- 06 Jul, 2013 1 commit
-
-
Morris Jette authored
-
- 05 Jul, 2013 2 commits
-
-
John Thiltges authored
When using ThreadsPerCore > 1, it appears that DefMemPerCPU is being scaled by slurmctld, but not by slurmd/slurmstepd. For example, we set ThreadsPerCore=2 and DefMemPerCPU=100. Running a single core job, we would expect two threads to be allocated and AllocMem on the assigned node to increase by 200MB. scontrol reports that AllocMem increased by 200MB, but the task/cgroup plugin only sees 100M of RAM. It looks like the problem may lie in common/slurm_cred.c:format_core_allocs(). The function counts the job/step cores and multiplies the mem_limit's, but it does not scale the CPU count like in slurmd/slurmd/req.c:_check_job_credential(). See bug 309
-
jette authored
-
- 28 Jun, 2013 3 commits
-
-
Morris Jette authored
Effects jobs with --exclusive and --cpus-per-task options bug 355
-
Danny Auble authored
-
Danny Auble authored
enum.
-
- 26 Jun, 2013 2 commits
-
-
Morris Jette authored
-
Dominik Friedrich authored
-
- 25 Jun, 2013 3 commits
-
-
Thomas Cadeau authored
-
Danny Auble authored
for as the job/step id.
-
David Gloe authored
The SLURM Makefile.am scripts use pkglibexecdir. One source indicates that this was not added until automake 1.10.2 (https://github.com/rerun/rerun/issues/167). So we just made that to be the minimum.
-
- 24 Jun, 2013 1 commit
-
-
jette authored
Under very heavy load with many thousands of batch job submissions or job signals, the write lock can be held for very long periods of time preventing job scheduling, squeue response, etc. This code inserts a timing break to permit other functions to get the locks.
-
- 21 Jun, 2013 4 commits
-
-
Danny Auble authored
-
Danny Auble authored
default and appeared to break other things.
-
Danny Auble authored
-
Martin Perry authored
specified.
-
- 18 Jun, 2013 2 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
- 12 Jun, 2013 1 commit
-
-
Morris Jette authored
if on "scontrol reconfig" when AllowNodes is manually changed using scontrol since last slurmctld restart.
-
- 10 Jun, 2013 1 commit
-
-
Morris Jette authored
due to either down nodes or explicit resizing. Generated slurmctld errors of this type: [2013-06-04T12:43:46+06:00] error: gres/gpu: step_test 68662.4294967294 gres_bit_alloc is NULL This is a movement of the logic introduced in commit https://github.com/SchedMD/slurm/commit/6fff97bb77d2d88aa808c47fd7880246a0c1d090 to eliminate a memory leak.
-
- 07 Jun, 2013 1 commit
-
-
Danny Auble authored
overwritten by the parallel versions thus making it so we need handle both cases.
-
- 06 Jun, 2013 1 commit
-
-
Mark Nelson authored
-
- 05 Jun, 2013 5 commits
-
-
Danny Auble authored
Since we don't currently track energy usage per task (only per step). Otherwise we get double the energy.
-
Danny Auble authored
-
Janne Blomqvist authored
Andy Wettstein (University of Chicago) reported privately to me that slurmctld 2.5.4 crashed after he enabled the priority/multifactor2 plugin due to a division by zero error. I was able to reproduce the crash by creating an account hierarchy where all the accounts and users had zero shares. See bug 315
-
David Bigagli authored
Revert premature change of META
-
jette authored
Without this change, it appears that POE ignores the -procs argument resulting in a job step request with multiple host names, but only one ntask required
-
- 04 Jun, 2013 3 commits
-
-
Morris Jette authored
-
jette authored
For example "host1*2" is equivalent to "host1,host1".
-
jette authored
Without this change, it appears that POE ignores the -procs argument resulting in a job step request with multiple host names, but only one ntask required
-
- 03 Jun, 2013 2 commits
-
-
Morris Jette authored
-
Hongjia Cao authored
We're having some trouble getting our slurm jobs to successfully restart after a checkpoint. For this test, I'm using sbatch and a simple, single-threaded executable. Slurm is 2.5.4, blcr is 0.8.5. I'm submitting the job using sbatch: $ sbatch -n 1 -t 12:00:00 bin/bowtie-ex.sh I am able to create the checkpoint and vacate the node: $ scontrol checkpoint create 137 .... time passes .... $ scontrol vacate 137 At that point, I see the checkpoint file from blcr in the current directory and the checkpoint file from Slurm in /var/spool/slurm-llnl/checkpoint. However, when I attempt to restart the job: $ scontrol checkpoint restart 137 scontrol_checkpoint error: Node count specification invalid In slurmctld's log (at level 7) I see: [2013-05-29T12:41:08-07:00] debug2: Processing RPC: REQUEST_CHECKPOINT(restart) from uid=***** [2013-05-29T12:41:08-07:00] debug3: Version string in job_ckpt header is JOB_CKPT_002 [2013-05-29T12:41:08-07:00] _job_create: max_nodes == 0 [2013-05-29T12:41:08-07:00] _slurm_rpc_checkpoint restart 137: Node count specification invalid
-
- 31 May, 2013 1 commit
-
-
Morris Jette authored
Rename slurm_step_ctx_params_t field from "mem_per_cpu" to "pn_min_memory". Job step now accepts memory specification in either per-cpu or per-node basis.
-
- 30 May, 2013 1 commit
-
-
Morris Jette authored
Uninitialized variables resulted in error of "cons_res: sync loop not progressing, holding job #"
-
- 29 May, 2013 1 commit
-
-
jette authored
The most notable problem case is on a cray where a job step specifically requests one or more node that are not the first nodes in the job allocation
-
- 24 May, 2013 2 commits
-
-
Morris Jette authored
-
Morris Jette authored
If set, the job's size component will be based upon not the job size alone, but the job's size divided by it's time limit.
-
- 23 May, 2013 3 commits
-
-
Morris Jette authored
The problem we have observed is the backfill scheduler temporarily gives up its locks (one second), but then reclaims them before the backlog of work completes, basically keeping the backfill scheduler running for a really long time when under a heavy load. bug 297
-
Morris Jette authored
-
Morris Jette authored
Defers (rather than forgets) reboot request with job running on the node within a reservation.
-