- 18 Feb, 2016 1 commit
-
-
Morris Jette authored
Fix for when a feature is available, but not active on any node.
-
- 17 Feb, 2016 14 commits
-
-
Morris Jette authored
Enforce a job's timelimit based upon when it actually beings execution, after node reboot completes. However the job will be changed for resource allocation from the start of allocation, include boot time.
-
Danny Auble authored
# Conflicts: # src/slurmctld/node_mgr.c
-
Morris Jette authored
-
Morris Jette authored
Move logic that re-calculates a job's end time based upon delayed launch into a separate function.
-
Tim Wickberg authored
Please submit patches through our bugzilla instance. (This would also show up on the issues page if we hadn't disabled it.)
-
Danny Auble authored
a parsing option. Dynamically set columns widths based on largest number in column.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
Previous logic was failing if feature name not found
-
Morris Jette authored
Add PID to both the slumctld code and cray capmc_suspend/resume programs
-
Morris Jette authored
-
Morris Jette authored
Previous logic would cause slurmctld shutdown to wait for completion of all power save (suspend and resume) programs according to their time limits, which could be huge. The new logic waits up to 10 second then orphans the processes.
-
Morris Jette authored
-
- 16 Feb, 2016 15 commits
-
-
Morris Jette authored
The parsing of the configuration parameter failed with a prefix of "node_features" due to vestigial logic for a plugin type of "knl".
-
Morris Jette authored
Was trying to set NUMA mode as MCNUMA and vise-versa Also change capmc to specify mode before nids, which seems more robust.
-
Tim Wickberg authored
-
Morris Jette authored
the "fputs" function was aborting, trying to write a NULL string pointer. Also found the log message was printing the configured and read names in the wrong order.
-
Morris Jette authored
-
Tim Wickberg authored
abort() rather than continue if pthread_mutex_ calls fail. better to die early rather than continue on and risk corruption. mirrors the (now removed) macro definitions from cbuf/hostlist/list.
-
Danny Auble authored
-
Danny Auble authored
accounting.
-
Danny Auble authored
AccountUtilizationByUser report a parent could potentially (correctly) be larger than the sum of it's children because of the way accounting for reservations work.
-
Danny Auble authored
was given with just an =, i.e. users= without specifying any users after the =.
-
Danny Auble authored
-
Danny Auble authored
period. This would only hit you if you rerolled a 15.08 prior to this commit.
-
Tim Wickberg authored
Don't expand on the format or arguments, these have changed in 16.05 already and may vary further with BB/generic support.
-
Alejandro Sanchez authored
-
Morris Jette authored
If job submit time is right at a second boundary, the test could fail TEST: 15.37 spawn /home/jette/SLURM/install_smd/bin/salloc --begin now+60 --deadline now+600 --time-min 10 sleep 1^M salloc: error: Job submit/allocate failed: Requested time limit is invalid (missing or exceeds some limit)^M FAILURE: batch not submitted with a deadline too short test15.37 FAILURE [2016-02-12T15:53:53.005] _valid_job_part: job's min_time greater than deadline (10 > 2016-02-12T16:03:52) [2016-02-12T15:53:53.005] _slurm_rpc_allocate_resources: Requested time limit is invalid (missing or exceeds some limit)
-
- 13 Feb, 2016 4 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
These are very unlikely to ever occur, but this helps harden the code.
-
- 12 Feb, 2016 6 commits
-
-
Brian Christiansen authored
-
Danny Auble authored
-
Brian Christiansen authored
-
Brian Christiansen authored
-
Brian Christiansen authored
-
Morris Jette authored
Add logic to read current KNL state information using Intel's syscfg command on the KNL node.
-