- 01 Dec, 2010 2 commits
-
-
-
Moe Jette authored
than midplane counts.
-
- 30 Nov, 2010 5 commits
-
-
Danny Auble authored
sacctmgr now has smarts to figure out if a qos is a default qos when modifing a user/acct or removing a qos.
-
Danny Auble authored
Made openssl not be required to build RPMs, it is not required anymore since munge is the default crypto plugin.
-
Moe Jette authored
salloc, sbatch and srun man pages. These options are defunct, Patch from Rod Schultz, Bull.
-
Danny Auble authored
-
Moe Jette authored
managed. Substantially improves performance for large numbers of tasks. Adds support for SLURM_PMI_KVS_NO_DUP_KEYS environment variable. Patch from Hongjia Cao, NUDT.
-
- 29 Nov, 2010 5 commits
-
-
Moe Jette authored
cancels a job while the node is not responding but slurmctld has not yet the node down. Patch from Hongjia Cao, NUDT.
-
Moe Jette authored
malloc function is interupted and called again. The malloc function is thread safe, but not reentrant.
-
Danny Auble authored
-
Danny Auble authored
-
Moe Jette authored
xmalloc, which is not reentrant. original patch made in revision 21605
-
- 24 Nov, 2010 3 commits
-
-
Moe Jette authored
state be NODE_FAIL rather than CANCELLED.
-
Moe Jette authored
tasks.
-
Danny Auble authored
-
- 23 Nov, 2010 7 commits
-
-
Moe Jette authored
a job and let the job's owner release it. The scontrol command of "hold <job_id>" when executed by a SLURM administrator can only be released by a SLURM administrator and not the job owner.
-
Danny Auble authored
Apply rlimits right before execing the users task so to lower the risk of the task exiting because the slurmstepd ran over a limit (log file size, etc...)
-
Moe Jette authored
-
Moe Jette authored
is exceeded.
-
Moe Jette authored
slurmctld is reconfigured while a job is in completing state.
-
-
Danny Auble authored
Fix for possible deadlock in the slurmstepd when cancelling a job that is also writing a large amount of data to stderr.
-
- 22 Nov, 2010 4 commits
-
-
Moe Jette authored
CEA.
-
Moe Jette authored
queue of jobs rather than restarting at the top if there are no changes in job, node, or partition state between runs. Patch from Hongjia Cao, NUDT.
-
Moe Jette authored
(5 minutes) instead of one minute.
-
Moe Jette authored
terminated on. Patch from Hongjia Cao, NUDT.
-
- 12 Nov, 2010 4 commits
-
-
Danny Auble authored
Fix for problems when adding a user for the first time to a new cluster with a 2.1 sacctmgr without specifying a default account.
-
Moe Jette authored
rather than requeue it.
-
Moe Jette authored
-
Danny Auble authored
Add support for fair-share scheduling to be based upon resource use at the level of bank accounts and ignore use of individual users. Patch by Par Andersson, National Supercomputer Centre, Sweden.
-
- 11 Nov, 2010 2 commits
-
-
Moe Jette authored
-
Danny Auble authored
-
- 10 Nov, 2010 2 commits
-
-
Danny Auble authored
-
Moe Jette authored
--nodes option was ignored).
-
- 09 Nov, 2010 3 commits
-
-
Danny Auble authored
-
Moe Jette authored
a job was requeued and started no job steps.
-
Don Lipari authored
-
- 08 Nov, 2010 3 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Don Lipari authored
-