- 18 Apr, 2012 1 commit
-
-
Danny Auble authored
-
- 17 Apr, 2012 11 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
catch, and fix, the problem when the next job comes through to use it.
-
Danny Auble authored
-
Danny Auble authored
larger than midplane jobs.
-
Danny Auble authored
This makes it easier to handle failover, if a slurmctld keeps failing (slurmdbd gets shutdown or something) this way we don't continually send a message over and over again in this senario, at least it drops down how many messages we send over and over again.
-
Danny Auble authored
-
Danny Auble authored
-
Bjørn-Helge Mevik authored
Add support for new SchedulerParameters of bf_max_job_user, maximum number of jobs to attempt backfilling per user. Work by Bjørn-Helge Mevik, University of Oslo.
-
Morris Jette authored
-
Morris Jette authored
Fix sched/wiki2 to support job account name, gres, partition name, wckey, or working directory that contains "#" (a job record separator). Without this patch, the parsing will probably stop once reaching the "#".
-
- 12 Apr, 2012 5 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
Conflicts: src/slurmd/slurmd/slurmd.c
-
Danny Auble authored
and cons_res or gang scheduling. The slurm.conf is always used for a node configuration in this case, and it ignores what the actual hardware is because in this situation the slurmctld needs to make a bunch of bitmaps before the slurmd's register.
-
Danny Auble authored
-
- 11 Apr, 2012 1 commit
-
-
Morris Jette authored
-
- 10 Apr, 2012 11 commits
-
-
Morris Jette authored
-
Danny Auble authored
and time limit where it was previously set by an admin.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
slurmdbd accounting and running large amounts of jobs (>50 sec). Job information could be corrupted before it had a chance to reach the DBD.
-
Danny Auble authored
qos.
-
Morris Jette authored
-
Morris Jette authored
-
jette authored
-
jette authored
Will set new scheduler timeout if MessageTimeout changes due to re-reading slurm.conf
-
jette authored
-
- 09 Apr, 2012 2 commits
-
-
Danny Auble authored
tasks than possible without overcommit the request would be allowed on more nodes than requested.
-
Morris Jette authored
-
- 07 Apr, 2012 1 commit
-
-
Morris Jette authored
These changes reduce the overhead of some functions for improved slurmctld throughput. Overhead of slurm_preempt_init() goes from 20% of CPU time to NIL and if preemption is disabled then slurm_job_preempt_check() goes from 12% of CPU time to NIL.
-
- 06 Apr, 2012 6 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Danny Auble authored
event comes in through the realtime interface.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
- 05 Apr, 2012 2 commits
-
-
Morris Jette authored
-
Don Lipari authored
While safeguards are in place to prevent unauthorized users from extending the TimeLimit of their running jobs, there were no such restrictions for extending the EndTime. This patch adds the same constraints to modifying EndTime that currently exists for modifying TimeLimit.
-