- 08 Nov, 2016 3 commits
-
-
Morris Jette authored
Different paths were used in various locations. All changed to use the /var/spool directory.
-
Morris Jette authored
-
Morris Jette authored
If a job is started by the main scheduling logic and requeued while the backfill scheduler has locks released, that can result in an invalid data structure in select/cons_res. Namely, the backfill scheduler's attempt to start the job would clear the job resources node_bitmap. That leaves a NULL pointer in the select/cons_res plugin generating an abort. (That pointer is needed to clean up the job allocation records when the Epilog or Cray Node Health Check, NHC, are complete and the resources become available for another job. bug 3230
-
- 07 Nov, 2016 6 commits
-
-
Danny Auble authored
task and can get quite verbose on a KNL node.
-
Danny Auble authored
warning fix.
-
Danny Auble authored
if the pointer is valid
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
Backup slurmctld will now 1. Not abort due to NULL pointer (needed to move code around on restart) 2. Recover KNL MCDRAM and NUMA modes from state save files if capmc and cnselect not available bug 3241
-
- 05 Nov, 2016 2 commits
-
-
Morris Jette authored
-
Morris Jette authored
cray/burst_buffer - Update "instance" parsing to match updated dw_wlm_cli output. bug 3222
-
- 04 Nov, 2016 22 commits
-
-
Morris Jette authored
No change to logic
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
cons_res.
-
Danny Auble authored
requests when they = 1. What we did was just increase the cpus_per_task to equal the amount of threads in the core|socket. This was done with an array to handle heterogeneous systems. Bug 2519
-
Danny Auble authored
Hopefully this will make alterations to this function easier as it doesn't now consider ntasks_per_*.
-
Brian Christiansen authored
-
Morris Jette authored
Expand the dw_wlm_cli script to include persistent and job-specific burst buffers. This script is used by burst_buffer/cray.
-
Morris Jette authored
Add "FreeSpace" information for each pool to the "scontrol show burstbuffer" output. Required changes to the burst_buffer_info_t data structure. bug 3222
-
Morris Jette authored
-
Morris Jette authored
This is a new field and the fix only applies to an emulated burst buffer configuration (i.e. dw_wlm_cli script made to look like a real DataWarp system)
-
Morris Jette authored
-
Morris Jette authored
Change error() to verbose(). New logic is needed to address this issue once we know how to determine the KNL MCDRAM size.
-
Morris Jette authored
-
Morris Jette authored
cray/burst_buffer - Preserve job ID and don't translate to job array ID after slurmctld restart. Prior logic would not set array_task_id to NO_VAL, so all job-buffer IDs would be reported in the form "JobID=0_0(123)" rather than "JobID=123"
-
Morris Jette authored
cray/busrt_buffer - Internally track both allocated and unusable space. The reported UsedSpace in a pool is now the allocated space (previously was unusable space). Base available space on whichever value leaves least free space. bug 3222
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
Previously disconnected from build system, and most code removed by commit 0b14a3a7 back on 15.08-pre1.
-
- 03 Nov, 2016 7 commits
-
-
Tim Wickberg authored
We don't build on Tru-64, and there are a lot more platform-dependent pieces of code within Slurm than we've indicated here.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
Remove stray Bull logo as well.
-
Tim Wickberg authored
Remove stray Bull logo as well.
-
Tim Wickberg authored
-
Tim Wickberg authored
They're full of dead links, and the plugins are deprecated (announced at SLUG16).
-