- 15 Jul, 2014 8 commits
-
-
Morris Jette authored
Conflicts: configure contribs/cray/Makefile.in src/common/stepd_api.c
-
Morris Jette authored
Fix race condition which could result in requeue if batch job exit and node registration occur at the same time.
-
Danny Auble authored
-
Danny Auble authored
(From that commit) There was a problem when building from source where for example @bindir@ would resolve to ${prefix}/bin. This patch fixes it, based on http://www.gnu.org/software/autoconf/manual/ autoconf-2.69/html_node/Installation-Directory-Variables.html It also changes opt_modulefiles_slurm to opt_modulefiles_slurm.in but I couldn't figure out how to get git diff to show that.
-
Danny Auble authored
only used in the slurmd and slurmstepd.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
and will avoid the need for all the daemons to link to libhwloc
-
- 14 Jul, 2014 14 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Rod Schultz authored
max_width parameter from split_hostlist.
-
Morris Jette authored
Without this change the function route_split_hostlist_treewidth() in src common would not be available to the route plugins
-
Morris Jette authored
-
Morris Jette authored
the slurm api can not be used without this change also, the route plugins were not in the RPMs
-
David Bigagli authored
and should there be problems accessing the state files.
-
Morris Jette authored
-
Morris Jette authored
Note that the map/mask specified applies to all allocated nodes.
-
Morris Jette authored
Previously the test could start and fail due to completing jobs
-
Morris Jette authored
-
Morris Jette authored
Fix for possible abort on change in GRES configuration. bug 958
-
Morris Jette authored
-
- 11 Jul, 2014 18 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Danny Auble authored
-
Rod Schultz authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Rod Schultz authored
-
Rod Schultz authored
-
Matthieu Hautreux authored
limitation in terms of scalability, I did some tests with cluster sizes varying from 64 to 64k nodes. I saw the same kind of performance issues that you saw with large sizes (more than a few thousands nodes). Looking at the code, I noticed that the hash table of the node records was not built before the construction of the topology information in slurmctld ! Making sure that the hash table is present really reduces the built time. This patch ensures that and enable to get sub-second built time for topology information in scenarios that were used to require a minute before. I think that using this patch, the load time of slurm confs is less problematic.
-
Matthieu Hautreux authored
implemented.
-
Rod Schultz authored
using switch topology information.
-
Morris Jette authored
Job requeue test assumed the job would not be started immediately after requeue, but that could happend and result in a test reporting FAILURE, when the job really was requeued.
-
Morris Jette authored
-
Danny Auble authored
Conflicts: testsuite/expect/test21.24
-