[Met_help] [rt.rap.ucar.edu #55564] History for FW: PB2NC Problem (UNCLASSIFIED)
John Halley Gotway via RT
met_help at ucar.edu
Tue Jun 5 06:50:15 MDT 2012
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Classification: UNCLASSIFIED
Caveats: NONE
Do you see anything obvious below which would explain why MET PB2NC fails?
This repeats for every hour 0-24.
I'm running MET V3.1 and checked the run command and the config file and
don't see anything obvious.
***Running PB2NC on PrepBufr files***
***Files for use with WRF initialized at 06Z on 20120110*** hour 0 DEBUG 1:
Reading Default Config File:
/opt/MET/MET_3.1/data/config/PB2NCConfig_default
DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
DEBUG 1: Creating NetCDF File: ../ncobs/20120110/LAXd02_06_00_pb.nc
DEBUG 1: Processing PrepBufr File:
/opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
***********BUFR ARCHIVE LIBRARY ABORT**************
BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
***********BUFR ARCHIVE LIBRARY ABORT**************
Thanks.
R/
John
Classification: UNCLASSIFIED
Caveats: NONE
Classification: UNCLASSIFIED
Caveats: NONE
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Thu Mar 29 10:26:50 2012
John,
Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully understand your description of the problem. I see the error
message your getting, but does that happen every time you run
METv3.1 PB2NC, or does it show up intermittently?
Off the top of my head, one possibility is that when you built
METv3.1, you continued to link against a version of BUFRLIB that may
have been compiled using an older version of the compiler. That's
really just a shot in the dark, but you could try recompiling BUFRLIB
using your latest compiler version and then recompiling MET, being
sure to do a "make clean" first:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
If that doesn't fix the problem, I'll need to try to replicate the
problem here to debug it. Can you send me your "user_defs.mk" file,
your PB2NC Config file, and the PREPBUFR file you're using?
Also, I wanted to mention that we posted an updated set of bugfixes
for METv3.1 this morning. You can access it here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
Thanks,
John Halley Gotway
On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>
> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: FW: PB2NC Problem (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Do you see anything obvious below which would explain why MET PB2NC
fails?
> This repeats for every hour 0-24.
>
> I'm running MET V3.1 and checked the run command and the config file
and
> don't see anything obvious.
>
> ***Running PB2NC on PrepBufr files***
> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
> Reading Default Config File:
> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
> DEBUG 1: Processing PrepBufr File:
> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
>
> Thanks.
>
> R/
> John
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Thu Mar 29 12:09:50 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks for your initial assessment. I have forwarded this to our
sysadm who
compiled MET for his analysis and comment.
I could also run PB2NC using an earlier version of MET (V3.0.1 or even
V3.0)
and see if that changes anything.
I run PB2NC 24 times in a script (one per hour) and the error repeats
all 24
times. It's not intermittent.
Will keep you posted on results here.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, March 29, 2012 10:27 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully
understand your description of the problem. I see the error message
your
getting, but does that happen every time you run
METv3.1 PB2NC, or does it show up intermittently?
Off the top of my head, one possibility is that when you built
METv3.1, you
continued to link against a version of BUFRLIB that may have been
compiled
using an older version of the compiler. That's really just a shot in
the
dark, but you could try recompiling BUFRLIB using your latest compiler
version
and then recompiling MET, being sure to do a "make clean" first:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
If that doesn't fix the problem, I'll need to try to replicate the
problem
here to debug it. Can you send me your "user_defs.mk" file, your
PB2NC Config
file, and the PREPBUFR file you're using?
Also, I wanted to mention that we posted an updated set of bugfixes
for
METv3.1 this morning. You can access it here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
Thanks,
John Halley Gotway
On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>
> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: FW: PB2NC Problem (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Do you see anything obvious below which would explain why MET PB2NC
fails?
> This repeats for every hour 0-24.
>
> I'm running MET V3.1 and checked the run command and the config file
> and don't see anything obvious.
>
> ***Running PB2NC on PrepBufr files***
> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
> Reading Default Config File:
> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
> DEBUG 1: Processing PrepBufr File:
> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
>
> Thanks.
>
> R/
> John
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Apr 03 14:39:06 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
I have done some testing here and found that PB2NC runs without error
if I
drop back to METV3.0. The error is the same as I described below if I
use
METV3.0.1 and METV3.1. Does this shed any light on the problem? If
not, I'll
provide the file you requested below.
Thanks.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, March 29, 2012 10:27 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully
understand your description of the problem. I see the error message
your
getting, but does that happen every time you run
METv3.1 PB2NC, or does it show up intermittently?
Off the top of my head, one possibility is that when you built
METv3.1, you
continued to link against a version of BUFRLIB that may have been
compiled
using an older version of the compiler. That's really just a shot in
the
dark, but you could try recompiling BUFRLIB using your latest compiler
version
and then recompiling MET, being sure to do a "make clean" first:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
If that doesn't fix the problem, I'll need to try to replicate the
problem
here to debug it. Can you send me your "user_defs.mk" file, your
PB2NC Config
file, and the PREPBUFR file you're using?
Also, I wanted to mention that we posted an updated set of bugfixes
for
METv3.1 this morning. You can access it here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
Thanks,
John Halley Gotway
On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>
> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: FW: PB2NC Problem (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Do you see anything obvious below which would explain why MET PB2NC
fails?
> This repeats for every hour 0-24.
>
> I'm running MET V3.1 and checked the run command and the config file
> and don't see anything obvious.
>
> ***Running PB2NC on PrepBufr files***
> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
> Reading Default Config File:
> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
> DEBUG 1: Processing PrepBufr File:
> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
>
> Thanks.
>
> R/
> John
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Apr 03 15:00:44 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
I forgot to ask another question.
If I process my PrepBUFR data using PB2NC for METV3.0 what is the
difference
between the results I would get by using the METV3.1 version?
Thanks.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, March 29, 2012 10:27 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully
understand your description of the problem. I see the error message
your
getting, but does that happen every time you run
METv3.1 PB2NC, or does it show up intermittently?
Off the top of my head, one possibility is that when you built
METv3.1, you
continued to link against a version of BUFRLIB that may have been
compiled
using an older version of the compiler. That's really just a shot in
the
dark, but you could try recompiling BUFRLIB using your latest compiler
version
and then recompiling MET, being sure to do a "make clean" first:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
If that doesn't fix the problem, I'll need to try to replicate the
problem
here to debug it. Can you send me your "user_defs.mk" file, your
PB2NC Config
file, and the PREPBUFR file you're using?
Also, I wanted to mention that we posted an updated set of bugfixes
for
METv3.1 this morning. You can access it here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
Thanks,
John Halley Gotway
On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>
> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: FW: PB2NC Problem (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Do you see anything obvious below which would explain why MET PB2NC
fails?
> This repeats for every hour 0-24.
>
> I'm running MET V3.1 and checked the run command and the config file
> and don't see anything obvious.
>
> ***Running PB2NC on PrepBufr files***
> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
> Reading Default Config File:
> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
> DEBUG 1: Processing PrepBufr File:
> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
>
> Thanks.
>
> R/
> John
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Wed Apr 04 12:50:11 2012
John -
I've attached the files you mention below. Could you see if you can
replicate the error?
Thanks.
R/
John
________________________________________
From: John Halley Gotway via RT [met_help at ucar.edu]
Sent: Thursday, March 29, 2012 10:26 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully understand your description of the problem. I see the error
message your getting, but does that happen every time you run
METv3.1 PB2NC, or does it show up intermittently?
Off the top of my head, one possibility is that when you built
METv3.1, you continued to link against a version of BUFRLIB that may
have been compiled using an older version of the compiler. That's
really just a shot in the dark, but you could try recompiling BUFRLIB
using your latest compiler version and then recompiling MET, being
sure to do a "make clean" first:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
If that doesn't fix the problem, I'll need to try to replicate the
problem here to debug it. Can you send me your "user_defs.mk" file,
your PB2NC Config file, and the PREPBUFR file you're using?
Also, I wanted to mention that we posted an updated set of bugfixes
for METv3.1 this morning. You can access it here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
Thanks,
John Halley Gotway
On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>
> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: FW: PB2NC Problem (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Do you see anything obvious below which would explain why MET PB2NC
fails?
> This repeats for every hour 0-24.
>
> I'm running MET V3.1 and checked the run command and the config file
and
> don't see anything obvious.
>
> ***Running PB2NC on PrepBufr files***
> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
> Reading Default Config File:
> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
> DEBUG 1: Processing PrepBufr File:
> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>
> ***********BUFR ARCHIVE LIBRARY ABORT**************
>
> Thanks.
>
> R/
> John
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Fri Apr 06 13:24:28 2012
John,
Sorry for the delay in getting back to you. I've been testing things
trying to replicate the error but am unable to do so. It's a BUFRLIB
error we have not previously encountered. I tried to test
my theory that it may be due to inconsistent compiler versions, but I
do not have adequate compiler versions available to test it out.
Generally, when there are issues with BUFRLIB, my two questions are:
- Are the versions of the C and FORTRAN compiler used to compile
BUFRLIB consistent with the C++ and FORTRAN compilers used to compile
MET?
- Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
I don't think the -DBLOCK4 is the issue. Having it set wrong produces
a different error from BUFRLIB.
In your user_defs.mk file, I see that you're compiling MET using GNU
version 4.4.4. We are using GNU version 4.4.5, and are having no
problems. And when we used earlier versions of the GNU
compilers, it worked fine too. I do see the following in that file:
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1 was
compiled using GNU 4.4.4? If not, try using
/opt/MET/BUFRLIB/v10.0.1_gcc444.
Sorry I can't be of more help at this point. Have you been able to
make any progress?
Thanks,
John
On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> John -
>
> I've attached the files you mention below. Could you see if you can
replicate the error?
> Thanks.
> R/
> John
> ________________________________________
> From: John Halley Gotway via RT [met_help at ucar.edu]
> Sent: Thursday, March 29, 2012 10:26 AM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
fully understand your description of the problem. I see the error
message your getting, but does that happen every time you run
> METv3.1 PB2NC, or does it show up intermittently?
>
> Off the top of my head, one possibility is that when you built
METv3.1, you continued to link against a version of BUFRLIB that may
have been compiled using an older version of the compiler. That's
> really just a shot in the dark, but you could try recompiling
BUFRLIB using your latest compiler version and then recompiling MET,
being sure to do a "make clean" first:
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
>
> If that doesn't fix the problem, I'll need to try to replicate the
problem here to debug it. Can you send me your "user_defs.mk" file,
your PB2NC Config file, and the PREPBUFR file you're using?
>
> Also, I wanted to mention that we posted an updated set of bugfixes
for METv3.1 this morning. You can access it here:
>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
>
> Thanks,
> John Halley Gotway
>
> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>
>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>> Queue: met_help
>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>> Owner: Nobody
>> Requestors: john.w.raby2.civ at mail.mil
>> Status: new
>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> Do you see anything obvious below which would explain why MET PB2NC
fails?
>> This repeats for every hour 0-24.
>>
>> I'm running MET V3.1 and checked the run command and the config
file and
>> don't see anything obvious.
>>
>> ***Running PB2NC on PrepBufr files***
>> ***Files for use with WRF initialized at 06Z on 20120110*** hour 0
DEBUG 1:
>> Reading Default Config File:
>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>> DEBUG 1: Creating NetCDF File:
../ncobs/20120110/LAXd02_06_00_pb.nc
>> DEBUG 1: Processing PrepBufr File:
>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>>
>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>>
>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>
>> Thanks.
>>
>> R/
>> John
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Fri Apr 06 13:38:13 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks for the analysis/discussion on possible things to look at. We
are
taking the tact that we may need to recompile METV3.1 using a
different
compiler, but I don't think this has been done yet. Will let you know
the
results.
I think I mentioned to you in another email that we were able to run
PB2NC
without error if we use METV3.0.
Will keep you posted.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Friday, April 06, 2012 1:24 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Sorry for the delay in getting back to you. I've been testing things
trying
to replicate the error but am unable to do so. It's a BUFRLIB error
we have
not previously encountered. I tried to test my theory that it may be
due to
inconsistent compiler versions, but I do not have adequate compiler
versions
available to test it out.
Generally, when there are issues with BUFRLIB, my two questions are:
- Are the versions of the C and FORTRAN compiler used to compile
BUFRLIB
consistent with the C++ and FORTRAN compilers used to compile MET?
- Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
I don't think the -DBLOCK4 is the issue. Having it set wrong produces
a
different error from BUFRLIB.
In your user_defs.mk file, I see that you're compiling MET using GNU
version
4.4.4. We are using GNU version 4.4.5, and are having no problems.
And when
we used earlier versions of the GNU compilers, it worked fine too. I
do see
the following in that file:
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1 was
compiled
using GNU 4.4.4? If not, try using /opt/MET/BUFRLIB/v10.0.1_gcc444.
Sorry I can't be of more help at this point. Have you been able to
make any
progress?
Thanks,
John
On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> John -
>
> I've attached the files you mention below. Could you see if you can
> replicate the error?
> Thanks.
> R/
> John
> ________________________________________
> From: John Halley Gotway via RT [met_help at ucar.edu]
> Sent: Thursday, March 29, 2012 10:26 AM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Hmmm, that's a new error message from PB2NC. Unfortunately, I don't
> fully understand your description of the problem. I see the error
> message your getting, but does that happen every time you run
> METv3.1 PB2NC, or does it show up intermittently?
>
> Off the top of my head, one possibility is that when you built
> METv3.1, you continued to link against a version of BUFRLIB that may
have
> been compiled using an older version of the compiler. That's really
just a
> shot in the dark, but you could try recompiling BUFRLIB using your
latest
> compiler version and then recompiling MET, being sure to do a "make
clean"
> first:
>
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/comp
> ilation/req_libs.php
>
> If that doesn't fix the problem, I'll need to try to replicate the
problem
> here to debug it. Can you send me your "user_defs.mk" file, your
PB2NC
> Config file, and the PREPBUFR file you're using?
>
> Also, I wanted to mention that we posted an updated set of bugfixes
for
> METv3.1 this morning. You can access it here:
>
>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.p
> hp
>
> Thanks,
> John Halley Gotway
>
> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>
>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>> Queue: met_help
>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>> Owner: Nobody
>> Requestors: john.w.raby2.civ at mail.mil
>> Status: new
>> Ticket<URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> Do you see anything obvious below which would explain why MET PB2NC
fails?
>> This repeats for every hour 0-24.
>>
>> I'm running MET V3.1 and checked the run command and the config
file
>> and don't see anything obvious.
>>
>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>> Reading Default Config File:
>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>> DEBUG 1: Creating NetCDF File:
>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>> DEBUG 1: Processing PrepBufr File:
>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>>
>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>>
>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>
>> Thanks.
>>
>> R/
>> John
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Fri Apr 06 13:44:47 2012
John,
Yes, I saw that about METv3.0. If I were in your position, I would
start from scratch with METv3.1. Just compile NetCDF, GSL, and
BUFRLIB using the current GNU compiler version (4.4.4 on your
system), and then compile MET using it as well. My theory is that one
of the packages is out of sync somewhere.
That whole recompiling process could be as quick as 15 minutes or so,
following the suggestions on the MET online tutorial:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
Thanks,
John
On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Thanks for the analysis/discussion on possible things to look at. We
are
> taking the tact that we may need to recompile METV3.1 using a
different
> compiler, but I don't think this has been done yet. Will let you
know the
> results.
>
> I think I mentioned to you in another email that we were able to run
PB2NC
> without error if we use METV3.0.
>
> Will keep you posted.
>
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Friday, April 06, 2012 1:24 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Sorry for the delay in getting back to you. I've been testing
things trying
> to replicate the error but am unable to do so. It's a BUFRLIB error
we have
> not previously encountered. I tried to test my theory that it may
be due to
> inconsistent compiler versions, but I do not have adequate compiler
versions
> available to test it out.
>
> Generally, when there are issues with BUFRLIB, my two questions are:
> - Are the versions of the C and FORTRAN compiler used to compile
BUFRLIB
> consistent with the C++ and FORTRAN compilers used to compile MET?
> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>
> I don't think the -DBLOCK4 is the issue. Having it set wrong
produces a
> different error from BUFRLIB.
>
> In your user_defs.mk file, I see that you're compiling MET using GNU
version
> 4.4.4. We are using GNU version 4.4.5, and are having no problems.
And when
> we used earlier versions of the GNU compilers, it worked fine too.
I do see
> the following in that file:
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>
> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was compiled
> using GNU 4.4.4? If not, try using /opt/MET/BUFRLIB/v10.0.1_gcc444.
>
> Sorry I can't be of more help at this point. Have you been able to
make any
> progress?
>
> Thanks,
> John
>
>
> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> John -
>>
>> I've attached the files you mention below. Could you see if you can
>> replicate the error?
>> Thanks.
>> R/
>> John
>> ________________________________________
>> From: John Halley Gotway via RT [met_help at ucar.edu]
>> Sent: Thursday, March 29, 2012 10:26 AM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>>
>> John,
>>
>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>> fully understand your description of the problem. I see the error
>> message your getting, but does that happen every time you run
>> METv3.1 PB2NC, or does it show up intermittently?
>>
>> Off the top of my head, one possibility is that when you built
>> METv3.1, you continued to link against a version of BUFRLIB that
may have
>> been compiled using an older version of the compiler. That's
really just a
>> shot in the dark, but you could try recompiling BUFRLIB using your
latest
>> compiler version and then recompiling MET, being sure to do a "make
clean"
>> first:
>>
>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/comp
>> ilation/req_libs.php
>>
>> If that doesn't fix the problem, I'll need to try to replicate the
problem
>> here to debug it. Can you send me your "user_defs.mk" file, your
PB2NC
>> Config file, and the PREPBUFR file you're using?
>>
>> Also, I wanted to mention that we posted an updated set of bugfixes
for
>> METv3.1 this morning. You can access it here:
>>
>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.p
>> hp
>>
>> Thanks,
>> John Halley Gotway
>>
>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>
>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket<URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> Do you see anything obvious below which would explain why MET
PB2NC fails?
>>> This repeats for every hour 0-24.
>>>
>>> I'm running MET V3.1 and checked the run command and the config
file
>>> and don't see anything obvious.
>>>
>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>> Reading Default Config File:
>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>> DEBUG 1: Creating NetCDF File:
>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>> DEBUG 1: Processing PrepBufr File:
>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Fri Apr 06 13:51:21 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks for your recommendation. I forwarded your email to our sysadm.
Have a great Easter weekend!
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Friday, April 06, 2012 1:45 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Yes, I saw that about METv3.0. If I were in your position, I would
start from
scratch with METv3.1. Just compile NetCDF, GSL, and BUFRLIB using the
current
GNU compiler version (4.4.4 on your system), and then compile MET
using it as
well. My theory is that one of the packages is out of sync somewhere.
That whole recompiling process could be as quick as 15 minutes or so,
following the suggestions on the MET online tutorial:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
Thanks,
John
On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Thanks for the analysis/discussion on possible things to look at. We
> are taking the tact that we may need to recompile METV3.1 using a
> different compiler, but I don't think this has been done yet. Will
let
> you know the results.
>
> I think I mentioned to you in another email that we were able to run
> PB2NC without error if we use METV3.0.
>
> Will keep you posted.
>
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Friday, April 06, 2012 1:24 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Sorry for the delay in getting back to you. I've been testing
things
> trying to replicate the error but am unable to do so. It's a
BUFRLIB
> error we have not previously encountered. I tried to test my theory
> that it may be due to inconsistent compiler versions, but I do not
> have adequate compiler versions available to test it out.
>
> Generally, when there are issues with BUFRLIB, my two questions are:
> - Are the versions of the C and FORTRAN compiler used to compile
> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>
> I don't think the -DBLOCK4 is the issue. Having it set wrong
produces
> a different error from BUFRLIB.
>
> In your user_defs.mk file, I see that you're compiling MET using GNU
> version 4.4.4. We are using GNU version 4.4.5, and are having no
> problems. And when we used earlier versions of the GNU compilers,
it
> worked fine too. I do see the following in that file:
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>
> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was
> compiled using GNU 4.4.4? If not, try using
> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>
> Sorry I can't be of more help at this point. Have you been able to
> make any progress?
>
> Thanks,
> John
>
>
> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> John -
>>
>> I've attached the files you mention below. Could you see if you can
>> replicate the error?
>> Thanks.
>> R/
>> John
>> ________________________________________
>> From: John Halley Gotway via RT [met_help at ucar.edu]
>> Sent: Thursday, March 29, 2012 10:26 AM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>> fully understand your description of the problem. I see the error
>> message your getting, but does that happen every time you run
>> METv3.1 PB2NC, or does it show up intermittently?
>>
>> Off the top of my head, one possibility is that when you built
>> METv3.1, you continued to link against a version of BUFRLIB that
may
>> have been compiled using an older version of the compiler. That's
>> really just a shot in the dark, but you could try recompiling
BUFRLIB
>> using your latest compiler version and then recompiling MET, being
sure to
>> do a "make clean"
>> first:
>>
>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/com
>> p
>> ilation/req_libs.php
>>
>> If that doesn't fix the problem, I'll need to try to replicate the
>> problem here to debug it. Can you send me your "user_defs.mk"
file,
>> your PB2NC Config file, and the PREPBUFR file you're using?
>>
>> Also, I wanted to mention that we posted an updated set of bugfixes
>> for
>> METv3.1 this morning. You can access it here:
>>
>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>> p
>> hp
>>
>> Thanks,
>> John Halley Gotway
>>
>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>
>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket<URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> Do you see anything obvious below which would explain why MET
PB2NC fails?
>>> This repeats for every hour 0-24.
>>>
>>> I'm running MET V3.1 and checked the run command and the config
file
>>> and don't see anything obvious.
>>>
>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>> Reading Default Config File:
>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>> DEBUG 1: Creating NetCDF File:
>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>> DEBUG 1: Processing PrepBufr File:
>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Apr 10 07:58:00 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Do you know if there is a significant difference between the version
of PB2NC
in METV3.0 and that of MET V3.1? My desire would be to perform some
processing
to get ahead on some PrepBUFR data I need, rather than wait to resolve
this
issue with our METV3.1. If there is a significant difference, then I
would
hold off.
Thanks.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Friday, April 06, 2012 1:45 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
Yes, I saw that about METv3.0. If I were in your position, I would
start from
scratch with METv3.1. Just compile NetCDF, GSL, and BUFRLIB using the
current
GNU compiler version (4.4.4 on your system), and then compile MET
using it as
well. My theory is that one of the packages is out of sync somewhere.
That whole recompiling process could be as quick as 15 minutes or so,
following the suggestions on the MET online tutorial:
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
Thanks,
John
On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Thanks for the analysis/discussion on possible things to look at. We
> are taking the tact that we may need to recompile METV3.1 using a
> different compiler, but I don't think this has been done yet. Will
let
> you know the results.
>
> I think I mentioned to you in another email that we were able to run
> PB2NC without error if we use METV3.0.
>
> Will keep you posted.
>
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Friday, April 06, 2012 1:24 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Sorry for the delay in getting back to you. I've been testing
things
> trying to replicate the error but am unable to do so. It's a
BUFRLIB
> error we have not previously encountered. I tried to test my theory
> that it may be due to inconsistent compiler versions, but I do not
> have adequate compiler versions available to test it out.
>
> Generally, when there are issues with BUFRLIB, my two questions are:
> - Are the versions of the C and FORTRAN compiler used to compile
> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>
> I don't think the -DBLOCK4 is the issue. Having it set wrong
produces
> a different error from BUFRLIB.
>
> In your user_defs.mk file, I see that you're compiling MET using GNU
> version 4.4.4. We are using GNU version 4.4.5, and are having no
> problems. And when we used earlier versions of the GNU compilers,
it
> worked fine too. I do see the following in that file:
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>
> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was
> compiled using GNU 4.4.4? If not, try using
> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>
> Sorry I can't be of more help at this point. Have you been able to
> make any progress?
>
> Thanks,
> John
>
>
> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> John -
>>
>> I've attached the files you mention below. Could you see if you can
>> replicate the error?
>> Thanks.
>> R/
>> John
>> ________________________________________
>> From: John Halley Gotway via RT [met_help at ucar.edu]
>> Sent: Thursday, March 29, 2012 10:26 AM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>> fully understand your description of the problem. I see the error
>> message your getting, but does that happen every time you run
>> METv3.1 PB2NC, or does it show up intermittently?
>>
>> Off the top of my head, one possibility is that when you built
>> METv3.1, you continued to link against a version of BUFRLIB that
may
>> have been compiled using an older version of the compiler. That's
>> really just a shot in the dark, but you could try recompiling
BUFRLIB
>> using your latest compiler version and then recompiling MET, being
sure to
>> do a "make clean"
>> first:
>>
>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/com
>> p
>> ilation/req_libs.php
>>
>> If that doesn't fix the problem, I'll need to try to replicate the
>> problem here to debug it. Can you send me your "user_defs.mk"
file,
>> your PB2NC Config file, and the PREPBUFR file you're using?
>>
>> Also, I wanted to mention that we posted an updated set of bugfixes
>> for
>> METv3.1 this morning. You can access it here:
>>
>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>> p
>> hp
>>
>> Thanks,
>> John Halley Gotway
>>
>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>
>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket<URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> Do you see anything obvious below which would explain why MET
PB2NC fails?
>>> This repeats for every hour 0-24.
>>>
>>> I'm running MET V3.1 and checked the run command and the config
file
>>> and don't see anything obvious.
>>>
>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>> Reading Default Config File:
>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>> DEBUG 1: Creating NetCDF File:
>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>> DEBUG 1: Processing PrepBufr File:
>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>> DEBUG 1: Blocking PrepBufr file to: /tmp/tmp_pb2nc_blk_31543_0
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC .REHOVI
>>>
>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Tue Apr 10 14:11:18 2012
John,
I wasn't sure on this, so I did the following - I used ncdump to dump
the contents of the output of the PB2NC test script to a text file for
both METv3.0.1 and METv3.1. Then I looked at their
differences and found the only differences in the MET_version
attribute and data of generation. All of the data values were
identical.
Based on that, I'd say that using PB2NC from METv3.0.1 would likely be
fine.
Thanks,
John
On 04/10/2012 07:58 AM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Do you know if there is a significant difference between the version
of PB2NC
> in METV3.0 and that of MET V3.1? My desire would be to perform some
processing
> to get ahead on some PrepBUFR data I need, rather than wait to
resolve this
> issue with our METV3.1. If there is a significant difference, then I
would
> hold off.
> Thanks.
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Friday, April 06, 2012 1:45 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Yes, I saw that about METv3.0. If I were in your position, I would
start from
> scratch with METv3.1. Just compile NetCDF, GSL, and BUFRLIB using
the current
> GNU compiler version (4.4.4 on your system), and then compile MET
using it as
> well. My theory is that one of the packages is out of sync
somewhere.
>
> That whole recompiling process could be as quick as 15 minutes or
so,
> following the suggestions on the MET online tutorial:
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/compilation/req_libs.php
>
> Thanks,
> John
>
> On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> John -
>>
>> Thanks for the analysis/discussion on possible things to look at.
We
>> are taking the tact that we may need to recompile METV3.1 using a
>> different compiler, but I don't think this has been done yet. Will
let
>> you know the results.
>>
>> I think I mentioned to you in another email that we were able to
run
>> PB2NC without error if we use METV3.0.
>>
>> Will keep you posted.
>>
>> R/
>> John
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Friday, April 06, 2012 1:24 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>>
>> John,
>>
>> Sorry for the delay in getting back to you. I've been testing
things
>> trying to replicate the error but am unable to do so. It's a
BUFRLIB
>> error we have not previously encountered. I tried to test my
theory
>> that it may be due to inconsistent compiler versions, but I do not
>> have adequate compiler versions available to test it out.
>>
>> Generally, when there are issues with BUFRLIB, my two questions
are:
>> - Are the versions of the C and FORTRAN compiler used to compile
>> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
>> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>>
>> I don't think the -DBLOCK4 is the issue. Having it set wrong
produces
>> a different error from BUFRLIB.
>>
>> In your user_defs.mk file, I see that you're compiling MET using
GNU
>> version 4.4.4. We are using GNU version 4.4.5, and are having no
>> problems. And when we used earlier versions of the GNU compilers,
it
>> worked fine too. I do see the following in that file:
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>>
>> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was
>> compiled using GNU 4.4.4? If not, try using
>> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>>
>> Sorry I can't be of more help at this point. Have you been able to
>> make any progress?
>>
>> Thanks,
>> John
>>
>>
>> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>> John -
>>>
>>> I've attached the files you mention below. Could you see if you
can
>>> replicate the error?
>>> Thanks.
>>> R/
>>> John
>>> ________________________________________
>>> From: John Halley Gotway via RT [met_help at ucar.edu]
>>> Sent: Thursday, March 29, 2012 10:26 AM
>>> To: Raby, John W CIV (US)
>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>> (UNCLASSIFIED)
>>>
>>> John,
>>>
>>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>>> fully understand your description of the problem. I see the error
>>> message your getting, but does that happen every time you run
>>> METv3.1 PB2NC, or does it show up intermittently?
>>>
>>> Off the top of my head, one possibility is that when you built
>>> METv3.1, you continued to link against a version of BUFRLIB that
may
>>> have been compiled using an older version of the compiler. That's
>>> really just a shot in the dark, but you could try recompiling
BUFRLIB
>>> using your latest compiler version and then recompiling MET, being
sure to
>>> do a "make clean"
>>> first:
>>>
>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/com
>>> p
>>> ilation/req_libs.php
>>>
>>> If that doesn't fix the problem, I'll need to try to replicate the
>>> problem here to debug it. Can you send me your "user_defs.mk"
file,
>>> your PB2NC Config file, and the PREPBUFR file you're using?
>>>
>>> Also, I wanted to mention that we posted an updated set of
bugfixes
>>> for
>>> METv3.1 this morning. You can access it here:
>>>
>>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>>> p
>>> hp
>>>
>>> Thanks,
>>> John Halley Gotway
>>>
>>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>>
>>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>>> Queue: met_help
>>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>>> Owner: Nobody
>>>> Requestors: john.w.raby2.civ at mail.mil
>>>> Status: new
>>>> Ticket<URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>> Do you see anything obvious below which would explain why MET
PB2NC fails?
>>>> This repeats for every hour 0-24.
>>>>
>>>> I'm running MET V3.1 and checked the run command and the config
file
>>>> and don't see anything obvious.
>>>>
>>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>>> Reading Default Config File:
>>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>>> DEBUG 1: Creating NetCDF File:
>>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>>> DEBUG 1: Processing PrepBufr File:
>>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>>> DEBUG 1: Blocking PrepBufr file to:
/tmp/tmp_pb2nc_blk_31543_0
>>>>
>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC
.REHOVI
>>>>
>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>
>>>> Thanks.
>>>>
>>>> R/
>>>> John
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>>
>>>>
>>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Apr 10 15:08:10 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks, but I can't get the V3.0.1 version to run either without that
error.
I can run it fine with V3.0.
New subj: How do you run an R Script (plot_cnt.R)? I have the R Gui up
and
can't seem to get it to load that script. See following dialog:
load("C:\\Users\\john.raby\\Documents\\Userdata_24FEB 11\\Old PC -
B\\Evaluation Project\\R Scripts for Plotting\\plot_cnt.R.txt")
Error: bad restore file magic number (file may be corrupted) -- no
data loaded
In addition: Warning message:
file 'plot_cnt.R.txt' has magic number '#####'
Use of save versions prior to 2 is deprecated
>
Thanks.
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, April 10, 2012 2:11 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
I wasn't sure on this, so I did the following - I used ncdump to dump
the
contents of the output of the PB2NC test script to a text file for
both
METv3.0.1 and METv3.1. Then I looked at their differences and found
the only
differences in the MET_version attribute and data of generation. All
of the
data values were identical.
Based on that, I'd say that using PB2NC from METv3.0.1 would likely be
fine.
Thanks,
John
On 04/10/2012 07:58 AM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Do you know if there is a significant difference between the version
> of PB2NC in METV3.0 and that of MET V3.1? My desire would be to
> perform some processing to get ahead on some PrepBUFR data I need,
> rather than wait to resolve this issue with our METV3.1. If there is
a
> significant difference, then I would hold off.
> Thanks.
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Friday, April 06, 2012 1:45 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> Yes, I saw that about METv3.0. If I were in your position, I would
> start from scratch with METv3.1. Just compile NetCDF, GSL, and
> BUFRLIB using the current GNU compiler version (4.4.4 on your
system),
> and then compile MET using it as well. My theory is that one of the
> packages is out of sync somewhere.
>
> That whole recompiling process could be as quick as 15 minutes or
so,
> following the suggestions on the MET online tutorial:
>
>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/comp
> ilation/req_libs.php
>
> Thanks,
> John
>
> On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> John -
>>
>> Thanks for the analysis/discussion on possible things to look at.
We
>> are taking the tact that we may need to recompile METV3.1 using a
>> different compiler, but I don't think this has been done yet. Will
>> let you know the results.
>>
>> I think I mentioned to you in another email that we were able to
run
>> PB2NC without error if we use METV3.0.
>>
>> Will keep you posted.
>>
>> R/
>> John
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Friday, April 06, 2012 1:24 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Sorry for the delay in getting back to you. I've been testing
things
>> trying to replicate the error but am unable to do so. It's a
BUFRLIB
>> error we have not previously encountered. I tried to test my
theory
>> that it may be due to inconsistent compiler versions, but I do not
>> have adequate compiler versions available to test it out.
>>
>> Generally, when there are issues with BUFRLIB, my two questions
are:
>> - Are the versions of the C and FORTRAN compiler used to compile
>> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
>> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>>
>> I don't think the -DBLOCK4 is the issue. Having it set wrong
>> produces a different error from BUFRLIB.
>>
>> In your user_defs.mk file, I see that you're compiling MET using
GNU
>> version 4.4.4. We are using GNU version 4.4.5, and are having no
>> problems. And when we used earlier versions of the GNU compilers,
it
>> worked fine too. I do see the following in that file:
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>>
>> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was
>> compiled using GNU 4.4.4? If not, try using
>> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>>
>> Sorry I can't be of more help at this point. Have you been able to
>> make any progress?
>>
>> Thanks,
>> John
>>
>>
>> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>> John -
>>>
>>> I've attached the files you mention below. Could you see if you
can
>>> replicate the error?
>>> Thanks.
>>> R/
>>> John
>>> ________________________________________
>>> From: John Halley Gotway via RT [met_help at ucar.edu]
>>> Sent: Thursday, March 29, 2012 10:26 AM
>>> To: Raby, John W CIV (US)
>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>> (UNCLASSIFIED)
>>>
>>> John,
>>>
>>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>>> fully understand your description of the problem. I see the error
>>> message your getting, but does that happen every time you run
>>> METv3.1 PB2NC, or does it show up intermittently?
>>>
>>> Off the top of my head, one possibility is that when you built
>>> METv3.1, you continued to link against a version of BUFRLIB that
may
>>> have been compiled using an older version of the compiler. That's
>>> really just a shot in the dark, but you could try recompiling
>>> BUFRLIB using your latest compiler version and then recompiling
MET,
>>> being sure to do a "make clean"
>>> first:
>>>
>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/co
>>> m
>>> p
>>> ilation/req_libs.php
>>>
>>> If that doesn't fix the problem, I'll need to try to replicate the
>>> problem here to debug it. Can you send me your "user_defs.mk"
file,
>>> your PB2NC Config file, and the PREPBUFR file you're using?
>>>
>>> Also, I wanted to mention that we posted an updated set of
bugfixes
>>> for
>>> METv3.1 this morning. You can access it here:
>>>
>>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>>> p
>>> hp
>>>
>>> Thanks,
>>> John Halley Gotway
>>>
>>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>>
>>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>>> Queue: met_help
>>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>>> Owner: Nobody
>>>> Requestors: john.w.raby2.civ at mail.mil
>>>> Status: new
>>>> Ticket<URL:
>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>> Do you see anything obvious below which would explain why MET
PB2NC
>>>> fails?
>>>> This repeats for every hour 0-24.
>>>>
>>>> I'm running MET V3.1 and checked the run command and the config
>>>> file and don't see anything obvious.
>>>>
>>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>>> Reading Default Config File:
>>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>>> DEBUG 1: Creating NetCDF File:
>>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>>> DEBUG 1: Processing PrepBufr File:
>>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>>> DEBUG 1: Blocking PrepBufr file to:
/tmp/tmp_pb2nc_blk_31543_0
>>>>
>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC
.REHOVI
>>>>
>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>
>>>> Thanks.
>>>>
>>>> R/
>>>> John
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>>
>>>>
>>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Tue Apr 10 15:15:17 2012
John,
I don't typically use an R GUI, so I'm not sure. We always run them
on a command line, using the Rscript command:
Rscript plot_cnt.R (whatever arguments you need to supply)
John
On 04/10/2012 03:08 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Thanks, but I can't get the V3.0.1 version to run either without
that error.
>
> I can run it fine with V3.0.
>
> New subj: How do you run an R Script (plot_cnt.R)? I have the R Gui
up and
> can't seem to get it to load that script. See following dialog:
>
> load("C:\\Users\\john.raby\\Documents\\Userdata_24FEB 11\\Old PC -
> B\\Evaluation Project\\R Scripts for Plotting\\plot_cnt.R.txt")
> Error: bad restore file magic number (file may be corrupted) -- no
data loaded
> In addition: Warning message:
> file 'plot_cnt.R.txt' has magic number '#####'
> Use of save versions prior to 2 is deprecated
>>
>
> Thanks.
>
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, April 10, 2012 2:11 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> I wasn't sure on this, so I did the following - I used ncdump to
dump the
> contents of the output of the PB2NC test script to a text file for
both
> METv3.0.1 and METv3.1. Then I looked at their differences and found
the only
> differences in the MET_version attribute and data of generation.
All of the
> data values were identical.
>
> Based on that, I'd say that using PB2NC from METv3.0.1 would likely
be fine.
>
> Thanks,
> John
>
> On 04/10/2012 07:58 AM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> John -
>>
>> Do you know if there is a significant difference between the
version
>> of PB2NC in METV3.0 and that of MET V3.1? My desire would be to
>> perform some processing to get ahead on some PrepBUFR data I need,
>> rather than wait to resolve this issue with our METV3.1. If there
is a
>> significant difference, then I would hold off.
>> Thanks.
>> R/
>> John
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Friday, April 06, 2012 1:45 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>>
>> John,
>>
>> Yes, I saw that about METv3.0. If I were in your position, I would
>> start from scratch with METv3.1. Just compile NetCDF, GSL, and
>> BUFRLIB using the current GNU compiler version (4.4.4 on your
system),
>> and then compile MET using it as well. My theory is that one of
the
>> packages is out of sync somewhere.
>>
>> That whole recompiling process could be as quick as 15 minutes or
so,
>> following the suggestions on the MET online tutorial:
>>
>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/comp
>> ilation/req_libs.php
>>
>> Thanks,
>> John
>>
>> On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> John -
>>>
>>> Thanks for the analysis/discussion on possible things to look at.
We
>>> are taking the tact that we may need to recompile METV3.1 using a
>>> different compiler, but I don't think this has been done yet. Will
>>> let you know the results.
>>>
>>> I think I mentioned to you in another email that we were able to
run
>>> PB2NC without error if we use METV3.0.
>>>
>>> Will keep you posted.
>>>
>>> R/
>>> John
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Friday, April 06, 2012 1:24 PM
>>> To: Raby, John W CIV (US)
>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>> (UNCLASSIFIED)
>>>
>>> John,
>>>
>>> Sorry for the delay in getting back to you. I've been testing
things
>>> trying to replicate the error but am unable to do so. It's a
BUFRLIB
>>> error we have not previously encountered. I tried to test my
theory
>>> that it may be due to inconsistent compiler versions, but I do not
>>> have adequate compiler versions available to test it out.
>>>
>>> Generally, when there are issues with BUFRLIB, my two questions
are:
>>> - Are the versions of the C and FORTRAN compiler used to compile
>>> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
>>> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>>>
>>> I don't think the -DBLOCK4 is the issue. Having it set wrong
>>> produces a different error from BUFRLIB.
>>>
>>> In your user_defs.mk file, I see that you're compiling MET using
GNU
>>> version 4.4.4. We are using GNU version 4.4.5, and are having no
>>> problems. And when we used earlier versions of the GNU compilers,
it
>>> worked fine too. I do see the following in that file:
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>>>
>>> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
was
>>> compiled using GNU 4.4.4? If not, try using
>>> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>>>
>>> Sorry I can't be of more help at this point. Have you been able
to
>>> make any progress?
>>>
>>> Thanks,
>>> John
>>>
>>>
>>> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>
>>>> John -
>>>>
>>>> I've attached the files you mention below. Could you see if you
can
>>>> replicate the error?
>>>> Thanks.
>>>> R/
>>>> John
>>>> ________________________________________
>>>> From: John Halley Gotway via RT [met_help at ucar.edu]
>>>> Sent: Thursday, March 29, 2012 10:26 AM
>>>> To: Raby, John W CIV (US)
>>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>>> (UNCLASSIFIED)
>>>>
>>>> John,
>>>>
>>>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
don't
>>>> fully understand your description of the problem. I see the
error
>>>> message your getting, but does that happen every time you run
>>>> METv3.1 PB2NC, or does it show up intermittently?
>>>>
>>>> Off the top of my head, one possibility is that when you built
>>>> METv3.1, you continued to link against a version of BUFRLIB that
may
>>>> have been compiled using an older version of the compiler.
That's
>>>> really just a shot in the dark, but you could try recompiling
>>>> BUFRLIB using your latest compiler version and then recompiling
MET,
>>>> being sure to do a "make clean"
>>>> first:
>>>>
>>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/co
>>>> m
>>>> p
>>>> ilation/req_libs.php
>>>>
>>>> If that doesn't fix the problem, I'll need to try to replicate
the
>>>> problem here to debug it. Can you send me your "user_defs.mk"
file,
>>>> your PB2NC Config file, and the PREPBUFR file you're using?
>>>>
>>>> Also, I wanted to mention that we posted an updated set of
bugfixes
>>>> for
>>>> METv3.1 this morning. You can access it here:
>>>>
>>>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>>>> p
>>>> hp
>>>>
>>>> Thanks,
>>>> John Halley Gotway
>>>>
>>>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>>>
>>>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>>>> Queue: met_help
>>>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>>>> Owner: Nobody
>>>>> Requestors: john.w.raby2.civ at mail.mil
>>>>> Status: new
>>>>> Ticket<URL:
>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>>
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>> Do you see anything obvious below which would explain why MET
PB2NC
>>>>> fails?
>>>>> This repeats for every hour 0-24.
>>>>>
>>>>> I'm running MET V3.1 and checked the run command and the config
>>>>> file and don't see anything obvious.
>>>>>
>>>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>>>> Reading Default Config File:
>>>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>>>> DEBUG 1: Creating NetCDF File:
>>>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>>>> DEBUG 1: Processing PrepBufr File:
>>>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>>>> DEBUG 1: Blocking PrepBufr file to:
/tmp/tmp_pb2nc_blk_31543_0
>>>>>
>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC
.REHOVI
>>>>>
>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>>
>>>>> Thanks.
>>>>>
>>>>> R/
>>>>> John
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>>
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: FW: PB2NC Problem (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Wed Apr 11 14:02:42 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Were you able to see if there is a difference between running the
PB2NC
version from METV3.0 and running the version from METV3.0.1? I can
only run
PB2NC in METV3.0.
Also regards your answer below on running the Rscript, are you running
this in
UNIX? I've tried running the script in windows using the R Gui and all
I can
get are errors.
Thanks.
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, April 10, 2012 3:15 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
John,
I don't typically use an R GUI, so I'm not sure. We always run them
on a
command line, using the Rscript command:
Rscript plot_cnt.R (whatever arguments you need to supply)
John
On 04/10/2012 03:08 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Thanks, but I can't get the V3.0.1 version to run either without
that error.
>
> I can run it fine with V3.0.
>
> New subj: How do you run an R Script (plot_cnt.R)? I have the R Gui
up
> and can't seem to get it to load that script. See following dialog:
>
> load("C:\\Users\\john.raby\\Documents\\Userdata_24FEB 11\\Old PC -
> B\\Evaluation Project\\R Scripts for Plotting\\plot_cnt.R.txt")
> Error: bad restore file magic number (file may be corrupted) -- no
> data loaded In addition: Warning message:
> file 'plot_cnt.R.txt' has magic number '#####'
> Use of save versions prior to 2 is deprecated
>>
>
> Thanks.
>
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, April 10, 2012 2:11 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> I wasn't sure on this, so I did the following - I used ncdump to
dump
> the contents of the output of the PB2NC test script to a text file
for
> both
> METv3.0.1 and METv3.1. Then I looked at their differences and found
> the only differences in the MET_version attribute and data of
> generation. All of the data values were identical.
>
> Based on that, I'd say that using PB2NC from METv3.0.1 would likely
be fine.
>
> Thanks,
> John
>
> On 04/10/2012 07:58 AM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> John -
>>
>> Do you know if there is a significant difference between the
version
>> of PB2NC in METV3.0 and that of MET V3.1? My desire would be to
>> perform some processing to get ahead on some PrepBUFR data I need,
>> rather than wait to resolve this issue with our METV3.1. If there
is
>> a significant difference, then I would hold off.
>> Thanks.
>> R/
>> John
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Friday, April 06, 2012 1:45 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Yes, I saw that about METv3.0. If I were in your position, I would
>> start from scratch with METv3.1. Just compile NetCDF, GSL, and
>> BUFRLIB using the current GNU compiler version (4.4.4 on your
>> system), and then compile MET using it as well. My theory is that
>> one of the packages is out of sync somewhere.
>>
>> That whole recompiling process could be as quick as 15 minutes or
so,
>> following the suggestions on the MET online tutorial:
>>
>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/com
>> p
>> ilation/req_libs.php
>>
>> Thanks,
>> John
>>
>> On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> John -
>>>
>>> Thanks for the analysis/discussion on possible things to look at.
We
>>> are taking the tact that we may need to recompile METV3.1 using a
>>> different compiler, but I don't think this has been done yet. Will
>>> let you know the results.
>>>
>>> I think I mentioned to you in another email that we were able to
run
>>> PB2NC without error if we use METV3.0.
>>>
>>> Will keep you posted.
>>>
>>> R/
>>> John
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Friday, April 06, 2012 1:24 PM
>>> To: Raby, John W CIV (US)
>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>> (UNCLASSIFIED)
>>>
>>> John,
>>>
>>> Sorry for the delay in getting back to you. I've been testing
>>> things trying to replicate the error but am unable to do so. It's
a
>>> BUFRLIB error we have not previously encountered. I tried to test
>>> my theory that it may be due to inconsistent compiler versions,
but
>>> I do not have adequate compiler versions available to test it out.
>>>
>>> Generally, when there are issues with BUFRLIB, my two questions
are:
>>> - Are the versions of the C and FORTRAN compiler used to compile
>>> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
>>> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>>>
>>> I don't think the -DBLOCK4 is the issue. Having it set wrong
>>> produces a different error from BUFRLIB.
>>>
>>> In your user_defs.mk file, I see that you're compiling MET using
GNU
>>> version 4.4.4. We are using GNU version 4.4.5, and are having no
>>> problems. And when we used earlier versions of the GNU compilers,
>>> it worked fine too. I do see the following in that file:
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>>>
>>> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
>>> was compiled using GNU 4.4.4? If not, try using
>>> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>>>
>>> Sorry I can't be of more help at this point. Have you been able
to
>>> make any progress?
>>>
>>> Thanks,
>>> John
>>>
>>>
>>> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>
>>>> John -
>>>>
>>>> I've attached the files you mention below. Could you see if you
can
>>>> replicate the error?
>>>> Thanks.
>>>> R/
>>>> John
>>>> ________________________________________
>>>> From: John Halley Gotway via RT [met_help at ucar.edu]
>>>> Sent: Thursday, March 29, 2012 10:26 AM
>>>> To: Raby, John W CIV (US)
>>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>>> (UNCLASSIFIED)
>>>>
>>>> John,
>>>>
>>>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
>>>> don't fully understand your description of the problem. I see
the
>>>> error message your getting, but does that happen every time you
run
>>>> METv3.1 PB2NC, or does it show up intermittently?
>>>>
>>>> Off the top of my head, one possibility is that when you built
>>>> METv3.1, you continued to link against a version of BUFRLIB that
>>>> may have been compiled using an older version of the compiler.
>>>> That's really just a shot in the dark, but you could try
>>>> recompiling BUFRLIB using your latest compiler version and then
>>>> recompiling MET, being sure to do a "make clean"
>>>> first:
>>>>
>>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/c
>>>> o
>>>> m
>>>> p
>>>> ilation/req_libs.php
>>>>
>>>> If that doesn't fix the problem, I'll need to try to replicate
the
>>>> problem here to debug it. Can you send me your "user_defs.mk"
>>>> file, your PB2NC Config file, and the PREPBUFR file you're using?
>>>>
>>>> Also, I wanted to mention that we posted an updated set of
bugfixes
>>>> for
>>>> METv3.1 this morning. You can access it here:
>>>>
>>>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>>>> p
>>>> hp
>>>>
>>>> Thanks,
>>>> John Halley Gotway
>>>>
>>>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>>>
>>>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>>>> Queue: met_help
>>>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>>>> Owner: Nobody
>>>>> Requestors: john.w.raby2.civ at mail.mil
>>>>> Status: new
>>>>> Ticket<URL:
>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>>
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>> Do you see anything obvious below which would explain why MET
>>>>> PB2NC fails?
>>>>> This repeats for every hour 0-24.
>>>>>
>>>>> I'm running MET V3.1 and checked the run command and the config
>>>>> file and don't see anything obvious.
>>>>>
>>>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>>>> Reading Default Config File:
>>>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>>>> DEBUG 1: Reading User Config File: ./PB2NCConfig_LAXd02_hr2
>>>>> DEBUG 1: Creating NetCDF File:
>>>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>>>> DEBUG 1: Processing PrepBufr File:
>>>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>>>> DEBUG 1: Blocking PrepBufr file to:
/tmp/tmp_pb2nc_blk_31543_0
>>>>>
>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC
>>>>> .REHOVI
>>>>>
>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>>
>>>>> Thanks.
>>>>>
>>>>> R/
>>>>> John
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>>
>>>>>
>>>>> Classification: UNCLASSIFIED
>>>>> Caveats: NONE
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem (UNCLASSIFIED)
From: John Halley Gotway
Time: Thu Apr 12 16:33:44 2012
John,
Sorry for the delay in getting back to you on this. There are in fact
real differences in the output of the PB2NC test script. The
differences are confined to relative humidity observations and are
due to the fact that we modified how it is derived from the observed
state variables. We modified that derivation to be consistent with
the algorithm NCEP uses.
Other than RH, the observations are the same.
Generally speaking, to see the list of changes from version to
version, I'd refer you to the release notes:
http://www.dtcenter.org/met/users/support/release_notes/index.php
Hope that helps clarify.
Thanks,
John
On 04/11/2012 02:02 PM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Were you able to see if there is a difference between running the
PB2NC
> version from METV3.0 and running the version from METV3.0.1? I can
only run
> PB2NC in METV3.0.
>
> Also regards your answer below on running the Rscript, are you
running this in
> UNIX? I've tried running the script in windows using the R Gui and
all I can
> get are errors.
>
> Thanks.
>
> John
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, April 10, 2012 3:15 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>
> John,
>
> I don't typically use an R GUI, so I'm not sure. We always run them
on a
> command line, using the Rscript command:
> Rscript plot_cnt.R (whatever arguments you need to supply)
>
> John
>
>
>
> On 04/10/2012 03:08 PM, Raby, John W USA CIV via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> John -
>>
>> Thanks, but I can't get the V3.0.1 version to run either without
that error.
>>
>> I can run it fine with V3.0.
>>
>> New subj: How do you run an R Script (plot_cnt.R)? I have the R Gui
up
>> and can't seem to get it to load that script. See following dialog:
>>
>> load("C:\\Users\\john.raby\\Documents\\Userdata_24FEB 11\\Old PC -
>> B\\Evaluation Project\\R Scripts for Plotting\\plot_cnt.R.txt")
>> Error: bad restore file magic number (file may be corrupted) -- no
>> data loaded In addition: Warning message:
>> file 'plot_cnt.R.txt' has magic number '#####'
>> Use of save versions prior to 2 is deprecated
>>>
>>
>> Thanks.
>>
>> John
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>> Sent: Tuesday, April 10, 2012 2:11 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
(UNCLASSIFIED)
>>
>> John,
>>
>> I wasn't sure on this, so I did the following - I used ncdump to
dump
>> the contents of the output of the PB2NC test script to a text file
for
>> both
>> METv3.0.1 and METv3.1. Then I looked at their differences and
found
>> the only differences in the MET_version attribute and data of
>> generation. All of the data values were identical.
>>
>> Based on that, I'd say that using PB2NC from METv3.0.1 would likely
be fine.
>>
>> Thanks,
>> John
>>
>> On 04/10/2012 07:58 AM, Raby, John W USA CIV via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> John -
>>>
>>> Do you know if there is a significant difference between the
version
>>> of PB2NC in METV3.0 and that of MET V3.1? My desire would be to
>>> perform some processing to get ahead on some PrepBUFR data I need,
>>> rather than wait to resolve this issue with our METV3.1. If there
is
>>> a significant difference, then I would hold off.
>>> Thanks.
>>> R/
>>> John
>>>
>>> -----Original Message-----
>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>> Sent: Friday, April 06, 2012 1:45 PM
>>> To: Raby, John W CIV (US)
>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>> (UNCLASSIFIED)
>>>
>>> John,
>>>
>>> Yes, I saw that about METv3.0. If I were in your position, I
would
>>> start from scratch with METv3.1. Just compile NetCDF, GSL, and
>>> BUFRLIB using the current GNU compiler version (4.4.4 on your
>>> system), and then compile MET using it as well. My theory is that
>>> one of the packages is out of sync somewhere.
>>>
>>> That whole recompiling process could be as quick as 15 minutes or
so,
>>> following the suggestions on the MET online tutorial:
>>>
>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/com
>>> p
>>> ilation/req_libs.php
>>>
>>> Thanks,
>>> John
>>>
>>> On 04/06/2012 01:38 PM, Raby, John W USA CIV via RT wrote:
>>>>
>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>> John -
>>>>
>>>> Thanks for the analysis/discussion on possible things to look at.
We
>>>> are taking the tact that we may need to recompile METV3.1 using a
>>>> different compiler, but I don't think this has been done yet.
Will
>>>> let you know the results.
>>>>
>>>> I think I mentioned to you in another email that we were able to
run
>>>> PB2NC without error if we use METV3.0.
>>>>
>>>> Will keep you posted.
>>>>
>>>> R/
>>>> John
>>>>
>>>> -----Original Message-----
>>>> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
>>>> Sent: Friday, April 06, 2012 1:24 PM
>>>> To: Raby, John W CIV (US)
>>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>>> (UNCLASSIFIED)
>>>>
>>>> John,
>>>>
>>>> Sorry for the delay in getting back to you. I've been testing
>>>> things trying to replicate the error but am unable to do so.
It's a
>>>> BUFRLIB error we have not previously encountered. I tried to
test
>>>> my theory that it may be due to inconsistent compiler versions,
but
>>>> I do not have adequate compiler versions available to test it
out.
>>>>
>>>> Generally, when there are issues with BUFRLIB, my two questions
are:
>>>> - Are the versions of the C and FORTRAN compiler used to compile
>>>> BUFRLIB consistent with the C++ and FORTRAN compilers used to
compile MET?
>>>> - Should we be using the "-DBLOCK4" option in user_def.mk on this
machine?
>>>>
>>>> I don't think the -DBLOCK4 is the issue. Having it set wrong
>>>> produces a different error from BUFRLIB.
>>>>
>>>> In your user_defs.mk file, I see that you're compiling MET using
GNU
>>>> version 4.4.4. We are using GNU version 4.4.5, and are having no
>>>> problems. And when we used earlier versions of the GNU
compilers,
>>>> it worked fine too. I do see the following in that file:
>>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc412
>>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1_gcc444
>>>> BUFR_BASE = /opt/MET/BUFRLIB/v10.0.1
>>>>
>>>> Do you know if the version of BUFRLIB in /opt/MET/BUFRLIB/v10.0.1
>>>> was compiled using GNU 4.4.4? If not, try using
>>>> /opt/MET/BUFRLIB/v10.0.1_gcc444.
>>>>
>>>> Sorry I can't be of more help at this point. Have you been able
to
>>>> make any progress?
>>>>
>>>> Thanks,
>>>> John
>>>>
>>>>
>>>> On 04/04/2012 12:50 PM, Raby, John W USA CIV via RT wrote:
>>>>>
>>>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>>
>>>>> John -
>>>>>
>>>>> I've attached the files you mention below. Could you see if you
can
>>>>> replicate the error?
>>>>> Thanks.
>>>>> R/
>>>>> John
>>>>> ________________________________________
>>>>> From: John Halley Gotway via RT [met_help at ucar.edu]
>>>>> Sent: Thursday, March 29, 2012 10:26 AM
>>>>> To: Raby, John W CIV (US)
>>>>> Subject: Re: [rt.rap.ucar.edu #55564] FW: PB2NC Problem
>>>>> (UNCLASSIFIED)
>>>>>
>>>>> John,
>>>>>
>>>>> Hmmm, that's a new error message from PB2NC. Unfortunately, I
>>>>> don't fully understand your description of the problem. I see
the
>>>>> error message your getting, but does that happen every time you
run
>>>>> METv3.1 PB2NC, or does it show up intermittently?
>>>>>
>>>>> Off the top of my head, one possibility is that when you built
>>>>> METv3.1, you continued to link against a version of BUFRLIB that
>>>>> may have been compiled using an older version of the compiler.
>>>>> That's really just a shot in the dark, but you could try
>>>>> recompiling BUFRLIB using your latest compiler version and then
>>>>> recompiling MET, being sure to do a "make clean"
>>>>> first:
>>>>>
>>>>>
http://www.dtcenter.org/met/users/support/online_tutorial/METv3.1/c
>>>>> o
>>>>> m
>>>>> p
>>>>> ilation/req_libs.php
>>>>>
>>>>> If that doesn't fix the problem, I'll need to try to replicate
the
>>>>> problem here to debug it. Can you send me your "user_defs.mk"
>>>>> file, your PB2NC Config file, and the PREPBUFR file you're
using?
>>>>>
>>>>> Also, I wanted to mention that we posted an updated set of
bugfixes
>>>>> for
>>>>> METv3.1 this morning. You can access it here:
>>>>>
>>>>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.
>>>>> p
>>>>> hp
>>>>>
>>>>> Thanks,
>>>>> John Halley Gotway
>>>>>
>>>>> On 03/29/2012 09:41 AM, Raby, John W USA CIV via RT wrote:
>>>>>>
>>>>>> Thu Mar 29 09:41:49 2012: Request 55564 was acted upon.
>>>>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>>>>> Queue: met_help
>>>>>> Subject: FW: PB2NC Problem (UNCLASSIFIED)
>>>>>> Owner: Nobody
>>>>>> Requestors: john.w.raby2.civ at mail.mil
>>>>>> Status: new
>>>>>> Ticket<URL:
>>>>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55564>
>>>>>>
>>>>>>
>>>>>> Classification: UNCLASSIFIED
>>>>>> Caveats: NONE
>>>>>>
>>>>>> Do you see anything obvious below which would explain why MET
>>>>>> PB2NC fails?
>>>>>> This repeats for every hour 0-24.
>>>>>>
>>>>>> I'm running MET V3.1 and checked the run command and the config
>>>>>> file and don't see anything obvious.
>>>>>>
>>>>>> ***Running PB2NC on PrepBufr files*** ***Files for use with WRF
>>>>>> initialized at 06Z on 20120110*** hour 0 DEBUG 1:
>>>>>> Reading Default Config File:
>>>>>> /opt/MET/MET_3.1/data/config/PB2NCConfig_default
>>>>>> DEBUG 1: Reading User Config File:
./PB2NCConfig_LAXd02_hr2
>>>>>> DEBUG 1: Creating NetCDF File:
>>>>>> ../ncobs/20120110/LAXd02_06_00_pb.nc
>>>>>> DEBUG 1: Processing PrepBufr File:
>>>>>> /opt3b/PrepBUFR/20120110/ndas.t12z.prepbufr.tm06.nr
>>>>>> DEBUG 1: Blocking PrepBufr file to:
/tmp/tmp_pb2nc_blk_31543_0
>>>>>>
>>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>>> BUFRLIB: TABENT - UNDEFINED TAG ( ) FOR INPUT MNEMONIC
>>>>>> .REHOVI
>>>>>>
>>>>>> ***********BUFR ARCHIVE LIBRARY ABORT**************
>>>>>>
>>>>>> Thanks.
>>>>>>
>>>>>> R/
>>>>>> John
>>>>>>
>>>>>> Classification: UNCLASSIFIED
>>>>>> Caveats: NONE
>>>>>>
>>>>>>
>>>>>>
>>>>>> Classification: UNCLASSIFIED
>>>>>> Caveats: NONE
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> Classification: UNCLASSIFIED
>>>> Caveats: NONE
>>>>
>>>>
>>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
More information about the Met_help
mailing list