- 15 May, 2014 4 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
task/cgroup - Correct specialized core task binding with user supplied invalid CPU mask or map. Rather than generating an error and ignoring user specification. Mask user supplied map against available CPUs or bind to all available CPUs. In either case, log the invalid CPU map or mask. bug 782
-
Morris Jette authored
-
- 14 May, 2014 4 commits
-
-
Morris Jette authored
Conflicts: src/slurmctld/job_scheduler.c
-
Morris Jette authored
Run EpilogSlurmctld for a job is killed during slurmctld reconfiguration. bug 806
-
Morris Jette authored
-
Morris Jette authored
Only if ALL of their partitions are hidden will a job be hidden by default. bug 812
-
- 13 May, 2014 12 commits
-
-
Morris Jette authored
If a batch job launch request can not be built (the script file is missing, a credential can not be created, or the user does not exist on the selected compute node), then cancel the job in a graceful fashion. Previously, the bad RPC would be sent to the compute node and that node DRAINED. see bug 807
-
Morris Jette authored
-
Morris Jette authored
Correct SelectTypeParameters=CR_LLN with job selecition of specific nodes. Previous logic would in most instances allocate resources on all nodes to the job.
-
Morris Jette authored
Correct squeue's job node and CPU counts for requeued jobs. Previously, when a job was requeued, its CPU count reported was that of the previous execution. When combined with the --ntasks-per-node option, squeue would compute the expected node count. If the --exclusive option is also used, the node count reported by squeue could be off by a large margin (e.g. "sbatch --exclusive --ntasks-per-node=1 -N1 .." on requeue would use the number of CPUs on the allocated node to recompute the expected node count). bug 756
-
David Gloe authored
req.c: In function ‘_launch_complete_rm’: req.c:5372: error: array subscript is above array bounds req.c: In function ‘_launch_complete_add’: req.c:5328: error: array subscript is above array bounds The lines are if (job_id != active_job_id[j]) { after the for loops in those functions. If no match is found in the loop, j will be JOB_STATE_CNT and overflow the array by one.
-
Morris Jette authored
-
Danny Auble authored
jobacct_gather/cgroup.
-
Morris Jette authored
Support SLURM_CONF path which does not have "slurm.conf" as the file name. bug 803
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
For a nested batch job (within an salloc, run "sbatch --jobid=$SLURM_JOBID ..."), report the completing node rank as 0, rather than -1
-
Morris Jette authored
-
- 12 May, 2014 13 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
If a job has non-responding node, retry job step create rather than returning with DOWN node error. bug 734
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Puenlap Lee authored
Also correct related documentation
-
Morris Jette authored
-
Nathan Yee authored
Add force option to all file removals ("rm ..." to "rm -f ..."). bug 673
-
Morris Jette authored
-
Morris Jette authored
-
Hongjia Cao authored
Completing nodes is removed when calling _try_sched() for a job, which is the case in select_nodes(). If _try_sched() thinks the job can run now but select_nodes() returns ESLURM_NODES_BUSY, the backfill loop will be ended.
-
- 09 May, 2014 7 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Martin Perry authored
-
Morris Jette authored
Do not resume a job with specialized cores on a node running another job with specialized cores (only one can run at a time). bug 792
-