- 10 Sep, 2015 1 commit
-
-
Morris Jette authored
-
- 09 Sep, 2015 5 commits
-
-
Morris Jette authored
Don't trucate task ID information in "squeue --array/-r" output. Task ID info in sview also expanded to 64 characters (from ~16 chars).
-
Danny Auble authored
Fix for bad cut and paste. Looks like the original code was just a copy of the access parsing.
-
Danny Auble authored
slurmctld.
-
Morris Jette authored
Use more flexible mechnanism to find json installation.
-
Brian Christiansen authored
Fix srun from inheriting the SLURM_CPU_BIND and SLURM_MEM_BIND environment variables when running in an existing srun (e.g. an srun within an salloc). Bug 1888
-
- 08 Sep, 2015 3 commits
-
-
Danny Auble authored
-
jette authored
Improve job state reason string when required nodes not available. bug 1920
-
David Bigagli authored
-
- 07 Sep, 2015 1 commit
-
-
David Bigagli authored
-
- 05 Sep, 2015 1 commit
-
-
Danny Auble authored
-
- 04 Sep, 2015 1 commit
-
-
Morris Jette authored
Add sbatch "--wait" option that waits for job completion before exiting. Exit code will match that of spawned job. but 1685
-
- 03 Sep, 2015 1 commit
-
-
Morris Jette authored
Remove our work around for a bug in the Cray API that has recently been fixed. Improve error message for root jobs trying to use DW.
-
- 02 Sep, 2015 6 commits
-
-
Morris Jette authored
This change is already logged in the NEWS for v14.11.10
-
Danny Auble authored
over the limit with grpwall on a QOS or association.
-
Danny Auble authored
limit.
-
Danny Auble authored
job QOS.
-
Morris Jette authored
Previous logic would set the avail_node_bitmap when a node was powered down, even if the initial state was DOWN or DRAINED. This made the node available for allocation to a job, which we don't want until the DOWN or DRAIN state is cleared. bug 1893
-
Morris Jette authored
This reverts commits 7660da9e 5c386455 and f6c5302b
-
- 01 Sep, 2015 9 commits
-
-
Brian Christiansen authored
Bug 1741
-
Danny Auble authored
-
David Bigagli authored
-
Danny Auble authored
-
Nathan Yee authored
-
David Bigagli authored
-
Danny Auble authored
-
Brian Christiansen authored
Bug 1896
-
Danny Auble authored
description.
-
- 31 Aug, 2015 2 commits
-
-
Brian Christiansen authored
-
Brian Christiansen authored
Bug 1867
-
- 28 Aug, 2015 1 commit
-
-
Morris Jette authored
This problem is reproducible by launching a job then killing the slurmstepd process. Under those conditions, requeue the job if possible (i.e. batch job with requeue option/configuration). This patch also improves the slurmctld logging when this happens. bug 1889
-
- 27 Aug, 2015 5 commits
-
-
Morris Jette authored
Correct RebootProgram logic when executed outside of a maintenance reservation. Previous logic would mark the node up upon response to the reboot RPC (from slurmctld to slurmc) and when the node actually rebooted, flag that as an unexpected reboot. This new logic checks the node's up time to not mark the compute node as being usable until the reboot actually takes place. but 1866
-
Morris Jette authored
Add environment variables SLURM_ARRAY_TASK_MAX, SLURM_ARRAY_TASK_MIN, SLURM_ARRAY_TASK_STEP for job arrays. bug 1600
-
Danny Auble authored
to free it otherwise.
-
Brian Christiansen authored
Bug 1826
-
Danny Auble authored
association manager.
-
- 26 Aug, 2015 4 commits
-
-
Danny Auble authored
-
Morris Jette authored
Prevent job array task ID from being reported as NO_VAL if last task in the array gets requeued. The problem is that when that task starts, the task bitmap entry for it stays set, but the task counter gets decremented. If that job then gets requeued, under some conditions a failure to schedule it results in the array_task_id in the job record getting set to NO_VAL. Then when building the job info to report for squeue/scontrol, the string showing the pending task ID's is not rebuilt due to that counter being zero. All indications are that the job runs fine, only the information reported to squeue/scontrol is wrong. bug 1790
-
Danny Auble authored
-
Danny Auble authored
-