- 18 Jul, 2012 2 commits
-
-
alejluther authored
'exclusive' flag
-
https://github.com/SchedMD/slurmalejluther authored
Conflicts: src/api/reservation_info.c src/common/node_select.c src/common/node_select.h src/plugins/sched/backfill/backfill.c src/plugins/sched/builtin/builtin.c src/plugins/sched/wiki2/job_will_run.c src/plugins/select/bluegene/select_bluegene.c src/plugins/select/cons_res/dist_tasks.c src/plugins/select/cons_res/job_test.c src/plugins/select/cons_res/select_cons_res.c src/plugins/select/cray/other_select.c src/plugins/select/cray/other_select.h src/plugins/select/cray/select_cray.c src/plugins/select/linear/select_linear.c src/scontrol/create_res.c src/slurmctld/job_scheduler.c src/slurmctld/node_scheduler.c src/slurmctld/reservation.c src/slurmctld/reservation.h src/slurmctld/slurmctld.h
-
- 17 Jul, 2012 16 commits
-
-
Morris Jette authored
-
Morris Jette authored
This corresponds to commit dd2dce54 from Mark Grondona's work in squeue, but applied to the sview command.
-
Morris Jette authored
Slurm 2.4 minor fixes
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Rod Schultz authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Batch job will have environment variable set: "SLURM_CPU_FREQ_REQ=0" and this will be loaded and reported as an invalid value by srun.
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Rod Schultz authored
-
- 16 Jul, 2012 7 commits
-
-
Morris Jette authored
-
Morris Jette authored
This addresses trouble ticket 85
-
Morris Jette authored
-
Don Albert authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
- 13 Jul, 2012 15 commits
-
-
Danny Auble authored
runjob_mux
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Danny Auble authored
is always set when sending or receiving a message.
-
Tim Wickberg authored
-
Mark A. Grondona authored
Set SLURM_CONF in default prolog/epilog environment instead of only in spank prolog/epilog environment. This change fixes a potential hang during spank prolog/epilog execution due to the possibility of memory allocation after fork(2) and before exec(2) when invoking slurmstepd spank prolog|epilog. This also has the benefit that SLURM commands used in prolog and epilog scripts will use the correct slurm.conf file.
-
Mark A. Grondona authored
If exec_wait_child_wait_for_parent() fails for any reason, it is safer to abort immediately rather than proceed to execute the user's job.
-
Mark A. Grondona authored
On a failure of fork(2), slurmstepd would print an error and exit, possibly leaving previously forked children waiting. Ensure a better cleanup by killing all active children on fork failure before exiting slurmstepd.
-
Mark A. Grondona authored
Close the read end of the pipe slurmstepd uses to notify children it is time to call exec(2) in order to save one file descriptor per task. (Previously, the read side of the pipe wasn't closed until exec_wait_info was destroyed)
-
Mark A. Grondona authored
For some reason squeue was treating completing jobs the same as pending jobs, and reported the number of nodes as the maximum of requested nodelist, requested node count or CPUs (divided into nodes?) This is in contrast to the squeue manpage which explicitly states that the number of nodes reported for completing jobs should be only the nodes that are still allocated to the job. This patch removes the special handling of completing jobs in src/squeue/print.c:_get_node_cnt(), so that the squeue output for completing jobs matches documentation. A comment is also added so that developers looking at the code understand what is going on.
-
Morris Jette authored
Conflicts: doc/html/high_throughput.shtml
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Without this change the values known to slurmctld do not change without a cold-start.
-