- 30 Apr, 2015 2 commits
-
-
Morris Jette authored
Fix scancel bug which could return an error on attempt to signal a job step. A simple "scancel 12.3" to signal a specific job step would fail. Adding another option (say "-i", "--partion=", etc.) would fix this.
-
David Bigagli authored
-
- 29 Apr, 2015 2 commits
-
-
Danny Auble authored
before ending the job.
-
Danny Auble authored
will make it so the slurmctld will not signal the apid's in a batch job. Instead it relies on the rpc coming from the slurmctld to kill the job to end things correctly.
-
- 28 Apr, 2015 3 commits
-
-
Morris Jette authored
Refactor scancel so that all pending jobs are cancelled before starting cancellation of running jobs. Otherwise they happen in parallel and the pending jobs can be scheduled on resources as the running jobs are being cancelled.
-
Danny Auble authored
-
jette authored
Add SchedulerParameters option of sched_min_interval that controls the minimum time interval between any job scheduling action. The default value is zero (disabled). bug 1623
-
- 24 Apr, 2015 2 commits
-
-
Morris Jette authored
Initialize some variables used with the srun --no-alloc option that may cause random failures.
-
Morris Jette authored
-
- 22 Apr, 2015 4 commits
-
-
Danny Auble authored
job. This is already handled on the stepd when the script finishes.
-
Danny Auble authored
often an inventory request is handled.
-
Brian Christiansen authored
-
Brian Christiansen authored
-
- 21 Apr, 2015 4 commits
-
-
Danny Auble authored
free afterwards would not have zeroed out memory on the variables that didn't get unpacked.
-
David Bigagli authored
-
Morris Jette authored
sbatch to stop parsing script for "#SBATCH" directives after first command. It keeps parsing so long as lines contain only white space or comments (first non-white space character is '#').
-
David Bigagli authored
-
- 20 Apr, 2015 3 commits
-
-
Morris Jette authored
Add SchedulerParameters option of "sched_max_job_start=" to limit the number of jobs that can be started in any single execution of the main scheduling logic.
-
Danny Auble authored
-
Danny Auble authored
-
- 18 Apr, 2015 1 commit
-
-
David Bigagli authored
-
- 17 Apr, 2015 3 commits
-
-
Brian Christiansen authored
Bug 1601
-
David Bigagli authored
-
Danny Auble authored
Bug 1603
-
- 16 Apr, 2015 1 commit
-
-
Danny Auble authored
and update NEWS
-
- 15 Apr, 2015 4 commits
-
-
Danny Auble authored
the same time, don't double bill the down time.
-
Danny Auble authored
database.
-
Danny Auble authored
list, but only the column list.
-
Morris Jette authored
Prevent slurmdbd error if cluster added or removed while rollup in progress. Removing a cluster can cause slurmdbd to abort. Adding a cluster can cause the slurmdbd rollup to hang.
-
- 14 Apr, 2015 2 commits
-
-
Morris Jette authored
When allocating resources with resolution of sockets, charge the job for a CPUs on allocated sockets rather than just the CPUs on used cores.
-
Nicolas Joly authored
bug 1595
-
- 13 Apr, 2015 1 commit
-
-
Morris Jette authored
The error was being triggered by logic to collect accounting information starting and not being completed when testing to start a node ping RPC. In other words, the logic wasn't testing for the completion of one RPC against the start of that same RPC, but against the start of a different RPC. This change moves all of the timeout logic into the ping_nodes.c module where we can make sure that the timing of different RPCs do not get confused with each other. bug 1190
-
- 11 Apr, 2015 3 commits
-
-
Morris Jette authored
slurmd was logging: "Error reading step 2031886.0 memory limits" (with various job/step IDs) because it was treating SLURM_SUCCESS as an error rather than SLURM_ERROR bug 1589
-
Danny Auble authored
-
Morris Jette authored
Disable changes to GRES count while jobs are running on the node. Previous logic could clear GRES count data and generate counter underflow then when attemtping to deallocate the job's resources (including future job terminations when computing when and where pending jobs will start as part of backfill scheduling logic). bug 1589
-
- 09 Apr, 2015 1 commit
-
-
Danny Auble authored
if many steps/jobs finish at once.
-
- 08 Apr, 2015 1 commit
-
-
David Bigagli authored
-
- 07 Apr, 2015 3 commits
-
-
Brian Christiansen authored
Bug 1587
-
Danny Auble authored
-
Morris Jette authored
switch/cray: If CR_PACK_NODES is configured, then set the environment variable "PMI_CRAY_NO_SMP_ENV=1" bug 1585
-