- 28 Nov, 2017 1 commit
-
-
Danny Auble authored
to start. Bug 4434
-
- 10 Nov, 2017 1 commit
-
-
Isaac Hartung authored
This now matches the sinfo documentation. Bug 4306.
-
- 07 Nov, 2017 2 commits
-
-
Alejandro Sanchez authored
Issue could be triggered when updating a partition node(s) with node(s) that were already in the partition, incorrectly increasing the node_record->part_cnt (number of associated partitions) and thus incorrectly extending the array of pointers to partitions associated with this node, leading to an array with repeated associated partitions pointers. Bug 4318.
-
Brian Gilmer authored
On CLE 6.0 mungedir is /usr; a 'module unload' call then removes /usr/bin from PATH which is rather inconvenient. Bug 4334.
-
- 03 Nov, 2017 1 commit
-
-
Isaac Hartung authored
Memory TRES was getting the pn_min_memroy value when updating the job. But the TRES memory value is the total memory of the job (e.g pn_min_memory * cpus || pn_min_memory * nodes). Bug 4177
-
- 01 Nov, 2017 3 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Ryan Day authored
CVE-2017-15566. Bug 4228.
-
- 30 Oct, 2017 2 commits
-
-
Danny Auble authored
Starting in MariaDB 10.2 many of the api commands started setting errno erroneously. Backport of 5b934425
-
Danny Auble authored
This reverts commit 7e5d3d15. Turns out the spank_task_privileged needs to execute inside the child process instead of the slurmstepd. Bug 4298
-
- 28 Oct, 2017 1 commit
-
-
Morris Jette authored
If configured with NodeFeatures=knl_cray and there are non-KNL nodes which include no features the slurmctld will abort without this patch when attempting strtok_r(NULL). bug 4294
-
- 27 Oct, 2017 1 commit
-
-
Isaac Hartung authored
Bug 4279
-
- 25 Oct, 2017 2 commits
-
-
Danny Auble authored
Before it would allow all sorts of things like adding/subtracting/multiplying/etc. It would cause warnings such as /home/bart/slurm-tmp5/src/common/layouts_mgr.c: In function ‘_layouts_load_automerge’: /home/bart/slurm-tmp5/src/common/layouts_mgr.c:363:21: error: ‘*’ in boolean context, suggest ‘&&’ instead [-Werror=int-in-bool-context] *lvalue = *lvalue * *rvalue; \ ~~~~~~~~^~~~ /home/bart/slurm-tmp5/src/common/layouts_mgr.c:1034:4: note: in expansion of macro ‘_entity_update_kv_helper’ _entity_update_kv_helper(type_t, operator); \ ^~~~~~~~~~~~~~~~~~~~~~~~ /home/bart/slurm-tmp5/src/common/layouts_mgr.c:1086:4: note: in expansion of macro ‘_layouts_load_merge’ _layouts_load_merge(bool, s_p_get_boolean); [tag] [reply] [−] Private Comment 16 Bug 4062
-
Felip Moll authored
Setting -Werror tricks the test into failing on the "error: In the GNU C Library, "major" is defined by <sys/sysmacros.h>. For historical compatibility, it is currently defined by <sys/types.h> as well, but we plan to remove this soon. To use "major", include <sys/sysmacros.h> directly. If you did not intend to use a system-defined macro "major", you should undefine it after including <sys/types.h>." error. Since the normal Slurm build uses -Werror, this warning on including both headers will then cause the build itself to fail. Bug 3982.
-
- 24 Oct, 2017 6 commits
-
-
Marshall Garey authored
Bug 4246
-
Brian Christiansen authored
Bug 4246
-
Brian Christiansen authored
Bug 4246
-
Brian Christiansen authored
when a job spans a reservation that was modified (multiple records for the same reservation). Bug 4246
-
Brian Christiansen authored
Bug 4246
-
Alejandro Sanchez authored
Down waking nodes right after ResumeTimeout has been reached if they are not responding. Otherwise we have to wait for ping_nodes() to handle this work, thus SlurmdTimeout comes into play giving the sensation to the end user that nodes got stuck in ALLOCATED# and job in CF state until ping_nodes() decides to mark them DOWN and requeue the job. Bug 4182
-
- 19 Oct, 2017 5 commits
-
-
Morris Jette authored
Update to commit 859f6c82
-
Tim Wickberg authored
-
Dominik Bartkiewicz authored
E.g., gpu:tesla:2 would have parsed as quantity "telsa" of gpu. Change the index value to the end of the array (-1 index value). Bug 4250.
-
Dominik Bartkiewicz authored
Rather than end up with "%.-1s" printed out in the output as snprintf refused to parse the format specifier. Bug 4164.
-
Felip Moll authored
bugzilla #4238 - Added a hint for the window manager to all popups and windows in order to get max, min, close buttons in Gnome, XFCE, and others.
-
- 18 Oct, 2017 1 commit
-
-
Danny Auble authored
Bug 4244
-
- 14 Oct, 2017 1 commit
-
-
Josh Samuelson authored
Pending job with administrator extended TimeLimit beyond partition's MaxTime remains pending with reason PartitionTimeLimit bug 4262
-
- 13 Oct, 2017 5 commits
-
-
Morris Jette authored
-
Brian Christiansen authored
The contoller will return node records with a NULL name for nodes that are hidden. This is so that you can map a partition_info's nodes -- using it's node_inx[] -- to a node record in the returned node_array from slurm_load_node(). Previously the perl api would leave an undefined object in the the node_array if the hidden nodes were found before a real node in the node_array and any hidden nodes at the end of the array from the controller wouldn't be counted for in the perl node_array. This patch adds empty hashes for hidden nodes and preserves the record_count and node_array from the slurmctld. Bug 4250
-
Brian Christiansen authored
The perl api leaves undefined objects in the node_array returned by load_nodes() for any node that is hidden. But 4250
-
Morris Jette authored
Bug 4003
-
Morris Jette authored
as process exits Bug 4003
-
- 10 Oct, 2017 4 commits
-
-
Brian Christiansen authored
when using xstrcasecmp. Matching up with other xstrcmp() functions.
-
Brian Christiansen authored
was missing
-
Isaac Hartung authored
Bug 4226
-
Tim Wickberg authored
that there was no bit_fmt was out of scope on the xfree. Passing a function address to xfree() predictably does not work very well. Change the variable name to avoid confusion. Bug 4241
-
- 05 Oct, 2017 1 commit
-
-
Brian Christiansen authored
Before: $ sbatch --wrap="sleep 300" Submitted batch job 228 $ squeue JOBID PARTITION NAME USER ST TIME CPUS NODELIST(REASON) 228 debug wrap brian PD 0:00 1 (AssocMaxUnknownPerNode) Fixed: $ squeue JOBID PARTITION NAME USER ST TIME CPUS NODELIST(REASON) 229 debug wrap brian PD 0:00 1 (AssocMaxCpuPerNode) $ sacctmgr mod account stuff set maxtrespernode=cpu=-1,mem=1 $ squeue JOBID PARTITION NAME USER ST TIME CPUS NODELIST(REASON) 229 debug wrap brian PD 0:00 1 (AssocMaxMemPerNode) $ sbatch --wrap="sleep 300" --gres=blah:2 -pgpu Submitted batch job 235 $ squeue JOBID PARTITION NAME USER ST TIME CPUS NODELIST(REASON) 235 gpu wrap brian PD 0:00 1 (AssocMaxGRESPerNode)
-
- 04 Oct, 2017 1 commit
-
-
Morris Jette authored
burst_buffer/cray plugin modified to work with changes in Cray UP06 software release. Specific changes: Cray software now returns an error if a state_in or stage_out script is processed that doesn't actually request a stage in or out (previously silently ignored). Also the warning message about tearing down a buffer that is already gone changed.
-
- 02 Oct, 2017 2 commits
-
-
Dominik Bartkiewicz authored
Move the check up a bit more where it'll do some good. Bug 4184.
-
Dominik Bartkiewicz authored
Bug 4146.
-