- 07 Jun, 2018 1 commit
-
-
Isaac Hartung authored
if any task in the array was requeued. This is a hint to use "sacct --duplicates" to see the whole picture of the array job. Bug 5105
-
- 06 Jun, 2018 3 commits
-
-
Morris Jette authored
burst_buffer.conf - Add SetExecHost flag to enable burst buffer access from the login node for interactive jobs.
-
Alejandro Sanchez authored
And remove the initialization before all the calls to the function. It is non-functional and the motivation is more a preventive thing so that if we ever use slurm_mktime() we know tm_isdst is consistently set to -1. Bug 5230.
-
Brian Christiansen authored
which were marked down due to ResumeTimeout. If a cloud node was marked down due to not responding by ResumeTimeout, the code inadvertently added the node back to the avail_node_bitmap -- after being cleared by set_node_down_ptr(). The scheduler would then attempt to allocate the node again, which would cause a loop of hitting ResumeTimeout and allocating the downed node again. Bug 5264
-
- 05 Jun, 2018 1 commit
-
-
Killian authored
Bug 5206.
-
- 04 Jun, 2018 1 commit
-
-
Morris Jette authored
-
- 02 Jun, 2018 1 commit
-
-
Brian Christiansen authored
srun would not return an exit code if a previous task exited before a latter task exited with a signal. If multiple tasks exit with a signal, srun returns the highest signal. Partially reverts commit 04b449e1 -- the setting of local_global_rc to NO_VAL as srun doesn't need to know whether it's been set or not anymore. srun always sets the signal if a task exited with a signal. Bug 5083
-
- 31 May, 2018 1 commit
-
-
Alejandro Sanchez authored
There were two code paths building an allocation response by calling its own static _build_alloc_msg() function: 1. src/slurmctld/proc_req.c 2. src/slurmctld/srun_comm.c These two functions diverged and both had members that were not filled in but were filled in the other. This patch makes it so we change the signature of the one in proc_req.c to make it extern and then in srun_comm.c we call this newly common function. Also added cpu_freq_[min|max|gov] members in the common one since these were the only members missing in proc_req.c function (the one in srun_comm.c had more members missing, like all the ntasks_per*, account, qos or resv_name). Bug 4999.
-
- 30 May, 2018 7 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Marshall Garey authored
Only trust MUNGE signed values, unless the RPC was signed by SlurmUser or root. CVE-2018-10995.
-
Tim Wickberg authored
Do not defer until later, and do not potentially miss out on proper validation of the user_name field which can lead to improper authentication handling. CVE-2018-10995.
-
Dominik Bartkiewicz authored
Bug 5038.
-
Tim Wickberg authored
Caused by pthread_cancel cleanup by commit e5f03971 in 17.11.6. Bug 5181.
-
Tim Wickberg authored
The race condition was created in a7c8964e in 17.11.6 when removing the (unsafe) pthread_cancel code handling thread termination. Bug 5164
-
- 24 May, 2018 1 commit
-
-
Brian Christiansen authored
Commits f18390e8 and eed76f85 modified the stepd so that if the stepd encountered an unkillable step timeout that the stepd would just exit the stepd. If the stepd is a batch step then it would reply back to the controller with a non-zero exit code which will drain the node. But if an srun allocation/step were to get into the unkillable step code, the steps wouldn't let the waiting srun or controller know about the step going away -- leaving a hanging srun and job. This patch enables the stepd to notify the waiting sruns and the ctld of the stepd being done and drains the node for srun'ed alloction and/or steps. Bug 5164
-
- 21 May, 2018 1 commit
-
-
Dominik Bartkiewicz authored
g_qos_count, g_qos_max_priority, must be call under qos write lock. Bug 5159.
-
- 17 May, 2018 2 commits
-
-
Danny Auble authored
PriorityFlags=ACCRUE_ALWAYS is set. Bug 5186
-
Morris Jette authored
Completely remove "gres" field from step record in slurmctld and step info message. Use "tres_per_node", "tres_per_socket", etc.
-
- 16 May, 2018 2 commits
-
-
Morris Jette authored
Add node_features plugin function "node_features_p_reboot_weight()" to return the node weight to be used for a compute node that requires reboot for use (e.g. to change the NUMA mode of a KNL node). Add NodeRebootWeight parameter to knl.conf configuration file.
-
Alejandro Sanchez authored
Previously the default paths continued to be tested even when new ones were requested. This had as a consequence that if any of the new paths was the same as any of the default ones (i.e. /usr or /usr/local), the configure script was incorrectly erroring out specifying that a version of PMIx was already found in a previous path. Bug 5168.
-
- 15 May, 2018 2 commits
-
-
Morris Jette authored
Add node_features plugin function "node_features_p_reboot_weight()" to return the node weight to be used for a compute node that requires reboot for use (e.g. to change the NUMA mode of a KNL node). Add NodeRebootWeight parameter to knl.conf configuration file.
-
Alejandro Sanchez authored
Previously the default paths continued to be tested even when new ones were requested. This had as a consequence that if any of the new paths was the same as any of the default ones (i.e. /usr or /usr/local), the configure script was incorrectly erroring out specifying that a version of PMIx was already found in a previous path. Bug 5168.
-
- 10 May, 2018 1 commit
-
-
Alejandro Sanchez authored
First issue was identified on multi partition requests. job_limits_check() was overriding the original memory requests, so the next partition Slurm validating limits against was not using the original values. The solution consists in adding three members to job_details struct to preserve the original requests. This issue is reported in bug 4895. Second issue was memory enforcement behavior being different depending on job the request issued against a reservation or not. Third issue had to do with the automatic adjustments Slurm did underneath when the memory request exceeded the limit. These adjustments included increasing pn_min_cpus (even incorrectly beyond the number of cpus available on the nodes) or different tricks increasing cpus_per_task and decreasing mem_per_cpu. Fourth issue was identified when requesting the special case of 0 memory, which was handled inside the select plugin after the partition validations and thus that could be used to incorrectly bypass the limits. Issues 2-4 were identified in bug 4976. Patch also includes an entire refactor on how and when job memory is is both set to default values (if not requested initially) and how and when limits are validated. Co-authored-by: Dominik Bartkiewicz <bart@schedmd.com>
-
- 09 May, 2018 9 commits
-
-
Morris Jette authored
If running without AccountingStorageEnforce but with the DBD and it isn't up when starting the slurmctld you could get into a corner case where you don't have a QOS list in the assoc_mgr. Thus no usage to transfer. Bug 5156
-
Tim Wickberg authored
-
Felip Moll authored
-
Morris Jette authored
Try to fill up each socket completely before moving into additional sockets. This will minimize the number of sockets needed, improving packing especially alongside MaxCPUsPerNode. Bug 4995.
-
Tim Wickberg authored
My mistake on commit 602817c8. Bug 4922.
-
Felip Moll authored
Without this, gang scheduling would incorrectly kick in for these jobs since active_resmap has not been updated appropriately. Bug 4922.
-
Tim Wickberg authored
Otherwise this will return the error message back to the next job submitter. Bug 5106.
-
Tim Wickberg authored
Bug 5106.
-
Alejandro Sanchez authored
job_ptr->part_ptr is NULL if the partition has been deleted. Crash only happens with PriorityFlags=CALCULATE_RUNNING enabled. Bug 5136.
-
- 08 May, 2018 3 commits
-
-
Brian Christiansen authored
Bug 5146
-
Tim Wickberg authored
Caused by a corrupted protocol_version field value being received by the slurmstepd, as we cannot safely write/read a uint16_t across the pipe as if it was an int. Regression caused by commit 90b116c2. Bug 5133.
-
Brian Christiansen authored
Requeued jobs are marked as PENDING|COMPLETING until the epilog checks in. The issue is that if job_set_alloc_tres gets called while in the PENDING|COMPLETING state, the job's alloc_tres_str will be free'd. If this job then gets checkpointed in this state (PENDING|COMPLETING + no tres_alloc_str) on startup the controller would crash because it expected the job to have a tres_alloc_str/cnt when in the COMPLETING state. This could be triggered if starting the controller without the dbd up. When the dbd comes up, the assoc_cache_mgr calls _update_job_tres() which calls job_set_alloc_tres. It could also be triggered by adding new tres. This most likely started happening in 17.11.5 because of commit 865b672f which introduced calling _update_job_tres() on each job after the dbd comes up. Bugs 5137,4522
-
- 04 May, 2018 1 commit
-
-
Brian Christiansen authored
Only when the connection has timedout. If the connection is timing out, consider increasing TCPTimeout in the slurm.conf Bug 4574
-
- 03 May, 2018 3 commits
-
-
Boris Karasev authored
Bug 5129.
-
Alejandro Sanchez authored
Use setenv() instead of setenvfs(), since setenvfs() memory allocation is implemented with xmalloc() and fini_setproctitle() (which is called on reconfigure) free's the memory with free(), leading to a: "free(): invalid size" malloc_printerr error. Continuation of dce83a23. Bug 5095.
-
Felip Moll authored
Due to current design the job limits are checked before the allocation is made when one specifies a generic gres and a specific gres type is configured. The workaround for now is to define a job submit plugin to control the user request and succesfully apply limits. Bug 4767
-