|
|
@1468
|
6 years |
mafoipsl |
On curie : set output files for all jobs.
On irene :
- set output files …
|
|
|
@1460
|
6 years |
jgipsl |
Change in headers at irene : removed temporary settings -E …
|
|
|
@1448
|
6 years |
jgipsl |
Change in headers at irene : temporary remove option -m and use a …
|
|
|
@1437
|
6 years |
jgipsl |
For Irene:
* ins_job now ask for project id. gch0316 for the grand …
|
|
|
@1409
|
7 years |
sdipsl |
- #337 and #338 source the appropriate .env_intel depending on compiler …
|
|
|
@1356
|
8 years |
sdipsl |
- comments for TaskType?
|
|
|
@1326
|
8 years |
sdipsl |
- Bugfix. IPSLCM6.0.2 ORCA1 monitoring was fully broken.
|
|
|
@1290
|
9 years |
sdipsl |
- broadcast postProcessingStopLevel value to every jobs. Default value is …
|
|
|
@1206
|
9 years |
sdipsl |
- Remove IGCM_debug_Verif_Exit_Post. Only IGCM_debug_Verif_Exit will …
|
|
|
@1201
|
9 years |
sdipsl |
- add jobWarningDelay in seconds to supervision messages see #253 …
|
|
|
@1198
|
9 years |
sdipsl |
- Supervize post-processing jobs see #252
- Homogenize post-processing …
|
|
|
@1192
|
9 years |
sdipsl |
Bugfix caused by r1189 side effect. On Curie job begins their execution in …
|
|
|
@1184
|
10 years |
sdipsl |
Add ciclad to the system. Episode II
|
|
|
@1167
|
10 years |
sdipsl |
run rebuild on standard node using iRAM high speed disk partition
|
|
|
@1154
|
10 years |
sdipsl |
- Add a pattern for node type. See #241
|
|
|
@1144
|
10 years |
acosce |
Correct comment on rebuild header
|
|
|
@1117
|
10 years |
sdipsl |
- Need to have access to IGCM_config_CommonConfiguration to know where to …
|
|
|
@1112
|
10 years |
sdipsl |
- clean up environment variable list flowing through jobs
- Some addons to …
|
|
|
@1070
|
10 years |
sdipsl |
- propagate supervision down to post-processing. See #210
|
|
|
@1059
|
10 years |
sdipsl |
- Attach the licence when it is not the case
|
|
|
@1031
|
10 years |
sdipsl |
See #187 RebuildFromArchive is deprecated
|
|
|
@1029
|
10 years |
sdipsl |
See #198: Add ssd header examples
|
|
|
@929
|
11 years |
sdipsl |
machine cleanup
whitespace cleanup
|
|
|
@928
|
11 years |
sdipsl |
clean-up
|
|
|
@906
|
11 years |
sdipsl |
- cleanup
- whitespace-cleanup
|
|
|
@880
|
11 years |
mafoipsl |
On curie, define time limit for post-processing jobs : rebuild and …
|
|
|
@848
|
11 years |
sdipsl |
- Use sequential post-processing class
|
|
|
@837
|
12 years |
labetoulle |
On Curie, ins_job sets up project ID. See #98.
Add
#-Q- curie #MSUB -A …
|
|
|
@832
|
12 years |
sdipsl |
- Trivial changes to accomodate new_libIGCM function defined here …
|
|
|
@785
|
12 years |
labetoulle |
Cosmetics :
- Convert tabs into space for better lisibility with different …
|
|
|
@770
|
12 years |
sdipsl |
Adapat headers for ada.
Clean up old headers (platine, cesium, brodie, …
|
|
|
@758
|
12 years |
mafoipsl |
On curie, delete #MSUB -E '--no-requeue' option. The system will requeue …
|
|
|
@741
|
12 years |
sdipsl |
- use xlarge class on curie
- comment header tweaking for ensemble due to …
|
|
|
@712
|
12 years |
aclsce |
- Added TaskType? variable in jobs to specify the type of the job …
|
|
|
@704
|
12 years |
jgipsl |
=> Added "#MSUB -Q normal" in headings for jobs at curie : When submitting …
|
|
|
@681
|
12 years |
mafoipsl |
On curie : Computing job on -q standard ie thin nodes, post-processing …
|
|
|
@643
|
12 years |
mafoipsl |
Add norerun options for jobs on titane and curie.
|
|
|
@640
|
12 years |
mafoipsl |
Add norerun option on SX front-end. Transfer specific files ie config.card …
|
|
|
@626
|
13 years |
mafoipsl |
On curie use the appropriate job name.
|
|
|
@622
|
13 years |
sdipsl |
- Hopefully end of the merge story
|
|
|
@616
|
13 years |
sdipsl |
- Merge libIGCM_MPI_OpenMP branch within the trunk
- trunk will shortly be …
|
|
|
@590
|
13 years |
sdipsl |
- can extract needed restart from a tar (up to 20 minutes on titane)
- …
|
|
|
@554
|
13 years |
sdipsl |
- Mechanism in place to submit pack jobs.
- Add a PackFrequency? ; default …
|
|
|
@544
|
13 years |
sdipsl |
- Prepare libIGCM for TGCC migration. For now work for CCRT machines ONLY …
|
|
|
@504
|
13 years |
mafoipsl |
vargas and brodie forgotten. Sorry.
On ulam (front-end for vargas and …
|
|
|
@500
|
13 years |
mafoipsl |
On ulam, set elapsed time limit to use all classes : t1 for ATLAS, …
|
|
|
@478
|
13 years |
sdipsl |
- ACo as we do not use TMPDIR on scalar mercure we have to clean the tmp
- …
|
|
|
@469
|
13 years |
mafoipsl |
On titane we have to use a workaround to reserve a full node.
This is …
|
|
|
@451
|
13 years |
sdipsl |
- Rebuilds and patchs are done on titane's mono class
- Standard error and …
|
|
|
@436
|
13 years |
sdipsl |
Some more break points ; just in case…
|
|
|
@426
|
14 years |
mmaipsl |
Add login shell path for platine jobs.
|
|
|
@423
|
14 years |
sdipsl |
- Vargas BIG_DIR was not set like the others.
- Bug with …
|
|
|
@419
|
14 years |
omamce |
Olivier Marti
Change rebuild_from[Workdir|Archive].job
For StandAlone? …
|
|
|
@387
|
14 years |
sdipsl |
- function IGCM_sys_Get_Master on cesium will try 10 times to scp files …
|
|
|
@386
|
14 years |
mafoipsl |
Add ulam header to facilitate post-processing on ulam directly.
|
|
|
@380
|
14 years |
sdipsl |
- Add a IGCM_config_Check function to test coherency between …
|
|
|
@373
|
14 years |
sdipsl |
Remove libIGCM_sys_claude.ksh and libIGCM_sys_calculo.ksh …
|
|
|
@330
|
14 years |
mafoipsl |
Correction of directory name for clean up. OK now.
|
|
|
@328
|
14 years |
mafoipsl |
add local clean up.
|
|
|
@327
|
14 years |
mafoipsl |
Add variables required to post-process on computer with MASTER server. For …
|
|
|
@320
|
14 years |
aclsce |
Changed "rm" by "IGCM_sys_RshMaster rm" in the case of …
|
|
|
@316
|
14 years |
mafoipsl |
Use cesium for SX9 post processing. Add hostname to know where we run.
|
|
|
@315
|
14 years |
mafoipsl |
Add mail when job aborts in mercure front-end header. Correct SendMail? …
|
|
|
@314
|
14 years |
mafoipsl |
Add cesium header to facilitate post-processing on cesium directly.
|
|
|
@299
|
14 years |
mafoipsl |
Use mercure (scalaire) new front end for SX8 post-processing.
|
|
|
@298
|
14 years |
mmaipsl |
Change mercure scalar queues like that :
noms machines | …
|
|
|
@281
|
14 years |
mmaipsl |
Add obelix headers.
|
|
|
@258
|
14 years |
sdipsl |
- Do not specify scalaire on the command line, it's in the header
- …
|
|
|
@257
|
14 years |
sdipsl |
- Add mercure-new : libIGCM_sys_mercure.ksh
- SX9 use mercure-new for …
|
|
|
@236
|
15 years |
sdipsl |
- Slight modification in the way we read and interpret card for …
|
|
|
@229
|
15 years |
mmaipsl |
Littles modifications in StandAlone? mode for post-treatment scripts.
…
|
|
|
@222
|
15 years |
aclsce |
- Added titane (CCRT Xeon machine) as target machine :
for the moment, …
|
|
|
@220
|
15 years |
sdipsl |
Use ulam's largest class for potentially heavy jobs
|
|
|
@216
|
15 years |
sdipsl |
IDRIS bugfix, typo in rebuild_FromArchive, rebuild_FromArchive parameter …
|
|
|
@213
|
15 years |
sdipsl |
- In AA_create_ts and AA_create_se no need to try to patch raw files when …
|
|
|
@207
|
15 years |
aclsce |
Modified mask patch to keep unlimited time_counter dimension.
Added …
|
|
|
@206
|
15 years |
sdipsl |
- Add LMDZ4 39 levels detection, launch atlas accordingly.
- Try to …
|
|
|
@205
|
15 years |
mafoipsl |
Add variables used by IGCM_Patch_20091118_mask in AA_rebuild_fromWorkdir …
|
|
|
@203
|
15 years |
sdipsl |
- With cesium we manage the TMPDIR; reflect that fact where needed
- …
|
|
|
@202
|
15 years |
sdipsl |
- With cesium we manage the TMPDIR; reflect that fact in some job
- …
|
|
|
@196
|
15 years |
sdipsl |
- Some more step towards post processing from cesium
- Wait from CCRT …
|
|
|
@192
|
15 years |
sdipsl |
- Cesium will post-process SX9 simulation
- Big clean in libIGCM_sys
- …
|
|
|
@178
|
15 years |
cetlod |
Add RESOL_MBG (Marine BioGeochemistry?) for use with TOP component
|
|
|
@172
|
15 years |
mafoipsl |
Add sx9mercure header in AA_ jobs.
|
|
|
@159
|
15 years |
sdipsl |
- Bugfix with test regarding Patch files and empty option
- Bugfix with …
|
|
|
@157
|
15 years |
sdipsl |
- RebuildFromWorkdir? is back. Will use BIG_DIR as temp storage directory
…
|
|
|
@147
|
15 years |
sdipsl |
- Do the correct pop Stack
- ADD variable MASTER n list of argument
|
|
|
@135
|
15 years |
sdipsl |
- Reflect the fact that monitoring should not depend on resolution for …
|
|
|
@108
|
15 years |
sdipsl |
May be last touch before tag 1.3
- some cosmetics
- prepare different …
|