[Met_help] [rt.rap.ucar.edu #38372] History for FW: Format of PrepBuffer Dump
RAL HelpDesk {for John Halley Gotway}
met_help at ucar.edu
Wed Jun 9 10:18:01 MDT 2010
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
John, that helped. Now I am able to progress to my next error. In the
config file (attached), I should only be processing ADPSFC data but in the
prepdump directory I get data for ADPUPA (List is below). Also not the
NVNWIN error. I was told that the following prepbuffer file is the one to
use: gdas1.t00z.prepbufr.nr. Is that true?
Thanks Bob
pb2nc_wrf
Reading Config File: /sphome/qcteam/METv2.0/data/config/PB2NCConfig_wrf
Creating NetCDF File: /h/data/global/WXQC/data/obs_prep/Jun07_2010_00z
Processing PrepBufr File:
/h/data/global/WXQC/data/obs_prep/raw/gdas1.t00z.prepbufr.nr
Dumping to ASCII output directory: /sphome/qcteam/prepdump
"dumppb.f", line 79: 1525-126 The file name provided in the OPEN statement
for unit 22 contains a null character. The program will re cover by using
the file name /h/data/global/WXQC/data/obs_prep/pb2nc_868416_0_blk.pb.
**************BUFR ARCHIVE LIBRARY ABORT*****************
BUFRLIB: NVNWIN - THE NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE LIMIT,
NMAX ( 10)
**************BUFR ARCHIVE LIBRARY ABORT*****************
ls -l
total 3072
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_ADPSFC.txt
-rw-rw-r-- 1 qcteam qc 2097152 Jun 07 20:49
dump_gdas1.t00z.prepbufr.nr_ADPUPA.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_AIRCAR.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_AIRCFT.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_ERS1DA.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_GOESND.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_GPSIPW.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_MSONET.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_PROFLR.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_QKSWND.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_RASSDA.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SATEMP.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SATWND.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SFCBOG.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SFCSHP.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SPSSMI.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_SYNDAT.txt
-rw-rw-r-- 1 qcteam qc 0 Jun 07 20:48
dump_gdas1.t00z.prepbufr.nr_VADWND.txt
-----Original Message-----
From: John Halley Gotway [mailto:johnhg at ucar.edu]
Sent: Monday, June 07, 2010 3:37 PM
To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Subject: Re: [Met_help] Format of PrepBuffer Dump
Bob,
It actually is a directory into which multiple text files are written - one
for each message type. Here's a snippet from the PB2NC usage statement
about it:
"-dump path" indicates that the entire contents of "prepbufr_file" should
also be dumped to text files in the directory specified (optional).
Just make a directory into which those files should be written:
mkdir /sphome/qcteam/pb2ncdump
Hopefully that'll solve it.
John
Craig, Robert J Civ USAF AFWA 16 WS/WXN wrote:
> John, I got pulled off of this for a while but now I'm back. When I
> run
the following code, I get the error shown below.
>
> /sphome/qcteam/METv2.0/bin/pb2nc
/h/data/global/WXQC/data/obs_prep/raw/gdas1.t00z.bf00
/h/data/global/WXQC/data/obs_prep/Jun07_2010_00z
/sphome/qcteam/METv2.0/data/config/PB2NCConfig_wrf -dump
/sphome/qcteam/pb2ncdump
>
> pb2nc_wrf
> Reading Config File: /sphome/qcteam/METv2.0/data/config/PB2NCConfig_wrf
> Creating NetCDF File: /h/data/global/WXQC/data/obs_prep/Jun07_2010_00z
> Processing PrepBufr File:
/h/data/global/WXQC/data/obs_prep/raw/gdas1.t00z.bf00
> Dumping to ASCII output directory: /sphome/qcteam/pb2ncdump
> "dumppb.f", line 72: 1525-014 The I/O operation on unit 51 cannot be
completed because an errno value of 2 (No such file or directory)
> was received while opening the file. The program will stop.
>
> I wanted to see the dump file to see what is in the nc file. The
permissions on the /sphome/qcteam directory are wide open. Since this is a
dump file, it shouldn't need to exist before this program is run.
>
> I am missing something?
>
> Thanks
>
> Bob
>
> -----Original Message-----
> From: John Halley Gotway [mailto:johnhg at ucar.edu]
> Sent: Friday, April 23, 2010 2:50 PM
> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
> Cc: met_help at ucar.edu
> Subject: Re: [Met_help] Format of PrepBuffer Dump
>
> Bob,
>
> Unfortunately no, we have no documentation on that ASCII dump.
> Basically,
we used some sample code from NCEP to dump the contents of a PREPBUFR file
to ASCII. They code is contained in the BUFRLIB
> user's guide: http://www.nco.ncep.noaa.gov/sib/decoders/BUFRLIB/toc/
>
> We found it to be useful in testing, so we included that functionality
> in
the PB2NC tool itself - primarily for testing purposes. The file of
interest in the PB2NC code that handles this
> functionality is: METv2.0/src/pb2nc/dumppb.f
>
> Sorry I can't be of more help.
>
> John
>
> Craig, Robert J Civ USAF AFWA 16 WS/WXN wrote:
>> John, what is the format of the PrepBuffer Dump file? Is there
documentation on this?
>>
>> Thanks
>>
>> Bob
>>
>>
>>
>> ---------------------------------------------------------------------
>> ---
>>
>> _______________________________________________
>> Met_help mailing list
>> Met_help at mailman.ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/met_help
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: FW: [Met_help] Format of PrepBuffer Dump
From: John Halley Gotway
Time: Tue Jun 08 09:41:36 2010
Bob,
The GDAS PrepBufr files contain global observations. The NDAS
PrepBufr files contain observations for only North America. Links to
those datasets are posted here:
http://www.dtcenter.org/met/users/downloads/observation_data.php
If your domain of interest is North America, use NDAS. Otherwise, use
GDAS. Unfortunately, the GDAS files are very large and take a long
time for PB2NC to work through. I see in the config file you sent
that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
Regarding using the "-dump" option, we intended that only as a sanity
check. It's fine to do it once, but it really take a long time to
write out all those observations to ASCII. In METv2.0, the -dump
option dumps all possible message types to ASCII - that's why you're
seeing output for ADPUPA instead of just ADPSFC. In the upcoming
METv3.0 release, we're changing that logic to only dump out the
specific message types the user has requested.
I just ran the same GDAS data that you're using through PB2NC, and I'm
not seeing that NVNWIN error. Are you still having problems with that
or no?
Thanks,
John
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
From: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Time: Tue Jun 08 10:10:17 2010
I switched to using the NDAS file and that worked. I am still getting
the error on the GDAS file. Most of what I do won't be in the
Northern Hemisphere so GDAS will need to work. In the literature
there is discussion about running prep buffer on 64 bit machine (that
is what we are running currently). Is this still an issue and would
the following error result from that?
**************BUFR ARCHIVE
LIBRARY ABORT*****************
BUFRLIB: NVNWIN - THE NUMBER OF
EVENTS, NVNWIN ( 25) EXCEEDS THE LIMIT, NMAX ( 10)
**************BUFR ARCHIVE LIBRARY ABORT*****************
Thanks
Bob
-----Original Message-----
From: RAL HelpDesk {for John Halley
Gotway} [mailto:met_help at ucar.edu]
Sent: Tuesday, June 08, 2010
10:42 AM
To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Subject:
[rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
Bob,
The GDAS PrepBufr files contain global observations. The NDAS
PrepBufr files contain observations for only North America. Links to
those datasets are posted here:
http://www.dtcenter.org/met/users/downloads/observation_data.php
If
your domain of interest is North America, use NDAS. Otherwise, use
GDAS. Unfortunately, the GDAS files are very large and take a long
time for PB2NC to work through. I see in the config file you sent
that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
Regarding using the
"-dump" option, we intended that only as a sanity check. It's fine to
do it once, but it really take a long time to write out all those
observations to ASCII. In METv2.0, the -dump option dumps all
possible message types to ASCII - that's why you're seeing output for
ADPUPA instead of just ADPSFC. In the upcoming METv3.0 release, we're
changing that logic to only dump out the specific message types the
user has requested.
I just ran the same GDAS data that you're using
through PB2NC, and I'm not seeing that NVNWIN error. Are you still
having problems with that or no?
Thanks,
John
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
From: John Halley Gotway
Time: Tue Jun 08 10:20:14 2010
Bob,
I really am not sure if this would be that would be the cause of the
error you're seeing. But I'm happy to try compiling and running on a
64-bit machine here to see if I can replicate the error. I
do have a few questions though.
- Just to be clear, this error only occurs when you're running GDAS
data, right? Does it occur both with and without the "-dump" option?
- Also, does the "METv2.0/scripts/test_pb2nc.sh" script run without
this error?
- Can you remind me what C++ and Fortran compilers you're using to
build MET - and also which version #'s if you have them?
Thanks,
John
RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>
> I switched to using the NDAS file and that worked. I am still
getting the error on the GDAS file. Most of what I do won't be in the
Northern Hemisphere so GDAS will need to work. In the literature
there is discussion about running prep buffer on 64 bit machine (that
is what we are running currently). Is this still an issue and would
the following error result from that?
>
> **************BUFR ARCHIVE LIBRARY ABORT*****************
> BUFRLIB: NVNWIN - THE NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE
LIMIT, NMAX ( 10)
> **************BUFR ARCHIVE LIBRARY ABORT*****************
>
> Thanks
>
> Bob
>
> -----Original Message-----
> From: RAL HelpDesk {for John Halley Gotway}
[mailto:met_help at ucar.edu]
> Sent: Tuesday, June 08, 2010 10:42 AM
> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
> Subject: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer Dump
>
> Bob,
>
> The GDAS PrepBufr files contain global observations. The NDAS
PrepBufr files contain observations for only North America. Links to
those datasets are posted here:
> http://www.dtcenter.org/met/users/downloads/observation_data.php
>
> If your domain of interest is North America, use NDAS. Otherwise,
use GDAS. Unfortunately, the GDAS files are very large and take a
long time for PB2NC to work through. I see in the config file you
sent that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
>
> Regarding using the "-dump" option, we intended that only as a
sanity check. It's fine to do it once, but it really take a long time
to write out all those observations to ASCII. In METv2.0, the -dump
option dumps all possible message types to ASCII - that's why you're
seeing output for ADPUPA instead of just ADPSFC. In the upcoming
METv3.0 release, we're changing that logic to only dump out the
specific message types the user has requested.
>
> I just ran the same GDAS data that you're using through PB2NC, and
I'm not seeing that NVNWIN error. Are you still having problems with
that or no?
>
> Thanks,
> John
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
From: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Time: Tue Jun 08 11:47:04 2010
John,
1) GDAS errors off with or without the dump option - same
error
2) I ran the test script and it worked but it is reading the
much smaller NDAS file
3) We used xlc and xlf compilers. I don't
know the version numbers but should have those later today.
When I
ran the GDAS version with the -v 2 option set, the last message it
printed before it errored is " Processing 352644 PrepBufr messages..."
Hope this helps.
Bob
-----Original Message-----
From:
RAL HelpDesk {for John Halley Gotway} [mailto:met_help at ucar.edu]
Sent: Tuesday, June 08, 2010 11:20 AM
To: Craig, Robert J Civ USAF
AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help]
Format of PrepBuffer Dump
Bob,
I really am not sure if this
would be that would be the cause of the error you're seeing. But I'm
happy to try compiling and running on a 64-bit machine here to see if
I can replicate the error. I
do have a few questions though.
- Just
to be clear, this error only occurs when you're running GDAS data,
right? Does it occur both with and without the "-dump" option?
-
Also, does the "METv2.0/scripts/test_pb2nc.sh" script run without this
error?
- Can you remind me what C++ and Fortran compilers you're
using to build MET - and also which version #'s if you have them?
Thanks,
John
RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16
WS/WXN} wrote:
> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>
> I
switched to using the NDAS file and that worked. I am still getting
the error on the GDAS file. Most of what I do won't be in the
Northern Hemisphere so GDAS will need to work. In the literature
there is discussion about running prep buffer on 64 bit machine (that
is what we are running currently). Is this still an issue and would
the following error result from that?
>
> **************BUFR
ARCHIVE LIBRARY ABORT*****************
> BUFRLIB: NVNWIN - THE
NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE LIMIT, NMAX ( 10)
> **************BUFR ARCHIVE LIBRARY ABORT*****************
>
>
Thanks
>
> Bob
>
> -----Original Message-----
> From: RAL
HelpDesk {for John Halley Gotway} [mailto:met_help at ucar.edu]
> Sent:
Tuesday, June 08, 2010 10:42 AM
> To: Craig, Robert J Civ USAF AFWA
16 WS/WXN
> Subject: [rt.rap.ucar.edu #38372] FW: [Met_help] Format
of PrepBuffer Dump
>
> Bob,
>
> The GDAS PrepBufr files contain
global observations. The NDAS PrepBufr files contain observations for
only North America. Links to those datasets are posted here:
>
http://www.dtcenter.org/met/users/downloads/observation_data.php
>
> If your domain of interest is North America, use NDAS. Otherwise,
use GDAS. Unfortunately, the GDAS files are very large and take a
long time for PB2NC to work through. I see in the config file you
sent that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
>
> Regarding using
the "-dump" option, we intended that only as a sanity check. It's
fine to do it once, but it really take a long time to write out all
those observations to ASCII. In METv2.0, the -dump option dumps all
possible message types to ASCII - that's why you're seeing output for
ADPUPA instead of just ADPSFC. In the upcoming METv3.0 release, we're
changing that logic to only dump out the specific message types the
user has requested.
>
> I just ran the same GDAS data that you're
using through PB2NC, and I'm not seeing that NVNWIN error. Are you
still having problems with that or no?
>
> Thanks,
> John
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
From: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Time: Tue Jun 08 14:39:52 2010
John, our compiler versions are:
XLC - 8.0.0.13
XLF - 11.1.0.6
Bob
-----Original Message-----
From: RAL HelpDesk {for John
Halley Gotway} [mailto:met_help at ucar.edu]
Sent: Tuesday, June 08,
2010 11:20 AM
To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Subject:
Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
Bob,
I really am not sure if this would be that would be the cause
of the error you're seeing. But I'm happy to try compiling and
running on a 64-bit machine here to see if I can replicate the error.
I
do have a few questions though.
- Just to be clear, this error
only occurs when you're running GDAS data, right? Does it occur both
with and without the "-dump" option?
- Also, does the
"METv2.0/scripts/test_pb2nc.sh" script run without this error?
- Can
you remind me what C++ and Fortran compilers you're using to build MET
- and also which version #'s if you have them?
Thanks,
John
RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
>
<URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>
>
I switched to using the NDAS file and that worked. I am still getting
the error on the GDAS file. Most of what I do won't be in the
Northern Hemisphere so GDAS will need to work. In the literature
there is discussion about running prep buffer on 64 bit machine (that
is what we are running currently). Is this still an issue and would
the following error result from that?
>
> **************BUFR
ARCHIVE LIBRARY ABORT*****************
> BUFRLIB: NVNWIN - THE
NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE LIMIT, NMAX ( 10)
> **************BUFR ARCHIVE LIBRARY ABORT*****************
>
>
Thanks
>
> Bob
>
> -----Original Message-----
> From: RAL
HelpDesk {for John Halley Gotway} [mailto:met_help at ucar.edu]
> Sent:
Tuesday, June 08, 2010 10:42 AM
> To: Craig, Robert J Civ USAF AFWA
16 WS/WXN
> Subject: [rt.rap.ucar.edu #38372] FW: [Met_help] Format
of PrepBuffer Dump
>
> Bob,
>
> The GDAS PrepBufr files contain
global observations. The NDAS PrepBufr files contain observations for
only North America. Links to those datasets are posted here:
>
http://www.dtcenter.org/met/users/downloads/observation_data.php
>
> If your domain of interest is North America, use NDAS. Otherwise,
use GDAS. Unfortunately, the GDAS files are very large and take a
long time for PB2NC to work through. I see in the config file you
sent that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
>
> Regarding using
the "-dump" option, we intended that only as a sanity check. It's
fine to do it once, but it really take a long time to write out all
those observations to ASCII. In METv2.0, the -dump option dumps all
possible message types to ASCII - that's why you're seeing output for
ADPUPA instead of just ADPSFC. In the upcoming METv3.0 release, we're
changing that logic to only dump out the specific message types the
user has requested.
>
> I just ran the same GDAS data that you're
using through PB2NC, and I'm not seeing that NVNWIN error. Are you
still having problems with that or no?
>
> Thanks,
> John
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of PrepBuffer Dump
From: John Halley Gotway
Time: Tue Jun 08 15:35:18 2010
Bob,
I tried running up on our supercomputer, bluefire, using the xlc/xlf
compilers. But I'm not seeing the nvnwin error you're seeing.
This may be tricky to track down. I'd like to rule out a difference
in BUFRLIB versions to make sure that isn't causing a problem. Would
you be able to send me the following:
(1) Your top-level MET Makefile.
(2) Go to the BUFRLIB directory you're using, tar it up, and send it
to me.
Those two things may help me track down what's going on.
Thanks,
John
RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>
> John, our compiler versions are:
>
> XLC - 8.0.0.13
> XLF - 11.1.0.6
>
> Bob
>
>
> -----Original Message-----
> From: RAL HelpDesk {for John Halley Gotway}
[mailto:met_help at ucar.edu]
> Sent: Tuesday, June 08, 2010 11:20 AM
> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer Dump
>
> Bob,
>
> I really am not sure if this would be that would be the cause of the
error you're seeing. But I'm happy to try compiling and running on a
64-bit machine here to see if I can replicate the error. I
> do have a few questions though.
> - Just to be clear, this error only occurs when you're running GDAS
data, right? Does it occur both with and without the "-dump" option?
> - Also, does the "METv2.0/scripts/test_pb2nc.sh" script run without
this error?
> - Can you remind me what C++ and Fortran compilers you're using to
build MET - and also which version #'s if you have them?
>
> Thanks,
> John
>
>
> RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>>
>> I switched to using the NDAS file and that worked. I am still
getting the error on the GDAS file. Most of what I do won't be in the
Northern Hemisphere so GDAS will need to work. In the literature
there is discussion about running prep buffer on 64 bit machine (that
is what we are running currently). Is this still an issue and would
the following error result from that?
>>
>> **************BUFR ARCHIVE LIBRARY ABORT*****************
>> BUFRLIB: NVNWIN - THE NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE
LIMIT, NMAX ( 10)
>> **************BUFR ARCHIVE LIBRARY ABORT*****************
>>
>> Thanks
>>
>> Bob
>>
>> -----Original Message-----
>> From: RAL HelpDesk {for John Halley Gotway}
[mailto:met_help at ucar.edu]
>> Sent: Tuesday, June 08, 2010 10:42 AM
>> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
>> Subject: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer Dump
>>
>> Bob,
>>
>> The GDAS PrepBufr files contain global observations. The NDAS
PrepBufr files contain observations for only North America. Links to
those datasets are posted here:
>> http://www.dtcenter.org/met/users/downloads/observation_data.php
>>
>> If your domain of interest is North America, use NDAS. Otherwise,
use GDAS. Unfortunately, the GDAS files are very large and take a
long time for PB2NC to work through. I see in the config file you
sent that you're masking out only observations over CONUS
(/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you could use the NDAS file instead.
>>
>> Regarding using the "-dump" option, we intended that only as a
sanity check. It's fine to do it once, but it really take a long time
to write out all those observations to ASCII. In METv2.0, the -dump
option dumps all possible message types to ASCII - that's why you're
seeing output for ADPUPA instead of just ADPSFC. In the upcoming
METv3.0 release, we're changing that logic to only dump out the
specific message types the user has requested.
>>
>> I just ran the same GDAS data that you're using through PB2NC, and
I'm not seeing that NVNWIN error. Are you still having problems with
that or no?
>>
>> Thanks,
>> John
>
------------------------------------------------
Subject: FW: [Met_help] Format of PrepBuffer Dump
From: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Time: Tue Jun 08 15:42:18 2010
Here they are. Thanks for your help with this.
Bob
-----Original Message-----
From: RAL HelpDesk {for John Halley Gotway} [mailto:met_help at ucar.edu]
Sent: Tuesday, June 08, 2010 4:35 PM
To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer Dump
Bob,
I tried running up on our supercomputer, bluefire, using the xlc/xlf
compilers. But I'm not seeing the nvnwin error you're seeing.
This may be tricky to track down. I'd like to rule out a difference
in
BUFRLIB versions to make sure that isn't causing a problem. Would you
be able
to send me the following:
(1) Your top-level MET Makefile.
(2) Go to the BUFRLIB directory you're using, tar it up, and send it
to me.
Those two things may help me track down what's going on.
Thanks,
John
RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>
> John, our compiler versions are:
>
> XLC - 8.0.0.13
> XLF - 11.1.0.6
>
> Bob
>
>
> -----Original Message-----
> From: RAL HelpDesk {for John Halley Gotway}
[mailto:met_help at ucar.edu]
> Sent: Tuesday, June 08, 2010 11:20 AM
> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
> Subject: Re: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer
> Dump
>
> Bob,
>
> I really am not sure if this would be that would be the cause of the
error
> you're seeing. But I'm happy to try compiling and running on a 64-
bit
> machine here to see if I can replicate the error. I
> do have a few questions though.
> - Just to be clear, this error only occurs when you're running GDAS
data,
> right? Does it occur both with and without the "-dump" option?
> - Also, does the "METv2.0/scripts/test_pb2nc.sh" script run without
this
> error?
> - Can you remind me what C++ and Fortran compilers you're using to
build
> MET - and also which version #'s if you have them?
>
> Thanks,
> John
>
>
> RAL HelpDesk {for Craig, Robert J Civ USAF AFWA 16 WS/WXN} wrote:
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38372 >
>>
>> I switched to using the NDAS file and that worked. I am still
getting the
>> error on the GDAS file. Most of what I do won't be in the Northern
>> Hemisphere so GDAS will need to work. In the literature there is
>> discussion about running prep buffer on 64 bit machine (that is
what we are
>> running currently). Is this still an issue and would the following
error
>> result from that?
>>
>> **************BUFR ARCHIVE LIBRARY ABORT*****************
>> BUFRLIB: NVNWIN - THE NUMBER OF EVENTS, NVNWIN ( 25) EXCEEDS THE
LIMIT,
>> NMAX ( 10)
>> **************BUFR ARCHIVE LIBRARY ABORT*****************
>>
>> Thanks
>>
>> Bob
>>
>> -----Original Message-----
>> From: RAL HelpDesk {for John Halley Gotway}
[mailto:met_help at ucar.edu]
>> Sent: Tuesday, June 08, 2010 10:42 AM
>> To: Craig, Robert J Civ USAF AFWA 16 WS/WXN
>> Subject: [rt.rap.ucar.edu #38372] FW: [Met_help] Format of
PrepBuffer Dump
>>
>> Bob,
>>
>> The GDAS PrepBufr files contain global observations. The NDAS
PrepBufr
>> files contain observations for only North America. Links to those
datasets
>> are posted here:
>> http://www.dtcenter.org/met/users/downloads/observation_data.php
>>
>> If your domain of interest is North America, use NDAS. Otherwise,
use
>> GDAS. Unfortunately, the GDAS files are very large and take a long
time
>> for PB2NC to work through. I see in the config file you sent that
you're
>> masking out only observations over CONUS
>> (/sphome/qcteam/METv2.0/data/poly/CONUS.poly). Therefore, I'm
guessing you
>> could use the NDAS file instead.
>>
>> Regarding using the "-dump" option, we intended that only as a
sanity
>> check. It's fine to do it once, but it really take a long time to
write
>> out all those observations to ASCII. In METv2.0, the -dump option
dumps
>> all possible message types to ASCII - that's why you're seeing
output for
>> ADPUPA instead of just ADPSFC. In the upcoming METv3.0 release,
we're
>> changing that logic to only dump out the specific message types the
user
>> has requested.
>>
>> I just ran the same GDAS data that you're using through PB2NC, and
I'm not
>> seeing that NVNWIN error. Are you still having problems with that
or no?
>>
>> Thanks,
>> John
>
------------------------------------------------
Subject: FW: [Met_help] Format of PrepBuffer Dump
From: John Halley Gotway
Time: Wed Jun 09 10:17:59 2010
Bob,
Great. Glad you got it working. I'm resolving this MET-Help ticket.
Thanks,
John
------------------------------------------------
More information about the Met_help
mailing list