- 21 Sep, 2016 1 commit
-
-
Morris Jette authored
Don't log error for job end_time being zero if node health check is still running. bug 3053
-
- 17 Sep, 2016 2 commits
-
-
Danny Auble authored
the same logic that was found in the slurmdbd. Now both functionalities share the same code. This was done with the merge right before this commit.
-
Morris Jette authored
Restore ability to manually power down nodes, broken in 15.08.12 in commit b4904661 The patch introduced in commit b4904661 (not powering down dead node) has a bad side effect. Adding the "(node_ptr->last_idle != 0)" condition prevents from powering down nodes with the following command: scontrol update nodename=nX state=power_down because the state update function relies on zeroing the "last_idle" variable when a power_down is requested (see src/slurmctld/node_mgr.c, line 1589). Reverting this commit should solve the problem...but I let you decide... Didier GAZEN
-
- 16 Sep, 2016 1 commit
-
-
Morris Jette authored
node_features/knl_cray: If a node is rebooted outside of Slurm's direction, update it's active features with current MCDRAM and NUMA mode information. bug 3071
-
- 15 Sep, 2016 3 commits
-
-
Tim Wickberg authored
Will be appended to usernames if --mail-user is not explicitly set for the job and email notifications are requested. Bug 3089.
-
Morris Jette authored
Fix race condition that could result in MCDRAM state information coming from capmc rather than cnselect (used state for next boot rather than latest boot). bug 3080
-
Nicolas Joly authored
-
- 14 Sep, 2016 2 commits
-
-
Alejandro Sanchez authored
No functional change, just silencing the warning message in this instance. Bug 3079.
-
Alejandro Sanchez authored
Bug 3073.
-
- 12 Sep, 2016 1 commit
-
-
Tim Wickberg authored
-
- 09 Sep, 2016 3 commits
-
-
Morris Jette authored
Modify srun task completion handling to only build the task/node string for logging purposes if it is needed. Modified for performance purposes. bug 3044
-
Tim Wickberg authored
This reverts commit 1ec2a4ae.
-
Alejandro Sanchez authored
Bug 3063.
-
- 08 Sep, 2016 2 commits
-
-
Brian Christiansen authored
In scontrol show nodes.
-
Morris Jette authored
Restructure srun command locking for task_exit processing logic for improved parallelism. This change decreases the amount of time consumed by serial logic by 2 orders of magnitude. bug 3044
-
- 07 Sep, 2016 2 commits
-
-
Morris Jette authored
Preserve node "RESERVATION" state when one of multiple overlapping reservations ends. Previous logic would clear the node's RESERVATION state flag when any one of the reservations on the node ended rather than keeping the node in RESERVATION state until the last reservation ended. bug 3057
-
Morris Jette authored
Handle case when slurmctld daemon restart while compute node reboot in progress. Return node to service rather than setting DOWN. bug 3042
-
- 06 Sep, 2016 3 commits
-
-
Morris Jette authored
Add salloc_wait_nodes option to the SchedulerParameters parameter in the slurm.conf file controlling when the salloc command returns in relation to when nodes are ready for use (i.e. booted). bug 3043
-
Gennaro Oliva authored
bug 3055
-
Gennaro Oliva authored
bug 3054
-
- 02 Sep, 2016 1 commit
-
-
Danny Auble authored
reservations.
-
- 01 Sep, 2016 3 commits
-
-
Morris Jette authored
sched/backfill - Check that a user's QOS is allowed to use a partition before trying to schedule resources on that partition for the job. bug 3039
-
Morris Jette authored
bug 3037
-
Morris Jette authored
bug 3035 and 3009
-
- 31 Aug, 2016 1 commit
-
-
Brian Christiansen authored
-
- 30 Aug, 2016 2 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
Otherwise blade_cnt is potentially greater than bit_size(jobinfo->blade_map) which leads to an assertion failure. Bug 3033.
-
- 27 Aug, 2016 2 commits
-
-
Artem Polyakov authored
with hwloc.
-
Morris Jette authored
This patch has two parts: 1. When a job is intially submitted, the Slurm was failing to set an initial reason for the job not starting. 2. After a job was submitted, it was sometimes failing to reset the job's reason. It was also failing to reset the "last_job_update" time, so something like "squeue -i1" would not get the new reason. bug 3025
-
- 26 Aug, 2016 2 commits
-
-
Alejandro Sanchez authored
Fix multipart srun submission with EnforcePartLimits=NO and job violating the partition limits. bug 3025
-
Alejandro Sanchez authored
bug 3011
-
- 25 Aug, 2016 1 commit
-
-
Morris Jette authored
If all GRES were not defined on all nodes OR if a regular expression was used for a GRES file configuration (e.g. in gres.conf "Type=gpu Files=/dev/nvidia[0-4]"), then memory corruption was likely. The logic has been bad since its inception several years ago.
-
- 24 Aug, 2016 1 commit
-
-
Joseph Mingrone authored
POLLRDHUP does not exist on BSD, define to POLLHUP as done elsewhere.
-
- 23 Aug, 2016 1 commit
-
-
David Gloe authored
The attached patch switches to a more reliable method of detecting service nodes, using xtcli status. In addition, it switches to the print function to be better compatible with python 3.
-
- 22 Aug, 2016 5 commits
-
-
Boris Karasev authored
-
Boris Karasev authored
To ease the distribution process, plugin names will be automatically adjusted to identify the version of API that it can support, ie: pmix_v1 and pmix_v2. This provides the ability for distro's to create separate non-conflicting packages for each API generation. Bug 2986
-
Morris Jette authored
Add job burst_buffer_state and delay_boot fields to scontrol and squeue output. Also add ability to modify delay_boot from scontrol.
-
Morris Jette authored
Add salloc, sbatch and srun option of --delay-boot=<time>, which will temporarily delay booting nodes into the desired state for a job in the hope of using nodes already
-
Morris Jette authored
Add job AdminComment field which can only be set by a Slurm administrator.
-
- 21 Aug, 2016 1 commit
-
-
Morris Jette authored
Add event trigger on burst buffer errors (see strigger man page, --burst_buffer option).
-