- 01 Oct, 2013 1 commit
-
-
Morris Jette authored
-
- 30 Sep, 2013 3 commits
-
-
Danny Auble authored
-
Morris Jette authored
sched/backfill - Prevent possible memory corruption due to use of bf_continue option and long running scheduling cycle (pending jobs could have been cancelled and purged).
-
Morris Jette authored
Change max message length from 100MB to 1GB before generating "Insane message length" error.
-
- 27 Sep, 2013 4 commits
-
-
Danny Auble authored
can only be a TORUS (1).
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
This specifies the maximum depth the backfill scheduler should go in any single partition.
-
- 25 Sep, 2013 1 commit
-
-
Morris Jette authored
-
- 24 Sep, 2013 4 commits
-
-
Danny Auble authored
srun is pending. The step creation would use the old values and possibly hang srun since the step wouldn't be able to be created in the modified allocation.
-
Morris Jette authored
-
Danny Auble authored
into the future.
-
Morris Jette authored
-
- 23 Sep, 2013 2 commits
-
-
Danny Auble authored
-
Morris Jette authored
bug 428
-
- 17 Sep, 2013 2 commits
-
-
Morris Jette authored
for setdebugflags command, avoid parsing "-flagname" as an scontrol command line option.
-
Armin Größlinger authored
for CR_Socket.
-
- 14 Sep, 2013 2 commits
-
-
Morris Jette authored
Add support for "on" and "before*" options
-
David Bigagli authored
bugfixes and the new item extraction feature.
-
- 13 Sep, 2013 4 commits
-
-
Morris Jette authored
-
Morris Jette authored
Set PBS_ACCOUNT, PBS_ENVIRONMENT, and PBS_QUEUE only for batch jobs and only if the user submission sets the account and partition.
-
Morris Jette authored
-
Morris Jette authored
No support for PBS "before" options
-
- 12 Sep, 2013 1 commit
-
-
Morris Jette authored
-l accelerator=true|false (GPU use) -l mpiprocs=# (processors per node) -l naccelerators=# (GPU count) -l select=# (node count) -l ncpus=# (task count) -v key=value (environment variable) -W umask=# (set job's umask) Note: the -v option does NOT support quoted commas.
-
- 11 Sep, 2013 3 commits
-
-
Morris Jette authored
-l accelerator=true|false (GPU use) -l mpiprocs=# (processors per node) -l naccelerators=# (GPU count) -l select=# (node count) -l ncpus=# (task count) -v key=value (environment variable) -W umask=# (set job's umask)
-
Morris Jette authored
-
Morris Jette authored
-
- 10 Sep, 2013 3 commits
-
-
Morris Jette authored
-
David Bigagli authored
that ended in the OverTimeLimit interval.
-
David Bigagli authored
-
- 09 Sep, 2013 2 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
- 06 Sep, 2013 2 commits
-
-
Danny Auble authored
-
Morris Jette authored
Caused by allocating single adapter per node of specific adapter type.
-
- 04 Sep, 2013 2 commits
-
-
David Bigagli authored
-
Morris Jette authored
Previous logic would pick CPUs then reject jobs that can not match GRES to the allocated CPUs. New logic first filters out CPUs that can not use the GRES, next picks CPUs for the job, and finally picks the GRES that best match those CPUs. bug 410
-
- 30 Aug, 2013 1 commit
-
-
Morris Jette authored
Report anything that is world writable.
-
- 29 Aug, 2013 3 commits
-
-
Danny Auble authored
/* Current code (<= 2.1) has it so we start the new * job with the next step id. This could be used * when restarting to figure out which step the * previous run of this job stopped on. */
-
Danny Auble authored
-
Morris Jette authored
The Cray version of those libraries must be used. bug 407
-