- 15 Nov, 2013 1 commit
-
-
Morris Jette authored
Add ability to clear a node's DRAIN flag using scontrol or sview by setting it's state to "UNDRAIN". The node's base state (e.g. "DOWN" or "IDLE") will not be changed. bug 514
-
- 13 Nov, 2013 3 commits
-
-
Morris Jette authored
This might have worked fine for core reservations or when there are sufficient idle nodes to use, the the select_g_resv_test() function clears the node bitmap for nodes that it can not use and the reservation create logic did not restore that bitmap after a failed resource selection attempt. This logic restores the node bitmap on a failed call to select_g_resv_test() so we can add nodes to the bitmap of available nodes rather than having it repeatedly cleared. The logic also adds some performance enhancements that I will add to in the next commit.
-
David Bigagli authored
-
Morris Jette authored
This fixes a bug where a system is enforcing memory limits and the job already has a step running on some of the nodes then tries to start another step using some of those nodes. For example wwith DefMemPerNode configured and the select plugin enforcing memory limits, try: salloc -N2 bash $ srun -N1 sleep 10& $ srun -N2 hostname Without this patch, the second srun would fail instead of pend.
-
- 09 Nov, 2013 1 commit
-
-
David Bigagli authored
-
- 08 Nov, 2013 1 commit
-
-
Danny Auble authored
-
- 05 Nov, 2013 1 commit
-
-
Morris Jette authored
Correction to hostlist parsing bug introduced in v2.6.4 for hostlists with more than one numeric range in brackets (e.g. rack[0-3]_blade[0-63]"). bug505
-
- 04 Nov, 2013 6 commits
-
-
Morris Jette authored
Just a start for now
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Added -S/--core-spec option to salloc, sbatch and srun commands to reserve specialized cores for system use. Modify sview and scontrol to set/get core_spec struct job_info / slurm_job_info_t: Added core_spec struct job_descriptorjob_desc_msg_t: Added core_spec
-
David Bigagli authored
-
- 03 Nov, 2013 1 commit
-
-
jette authored
The system really can not handle larger job arrays without adding a job array data structure, but this puts some of the infrastructure in place now.
-
- 02 Nov, 2013 1 commit
-
-
Martins Innus authored
to offer a greater priority range based upon utilization.
-
- 01 Nov, 2013 3 commits
-
-
Morris Jette authored
Add argument to priority plugin's priority_p_reconfig function to note when the association and QOS used_cpu_run_secs field has been reset. Without this flag, we remove time on "scontrol setdebug" or "scontrol setdebugflag" that can result in used_cpu_run_secs going negative or otherwise get bad values. Correction to logic added in commit 6d793189 bug 423
-
Morris Jette authored
Add argument to priority plugin's priority_p_reconfig function to note when the association and QOS used_cpu_run_secs field has been reset. Without this flag, we remove time on "scontrol setdebug" or "scontrol setdebugflag" that can result in used_cpu_run_secs going negative or otherwise get bad values. Correction to logic added in commit 6d793189 bug 423
-
Morris Jette authored
Fix to work with change logic introduced in Slurm version 2.6.3 scheduling logic which prevented Maui/Moab from starting jobs.
-
- 31 Oct, 2013 1 commit
-
-
David Bigagli authored
-
- 30 Oct, 2013 3 commits
-
-
Matthias Jurenz authored
-
Morris Jette authored
Add support for -W block=true (wait for job completion) Clear PBS_NODEFILE environment variable Credit to NCSC
-
Morris Jette authored
-
- 29 Oct, 2013 3 commits
-
-
David Bigagli authored
-
Morris Jette authored
Add support for -W block=true (wait for job completion) Clear PBS_NODEFILE environment variable Credit to NCSC
-
Morris Jette authored
-
- 28 Oct, 2013 5 commits
-
-
Morris Jette authored
Add support for dependencies of job array elements (e.g. "sbatch --depend=afterok:123_4 ..."). This does not support depenendencies of ALL job array elements, only individual job array elements.
-
Danny Auble authored
preforms the allocation, regression in 2.6.3.
-
Morris Jette authored
-
Danny Auble authored
-
Morris Jette authored
-
- 25 Oct, 2013 4 commits
-
-
Morris Jette authored
Previously the SlurmdDebug value in slurm.conf was ignored if the previous value was not 3/init
-
Morris Jette authored
Effect is minimal without multiple partitions and larger system sizes. With 40 partitions and about 600 nodes each, time goes from about 13 secs to 4 secs).
-
Morris Jette authored
This avoids building hostlist information with NodeHostName and NodeAddr information unless explisitly requested and can improve performance for the default mode of operation by about 65%.
-
Morris Jette authored
Correct sbatch documentation and job_submit/pbs plugin "%j" is job ID, not "%J" (which is job_id.step_id).
-
- 24 Oct, 2013 1 commit
-
-
Morris Jette authored
Without this change a job with a reason of WAIT_PART_DOWN, WAIT_PART_INACTIVE, WAIT_PART_NODE_LIMIT, WAIT_PART_TIME_LIMIT, or WAIT_QOS_THRES would not be cleared when that reason no longer applied.
-
- 23 Oct, 2013 4 commits
-
-
Morris Jette authored
Add cgroup create retry logic in case one step is starting at the same time as another step is ending and the logic to create and delete cgroups overlaps. bug 447
-
Morris Jette authored
If a node has GRES and multiple threads per core the select/cons_res plugin can get stuck in an infinite loop. See bug 475 Contributed by: PREVOST Ludovic NEC HPC Europe
-
Morris Jette authored
-
Morris Jette authored
Previously a node failure would always requeue the job
-
- 22 Oct, 2013 1 commit
-
-
Morris Jette authored
Add cgroup create retry logic in case one step is starting at the same time as another step is ending and the logic to create and delete cgroups overlaps. bug 447
-