- 04 Mar, 2014 9 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Magnus Jonsson authored
-
Carlos Bederián authored
-
Danny Auble authored
on BGQ systems.
-
Don Lipari authored
and SLURM_STEP_ID env vars are set.
-
Danny Auble authored
-
Morris Jette authored
Based upon a study of the preemption logic, putting smaller jobs at the top of the queue (lower priority) will improve algorithm efficiency in terms of not releasing more resources than needed.
-
- 03 Mar, 2014 12 commits
-
-
David Bigagli authored
merged together.
-
Danny Auble authored
Conflicts: src/sacct/print.c
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
4TB of memory on them.
-
Danny Auble authored
-
Josko Plazonic authored
sacct.
-
Morris Jette authored
The previous logic would get an authentication credential while waiting for a message to forward. This change will re-read an authtentication credential if the wait is too long (greater than or equal to the message timeout). Replace white space with tabs
-
Morris Jette authored
Add support for the new ScheduleParameters of preempt_strict_order and preempt_reorder_count
-
Morris Jette authored
Add SchedulerParameters options of preempt_reorder_count and preempt_strict_order.
-
Morris Jette authored
-
- 02 Mar, 2014 5 commits
-
-
jette authored
This is for consistency. All of the other backfill parameters start with "bf_".
-
jette authored
Conflicts: NEWS src/slurmd/slurmd/req.c
-
jette authored
Add support for SchedulerParameters value of bf_max_job_start that limits the total number of jobs that can be started in a single iteration of the backfill scheduler. bug 607
-
jette authored
A test for bf_max_job_user was > rather than >=
-
jette authored
-
- 28 Feb, 2014 14 commits
-
-
Danny Auble authored
-
Danny Auble authored
the correct lib.
-
Morris Jette authored
Since the QOS priority is 32-bits, base the preemption priority upon that 32-bit field and ignore the job size. Variant of commit a3e74ef2
-
Danny Auble authored
-
Danny Auble authored
completes
-
Morris Jette authored
This avoids some misleading error message
-
Morris Jette authored
The job preemption order was based upon the lowest 16 bits of a QOS priority, but the field holds 32 bits. This uses a value of 0xffff if the QOS priority is higher.
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
If the slurm.conf changes to remove nodes from a job step in the process of running Node Health Check, the step record would be purged. Then when the NHC completes, it's step record pointer would be invalid. The change here will prevent such a job step record from being purged. This was found while investigating bug 612, but would be a very unusual condition and does not fix this bug.
-
Morris Jette authored
-
Danny Auble authored
-
Danny Auble authored
always be set to SLURM_SUCCESS since it isn't used before.
-
Morris Jette authored
If the switch/cray global state save/restore fails, avoid re-allocating ports based upon the job steps that we can recover
-