- 19 Aug, 2013 3 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
- 16 Aug, 2013 2 commits
-
-
Morris Jette authored
This makes it consistent with the value of default_queue_depth. The backfill scheduler should be able to easily handle this value (or much higher for pretty much any configuration).
-
Danny Auble authored
-
- 15 Aug, 2013 5 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
could end up before the job started. Bug 371
-
Morris Jette authored
This function can now be called to test for processes which are dumping in order to avoid sending them a SIGKILL until dump completes. Change in logic required for job_container/cray.
-
Danny Auble authored
-
- 14 Aug, 2013 5 commits
-
-
Danny Auble authored
plugins more maintainable.
-
Morris Jette authored
This avoids waiting for the job's initiation to fail.
-
Morris Jette authored
Fix job state recovery logic in which a job's accounting frequency was not set. This would result in a value of 65534 seconds being used (the equivalent of NO_VAL in uint16_t), which could result in the job being requeued or aborted.
-
David Bigagli authored
-
Morris Jette authored
Problem reported by BYU. slurm.conf included a file one byte in length. Logic created a buffer one byte long and used fgets() to read the file. fgets() reads one byte less than the buffer size to include a trailing '\0', so it fails to read the file.
-
- 13 Aug, 2013 3 commits
-
-
Morris Jette authored
-
jette authored
This problem was reported by Harvard University and could be reproduced with a command line of "srun -N1 --tasks-per-node=2 -O id". With other job types, the error message could be logged many times for each job. This change logs the error once per job and only if the job request does not include the -O/--overcommit option.
-
Danny Auble authored
was down (slurmctld not running) during that time period.
-
- 09 Aug, 2013 2 commits
-
-
Danny Auble authored
version of Slurm.
-
Danny Auble authored
the reverse of this comment to return the code to the mix.
-
- 08 Aug, 2013 3 commits
-
-
Mark Nelson authored
-
Morris Jette authored
-
Morris Jette authored
Visible only using "scontrol show job" today and these fields are only relevant for batch jobs.
-
- 07 Aug, 2013 3 commits
-
-
Morris Jette authored
Add mechanism for job_submit plugin to generate error message for srun salloc or sbatch to log. New argument added to job_submit function in the plugin. bug 278
-
Danny Auble authored
-
Morris Jette authored
with a vertical bar separating each field.
-
- 06 Aug, 2013 4 commits
-
-
Morris Jette authored
Previously only the spawned job steps would be signalled rather than the batch shell
-
Danny Auble authored
of at multifactor poll.
-
Morris Jette authored
Used to extract plugin specific fields from the job's or step's GRES data structure.
-
Stephen Trofinoff authored
Allocate jobs to nodes with the largest number of available CPUs. Added SchedulingParameters paramter of "CR_LLN" and partition parameter of "LLN=yes|no".
-
- 02 Aug, 2013 1 commit
-
-
Morris Jette authored
Generate an error return rather than fatal error and exit if the configuraiton file is absent or invalid. This will permit Slurm APIs to be more reliably used by other programs. bug 374
-
- 01 Aug, 2013 1 commit
-
-
David Bigagli authored
to drain the node and log error slurmd log file.
-
- 31 Jul, 2013 2 commits
-
-
Danny Auble authored
MaxBlockInError set the job wait reason appropriately.
-
David Bigagli authored
-
- 30 Jul, 2013 1 commit
-
-
Thomas Cadeau authored
-
- 27 Jul, 2013 1 commit
-
-
Morris Jette authored
Just a bunch of empty functions now. See bug 77
-
- 26 Jul, 2013 2 commits
-
-
David Bigagli authored
-
Morris Jette authored
-
- 25 Jul, 2013 2 commits
-
-
David Bigagli authored
-
Danny Auble authored
%_with_cray to %_with_cray_alps.
-