxr-xr-x 1 ramos
> > >>>>>> ramos 7.6M Mar
> > >>>>>> 6
> > >>>>>> 16:44 regrid_data_plane -rwxr-xr-x 1 ramos ramos 7.2M Mar
6
> > >>>>>> 16:44 shift_data_plane -rwxr-xr-x 1 ramos ramos 3.6M Mar 6
> > >>>>>> 16:44 tc_dland -rwxr-xr-x 1 ramos ramos 5.5M Mar 6 16:44
> > >>>>>> tc_pairs drwxrwxr-x 2 ramos
> > >>>> ramos 4.0K Mar 6 16:44 .
> > >>>>>> -rwxr-xr-x 1 ramos ramos 8.6M Mar 6 16:44 tc_stat
> > >>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>
> > >>>>>> -----Original Message-----
> > >>>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> > >>>>>> Sent: Tuesday, March 6, 2018 9:12 AM
> > >>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>
> > >>>>>> Hi Greg,
> > >>>>>>
> > >>>>>> This is John, I work with Howard. Well I'm glad to hear
that
> > >>>>>> the behavior is consistent across versions. If not, I
would
> > >>>>>> have been
> > >>>>> confused.
> > >>>>>> So the very first command that's being run by the test
script
> > >>>>>> is failing because it isn't able to create an output file.
> > >>>>>>
> > >>>>>> As Howard mentioned, the most likely cause for this type of
> > >>>>>> error is a permissions issue. So let's rule that out
first.
> > >>>>>> Please try running the following commands:
> > >>>>>> cd /users/ramos/met/met-7.0/scripts
> > >>>>>> touch ../out/ascii2nc/test_output_file
> > >>>>>>
> > >>>>>> The touch command just creates an empty file. If you get
an
> > >>>>>> error message about permissions, then we've found the
problem.
> > >>>>>> I tried running the exact same ascii2nc command but writing
> > >>>>>> my output to "/usr/bin"... where of course I don't have
> > >>>>>> permission to
> > >> write.
> > >>>>>> Here's
> > >>>>> the error message I got:
> > >>>>>> cd met-7.0/scripts
> > >>>>>> ../src/tools/other/ascii2nc/ascii2nc
> > >>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt
> > >>>>>> /usr/bin/sample_ascii_obs.txt -v 2
> > >>>>>>
> > >>>>>> ERROR :
> > >>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>> NetCDF file "/usr/bin/sample_ascii_obs.txt" for writing
ERROR :
> > >>>>>>
> > >>>>>> Do a long listing of that directory to see the permissions:
> > >>>>>> ls -la /users/ramos/met/met-7.0/out/ascii2nc
> > >>>>>>
> > >>>>>> What do you see?
> > >>>>>>
> > >>>>>> Thanks,
> > >>>>>> John
> > >>>>>>
> > >>>>>>
> > >>>>>> On Tue, Mar 6, 2018 at 9:49 AM, Ramos, Mr. Gregory,
> > >>>>>> Contractor, Code
> > >>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>
> > >>>>>>> <URL:
> > >>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>> This morning I downloaded the latest version of MET. The
> > >>>>>>> tests still don't
> > >>>>>>> run:
> > >>>>>>>
> > >>>>>>> [ramos at maury2 met-7.0]$ make test cd scripts; make
> > >>>>>>> make[1]: Entering directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>>
> > >>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> > >>>>>>> into NetCDF
> > >>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>> -v 2
> > >>>>>>> DEBUG 1: Config File: /software/depot/met-7.0/share/
> > >>>>>>> met/config/Ascii2NcConfig_default DEBUG 2: Rejected 0
> > >>>>>>> observations off the masking grid.
> > >>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >> ERROR :
> > >>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>> make[1]: Leaving directory `/users/ramos/met/met-
7.0/scripts'
> > >>>>>>> make: *** [test] Error 2
> > >>>>>>> [ramos at maury2 met-7.0]$
> > >>>>>>>
> > >>>>>>> -----Original Message-----
> > >>>>>>> From: Howard Soh via RT [mailto:met_help at ucar.edu]
> > >>>>>>> Sent: Tuesday, March 6, 2018 8:23 AM
> > >>>>>>> To: Ramos, Mr. Gregory, Contractor, Code 7501.1
> > >>>>>>> Subject: Re: [rt.rap.ucar.edu #84316] MET 6.1 bugfix
> > >>>>>>>
> > >>>>>>> Would you check if ../out/ascii2nc exists?
> > >>>>>>>
> > >>>>>>> The common cause are 1) when the output directory does not
> > >>>>>>> exist or
> > >>>>>>> 2) the user does not have the proper privileges (write
> > >> privilege).
> > >>>>>>> Would you check these?
> > >>>>>>>
> > >>>>>>> ls ../out/ascii2nc/
> > >>>>>>> touch ../out/ascii2nc/test.txt rm
../out/ascii2nc/test.txt
> > >>>>>>>
> > >>>>>>> Cheers,
> > >>>>>>> Howard
> > >>>>>>>
> > >>>>>>>
> > >>>>>>> On Tue, Mar 6, 2018 at 8:31 AM, Ramos, Mr. Gregory,
> > >>>>>>> Contractor, Code
> > >>>>>>> 7501.1 via RT <met_help at ucar.edu> wrote:
> > >>>>>>>
> > >>>>>>>> Tue Mar 06 08:31:27 2018: Request 84316 was acted upon.
> > >>>>>>>> Transaction: Ticket created by
Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Queue: met_help
> > >>>>>>>> Subject: MET 6.1 bugfix
> > >>>>>>>> Owner: Nobody
> > >>>>>>>> Requestors: Greg.Ramos.ctr at nrlmry.navy.mil
> > >>>>>>>> Status: new
> > >>>>>>>> Ticket <URL:
> > >>>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=84316
> > >>>>>>>>
> > >>>>>>>> I built the MET 6.1, but the tests won't run. My user is
> > >>>>>>>> getting the same error in trying to open the file for
> > >>>>>>>> writing using the
> > >>>>> tutorial:
> > >>>>>>>>
> > >>>>>>>> cd scripts; make
> > >>>>>>>> make[1]: Entering directory `/users/ramos/met/met-6.1_
> > >>>>> bugfix/scripts'
> > >>>>>>>> *** Running ASCII2NC to reformat ASCII point observations
> > >>>>>>>> into NetCDF
> > >>>>>>>> *** ../src/tools/other/ascii2nc/ascii2nc \
> > >>>>>>>> ../data/sample_obs/ascii/sample_ascii_obs.txt \
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc \
> > >>>>>>>> -v 2
> > >>>>>>>> DEBUG 1: Config File:
> > >>>>>>>> /software/depot/met-6.1_bugfix/share/met/config/
> > >>>>>>>> Ascii2NcConfig_default
> > >>>>>>>> DEBUG 2: Rejected 0 observations off the masking grid.
> > >>>>>>>> DEBUG 2: Rejected 0 observations outside the masking
polyline.
> > >>>>>>>> DEBUG 2: Rejected 0 observations not matched with station
ID's.
> > >>>>>>>> DEBUG 2: Processing observations for 5 headers.
> > >>>>>>>> DEBUG 1: Creating NetCDF Observation file:
> > >>>>>>>> ../out/ascii2nc/sample_ascii.nc ERROR :
> > >>>>>>>> ERROR : FileHandler::_openNetcdf() -> can't open output
> > >>>>>>>> NetCDF file "../out/ascii2nc/sample_ascii.nc" for writing
> > >>> ERROR :
> > >>>>>>>> make[1]: *** [ascii2nc] Error 1
> > >>>>>>>> make[1]: Leaving directory `/users/ramos/met/met-6.1_
> > >>>> bugfix/scripts'
> > >>>>>>>> make: *** [test] Error 2
> > >>>>>>>> [ramos at maury2 met-6.1_bugfix]$
> > >>>>>>>>
> > >>>>>>>> I am guessing this is a common error.
> > >>>>>>>>
> > >>>>>>>> Greg Ramos
> > >>>>>>>>
> > >>>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>>
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>
> > >>>
> > >>>
> > >>
> > >>
> > >>
> >
> >
> >
> >
> >
>
>
>
------------------------------------------------
More information about the Met_help
mailing list