- 31 May, 2014 5 commits
-
-
Danny Auble authored
-
David Gloe authored
messages, moving some messages to higher debug levels, and consolidating some needlessly verbose messages.
-
David Gloe authored
applications from the minimum (1%) to the more appropriate 25%.
-
David Gloe authored
the Aries network.
-
David Gloe authored
network systems - Resolves a bug involving nested batch steps and aeld.
-
- 30 May, 2014 9 commits
-
-
David Gloe authored
(mostly useful for debugging, it isn't used for much) Sets task_is_app, which tells the kernel that this is an application process (for core specialization)
-
David Gloe authored
limits to 4 concurrent jobs per node for our network resources.
-
David Gloe authored
-
David Gloe authored
-
David Gloe authored
generic resources and removing unused warmswap code.
-
David Gloe authored
@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.
-
Morris Jette authored
If a job allocates whole nodes (with --core-spec or --exclusive option) and launches a step within that allocation all in a single command (the srun creates the allocation and step at the same time) and requests a specific CPU count (e.g. "-ntasks=# --cpus-per-task=#") then allocate the job step only the requested CPU count, which may be less than the job's allocation. Bug 843
-
Morris Jette authored
-
Morris Jette authored
If shutdown of the slurmctld daemon is in progress, then stop trying to schedule jobs or process reconfigure requests. These are the only operations that take a significant amount of time and only service to slow down the shutdown process. We want the daemon to stop processing incoming RPCs and save state as soon as possible.
-
- 29 May, 2014 14 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Previous limit was 4 secs, raised to 10.
-
Morris Jette authored
Increase delays if memory leak test Disable one test if tracking lustre file system use (leaves open file)
-
Morris Jette authored
Eliminate one-time memory leak by freeing RPC statistics table at slurmctld shutdown
-
Morris Jette authored
Eliminate a one-time memory leack of the assocation hash table at shutdown.
-
Morris Jette authored
-
Morris Jette authored
select/cons_res plugin: Fix memory leak related to job preemption. bug 837
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
Conflicts: src/slurmd/slurmd/req.c
-
Danny Auble authored
d75bcaa5
-
- 28 May, 2014 12 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Marlys Kohnke authored
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
Show exactly waht the bad ProfileHDF5Default value is.
-
Morris Jette authored
ProfileHDF5Default=Filesystem should be ProfileHDF5Default=Lustre
-
Morris Jette authored
Added double brackets so brackets could be used within test program
-
Morris Jette authored
Fix for commit 3f7bd421
-
Morris Jette authored
Replace AC_TRY_RUN with more modern AC_RUN_IFELSE and assume printf of NULL does not seg fault
-
Morris Jette authored
This give system administrators the option on AMD Opteron 6000 series processors of either considering each NUMA node on a socket as a separate socket (resulting in some incorrect logging of socket count information) or not (resulting in sub-optimal job allocations since each core in the socket will be considered equivalent, even if on different NUMA nodes within the socket). bug 838
-
Morris Jette authored
-