Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions doc/html/sched_config.shtml
Original file line number Diff line number Diff line change
Expand Up @@ -129,6 +129,10 @@ see the <a href="slurm.conf.html">slurm.conf(5)</a> man page.
<UL>
<LI><B>bf_continue</B> - If set, then continue backfill scheduling after
periodically releasing locks for other operations.</LI>
<LI><B>bf_hetjob_sticky_preempt_timeout=#</B> - Number of seconds the backfill
scheduler will keep retrying the same heterogeneous job start attempt after
preemption has started for one of its components. Default value is 1800 seconds.
Set to 0 to disable.</LI>
<LI><B>bf_interval=#</B> - Interval between backfill scheduling attempts.
Default value is 30 seconds.</LI>
<LI><B>bf_max_job_part=#</B> - Maximum number of jobs to initiate per partition
Expand Down
11 changes: 11 additions & 0 deletions doc/man/man5/slurm.conf.5
Original file line number Diff line number Diff line change
Expand Up @@ -3969,6 +3969,17 @@ option is disabled by default. If enabled and \fBbf_hetjob_prio=min\fR is not
set, then it would be automatically set.
.IP

.TP
\fBbf_hetjob_sticky_preempt_timeout=#\fR
Number of seconds the backfill scheduler will keep retrying the same
heterogeneous job start attempt after preemption has started for one of its
components. During this window, the scheduler waits for preempted jobs on the
planned nodes to finish cleanup instead of immediately rolling back the
heterogeneous job start and replanning against a different node set.
Set to 0 to disable this sticky preemption wait.
The default value is 1800 seconds.
.IP

.TP
\fBbf_hetjob_prio=[min|avg|max]\fR
At the beginning of each backfill scheduling cycle, a list of pending to be
Expand Down
Loading