- 14 Feb, 2014 1 commit
-
-
Danny Auble authored
needed to forward a message the slurmd would core dump.
-
- 13 Feb, 2014 1 commit
-
-
David Bigagli authored
describing that jobs must be drained from cluster before deploying any checkpoint plugin.
-
- 12 Feb, 2014 2 commits
-
-
Morris Jette authored
Re-order existing code so that per-cpu memory limits will be enforce with both cpus_per_task and ntasks_per_node limits.
-
Morris Jette authored
Properly enforce a job's cpus-per-task option when a job's allocation is constrained on some nodes by the mem-per-cpu option. bug 590
-
- 10 Feb, 2014 2 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
- 09 Feb, 2014 1 commit
-
-
Moe Jette authored
-
- 08 Feb, 2014 3 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
This just shuts down the underlying select plugin used by select/cray in order for it to free all of its allocated memory
-
- 07 Feb, 2014 3 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
bug 586
-
- 05 Feb, 2014 8 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Dominik Bartkiewicz authored
Set GPU_DEVICE_ORDINAL environment variable.
-
Morris Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
- 04 Feb, 2014 5 commits
-
-
Morris Jette authored
-
Morris Jette authored
Previous logic would try to pick a specific node count and on a heterogeneous system, this would cause a problem. This change largely reverts commit a270417b
-
Morris Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
- 03 Feb, 2014 1 commit
-
-
Danny Auble authored
-
- 01 Feb, 2014 1 commit
-
-
Morris Jette authored
-
- 31 Jan, 2014 7 commits
-
-
David Bigagli authored
-
Danny Auble authored
starts and then a minor typo fix
-
Danny Auble authored
i.e. salloc -n32 doesn't request the number of nodes and with the previous code if this request used 4 nodes and only 1 was left in GrpNodes it would just run with no issue since we were checking things before we selected how many nodes it ran on. Now we check this afterwards so we always check the limits on how many nodes, cpus and how much memory is to be used.
-
Morris Jette authored
Fix step allocation when some CPUs are not available due to memory limits. This happens when one step is active and using memory that blocks the scheduling of another step on a portion of the CPUs needed. The new step is now delayed rather than aborting with "Requested node configuration is not available". bug 577
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
For Kelly ;)
-
- 30 Jan, 2014 3 commits
-
-
Morris Jette authored
-
David Bigagli authored
-
Morris Jette authored
No change in logic, just rename a variable for better clarity.
-
- 28 Jan, 2014 1 commit
-
-
Danny Auble authored
based on ionode count correctly on slurmctld restart.
-
- 24 Jan, 2014 1 commit
-
-
Danny Auble authored
would be set, but this is safer.
-