- 03 Dec, 2015 5 commits
-
-
Morris Jette authored
Cray job NHC delayed until after burst buffer released and epilog completes on all allocated nodes. bugs 2099 and 2192
-
Morris Jette authored
Release a job's allocated licenses only after epilog runs on all nodes rather than at start of termination process. bug 2192
-
Morris Jette authored
sched/backfill - Delay backfill scheduler for completing jobs only if CompleteWait configuration parameter is set (make code match documentation).
-
David Bigagli authored
-
Tim Wickberg authored
-
- 02 Dec, 2015 1 commit
-
-
Josko Plazonic authored
Bug 2030
-
- 01 Dec, 2015 4 commits
-
-
Danny Auble authored
correctly when PriorityFlags=CALCULATE_RUNNING is set. Previously the slurmctld could seg fault if the tres_alloc_str is NULL.
-
Morris Jette authored
Prevent slurmdbd divide by zero if no associations defined at rollup time.
-
Danny Auble authored
else behind.
-
David Bigagli authored
-
- 30 Nov, 2015 5 commits
-
-
Danny Auble authored
uint64_t.
-
Danny Auble authored
-
Danny Auble authored
as all the associations from the database will be lower case.
-
Thomas Cadeau authored
Correct job task count calcuation if only node count and ntasks-per-node options supplied. bug 2196
-
David Bigagli authored
-
- 26 Nov, 2015 1 commit
-
-
jette authored
sched/backfill: If max_rpc_cnt is configured and the backlog of RPCs has not cleared after yielding locks, then continue to sleep.
-
- 25 Nov, 2015 1 commit
-
-
Danny Auble authored
requesting any specific association.
-
- 23 Nov, 2015 1 commit
-
-
Danny Auble authored
-
- 19 Nov, 2015 3 commits
-
-
Morris Jette authored
BurstBuffer/cray: Fix job record purging if cancelled from pending state. The problem can occur when the a burst buffer record was created for the job in the plugin data structure, but no burst buffers were actually allocated for it. bug 2165
-
David Bigagli authored
-
Morris Jette authored
BurstBuffer/cray: Enable clearing of burst buffer string on completed job as a means of recovering from a failure mode. Format is "scontrol update jobid=### burstbuffer=". partial resolution of bug 2165
-
- 18 Nov, 2015 4 commits
-
-
Morris Jette authored
bug 2028
-
Morris Jette authored
BurstBuffer/cray: Add logic to terminate dw_wlm_cli child processes at shutdown. bug 2166
-
Morris Jette authored
Previous logic required the buffer name to work bug 2167
-
Morris Jette authored
Added srun option of --bcast to move executable file to compute nodes
-
- 17 Nov, 2015 2 commits
-
-
Morris Jette authored
burst_buffer/cray: Support file staging when job lacks job-specific buffer (i.e. only persistent burst buffers). bug 2113
-
David Bigagli authored
-
- 16 Nov, 2015 2 commits
-
-
Morris Jette authored
bug 2143
-
Morris Jette authored
Backfill scheduler: Test association and QOS node limits before reserving resources for pending job. bug 2129
-
- 13 Nov, 2015 8 commits
-
-
Brian Christiansen authored
-
Brian Christiansen authored
Prevents the following sequence from causing a segfault: $ scontrol create partitionname=stuff nodes=ALL $ sbatch --wrap="hostname" -o/dev/null -p stuff Submitted batch job 1047468 $ scontrol delete partitionname=stuff $ scontrol update jobid=1047468 partition=stuff
-
Danny Auble authored
-
Danny Auble authored
tree.
-
Danny Auble authored
-
Brian Christiansen authored
Bug 2006
-
Danny Auble authored
step.
-
Morris Jette authored
-
- 12 Nov, 2015 3 commits
-
-
Mark Roberts authored
-
Morris Jette authored
-
Morris Jette authored
Previously only supported by SlurmUser and root.
-