[Ncep.list.nems.announce] Upcoming NEMS commit
Raghu Reddy
raghu.reddy at noaa.gov
Mon Mar 4 17:03:59 UTC 2019
Sorry about that! The message got sent out before it was complete.
Hi all,
Thanks to Terry for showing me how to duplicate the problem, I was able to trace the problem to the following module file:
The file: /scratch3/NCEPDEV/nwprod/lib/modulefiles/EnvVars/v1.0.2
It is likely caused by one of the following settings in that file:
setenv LMOD_AVAIL_STYLE "system:<en_grouped>"
setenv LMOD_EXACT_MATCH yes
setenv LMOD_PACKAGE_PATH $NCEPLIBS/modulefiles
Most likely the setting and it’s causing the problem is:
setenv LMOD_EXACT_MATCH yes
In the following file: /scratch4/NCEPDEV/global/save/Terry.McGuinness/git/global-workflow/modulefiles/module_base.theia
If you comment out the line:
module load EnvVars/v1.0.2
Then you don’t see the problem. So most likely it is that setting where you’re explicitly asking it for an exact match!
Can you please comment it out and see if that resolves the problem?
It is possible that you need that for some other reasons, but that is an issue that should be resolved by the team.
Thanks,
Raghu
From: Raghu Reddy [mailto:raghu.reddy at noaa.gov]
Sent: Monday, March 04, 2019 12:02 PM
To: 'Samuel Trahan - NOAA Affiliate'
Cc: 'Ratko Vasic'; 'NEMS developers'; Raghu Reddy - NOAA Affiliate
Subject: RE: [Ncep.list.nems.announce] Upcoming NEMS commit
Hi all,
Thanks to Terry for showing me how to duplicate the problem, I was able to trace the problem to the following module file:
The file: /scratch3/NCEPDEV/nwprod/lib/modulefiles/EnvVars/v1.0.2
It is likely caused by one of the following settings in that file:
setenv LMOD_AVAIL_STYLE "system:<en_grouped>"
setenv LMOD_EXACT_MATCH yes
setenv LMOD_PACKAGE_PATH $NCEPLIBS/modulefiles
Most likely the setting and it’s causing the problem is:
setenv LMOD_EXACT_MATCH yes
In the following file: /scratch4/NCEPDEV/global/save/Terry.McGuinness/git/global-workflow/modulefiles/module_base.theia
If you comment out the line:
From: Samuel Trahan - NOAA Affiliate [mailto:samuel.trahan at noaa.gov]
Sent: Tuesday, February 26, 2019 11:16 AM
To: Raghu Reddy
Cc: Ratko Vasic; NEMS developers
Subject: Re: [Ncep.list.nems.announce] Upcoming NEMS commit
Raghu,
Both the NEMSfv3gfs compsets (AKA regression tests) and the FV3 GFS workflow have seen "module load slurm" fail to find anything, while "module load slurm/default" (or another explicit version) doesn't fail (yet).
Sincerely,
Sam Trahan
On Tue, 26 Feb 2019 at 10:36, Raghu Reddy <raghu.reddy at noaa.gov> wrote:
Hi Ratko,
I have a quick note and a question about the last item:
>>>> 6. NEMSfv3gfs: On Jet, "module load slurm" is replaced by "module load slurm/default"
The need to load the module is a temporary one, and I agree that even if it is temporary you need to do this until this is no longer needed. The requirement to load the slurm module will be removed after the transition.
One question I have is why the change from loading "slurm" to "slurm/default"? I'm just trying to find out if from a system administration perspective if this is a difference only between Theia and Jet? We are generally trying to keep them consistent, but I agree there may be times where we cannot, but this difference should not be there. So I was just curious.
Thanks,
Raghu
-----Original Message-----
From: ncep.list.nems.announce-bounces at lstsrv.ncep.noaa.gov [mailto:ncep.list.nems.announce-bounces at lstsrv.ncep.noaa.gov] On Behalf Of Ratko Vasic
Sent: Tuesday, February 26, 2019 10:31 AM
To: NEMS developers
Subject: [Ncep.list.nems.announce] Upcoming NEMS commit
Dear NEMS developers,
Wednesday between 2-3 PM EST we will commit following changes to these
repositories:
The branches:
produtil repo - master is already updated
NEMSfv3gfs repo - pull from gaea-f2 branch into master
NEMS repo - pull from gaea-f2 branch into master
The changes:
1. NEMS: location of rocoto is updated to 1.3.0rc2 in nemspara or
hurdata areas. Note that rc2 and rc3 are the same; the increment in
some places was due to a miscommunication.
2. All three repos: Everything on GAEA is switched to /lustre/f2 except
libraries which are expected to be ready soon.
3. NEMSfv3gfs: The workaround for Theia SLURM using 12 cores per node
has been removed now that the bug in SLURM is fixed.
4. NEMS: Documentation in NEMS/doc is updated
5. NEMSfv3gfs: The wallclock limit on WCOSS Phase 2 is increased from 40
minutes to 50 so that the expensive physics schemes will stop reaching
their wallclock limit.
6. NEMSfv3gfs: On Jet, "module load slurm" is replaced by "module load
slurm/default"
-Ratko
--
Ratko Vasic
Meteorologist
301-683-3814
National Oceanic and Atmospheric Administration
NCEP/EMC, Room 2791
NCWCP W/NP2
5830 University Research Court
College Park, MD 20740-3818
_______________________________________________
Ncep.list.nems.announce mailing list
Ncep.list.nems.announce at lstsrv.ncep.noaa.gov
https://www.lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.nems.announce
_______________________________________________
Ncep.list.nems.announce mailing list
Ncep.list.nems.announce at lstsrv.ncep.noaa.gov
https://www.lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.nems.announce
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www.lstsrv.ncep.noaa.gov/pipermail/ncep.list.nems.announce/attachments/20190304/01b8da29/attachment-0001.html
More information about the Ncep.list.nems.announce
mailing list