- 21 Mar, 2014 6 commits
-
-
Danny Auble authored
-
Nathan Yee authored
-
Morris Jette authored
If srun invoked with the --multi-prog option, but no task count, then use the task count provided in the MPMD configuration file.
-
Danny Auble authored
suite.
-
Morris Jette authored
-
David Bigagli authored
-
- 20 Mar, 2014 6 commits
-
-
Danny Auble authored
-
Nathan Yee authored
-
Hongjia Cao authored
performance.
-
Danny Auble authored
Conflicts: src/slurmctld/job_scheduler.c
-
Danny Auble authored
than you really have.
-
Danny Auble authored
doesn't get chopped off.
-
- 19 Mar, 2014 2 commits
-
-
David Bigagli authored
-
Gennaro Oliva authored
a minus sign for options was intended.
-
- 18 Mar, 2014 7 commits
-
-
Danny Auble authored
-
David Bigagli authored
-
Tommi T authored
correctly.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
Some of these were resulting in the state of a job not being updated correctly to tools like sview.
-
Danny Auble authored
in waiting reason ReqNodeNotAvail.
-
- 17 Mar, 2014 8 commits
-
-
David Bigagli authored
-
David Bigagli authored
-
David Bigagli authored
protocol version to be SLURM_2_5_PROTOCOL_VERSION which is the minimum supported version.
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
fix for bug 647
-
Morris Jette authored
See bug 644
-
- 16 Mar, 2014 4 commits
-
-
Morris Jette authored
Previously if the sbatch --export=NONE option was used then several Slurm environment variables were not propagated from the sbatch command (SLURM_SUBMIT_DIR, SLURM_SUBMIT_HOST, SLURM_JOB_NAME, etc.)
-
Morris Jette authored
Scheduler enhancements for reservations: When a job needs to run in reservation, but can not due to busy resources, then do not block all jobs in that partition from being scheduled, but only the jobs in that reservation.
-
Morris Jette authored
This corrects logic in commit fae55cbe to properly support front-end system configurations
-
Morris Jette authored
Reset a node's CpuLoad value at least once each SlurmdTimeout seconds. Previously the value would not be reset unless communications with the slurmd did not happen for at least 1/3 of the SlurmdTimeout value. That means nodes that were actively running and terminating jobs would not get the CpuLoad value reset in a timely fashion. Added a CpuLoad reset timer to prevent this.
-
- 15 Mar, 2014 7 commits
-
-
Morris Jette authored
Add logic to sleep and retry if slurm.conf can't be read. Without this, the slurmd daemons may die and when the SlurmdTimeout is reached, the nodes will be marked DOWN and their jobs will be killed. In the long term, it would be good to exit only if the read files on program startup, and the daemons keep running with old configuration on reconfiguration, but I don't have time to do that work now.
-
Morris Jette authored
The function slurm_api_set_conf_file() is never referenced. Remove it. No change in logic.
-
Morris Jette authored
No change in logic. Just remove redundant function return code.
-
Morris Jette authored
Fix invalid memory reference if script returns error message for user. Previous code failed to set static variable to NULL resulting in xfree of memory previously freed elsewhere.
-
Morris Jette authored
Add requeuehold command to information generated by scontrol's help command bug 642
-
Morris Jette authored
Conflicts: META NEWS
-
Morris Jette authored
-