- 20 Feb, 2015 3 commits
-
-
Morris Jette authored
-
Dorian Krause authored
we came across the following error message in the slurmctld logs when using non-consumable resources: error: gres/potion: job 39 dealloc of node node1 bad node_offset 0 count is 0 The error comes from _job_dealloc(): node_gres_data=0x7f8a18000b70, node_offset=0, gres_name=0x1999e00 "potion", job_id=46, node_name=0x1987ab0 "node1") at gres.c:3980 (job_gres_list=0x199b7c0, node_gres_list=0x199bc38, node_offset=0, job_id=46, node_name=0x1987ab0 "node1") at gres.c:4190 job_ptr=0x19e9d50, pre_err=0x7f8a31353cb0 "_will_run_test", remove_all=true) at select_linear.c:2091 bitmap=0x7f8a18001ad0, min_nodes=1, max_nodes=1, max_share=1, req_nodes=1, preemptee_candidates=0x0, preemptee_job_list=0x7f8a2f910c40) at select_linear.c:3176 bitmap=0x7f8a18001ad0, min_nodes=1, max_nodes=1, req_nodes=1, mode=2, preemptee_candidates=0x0, preemptee_job_list=0x7f8a2f910c40, exc_core_bitmap=0x0) at select_linear.c:3390 bitmap=0x7f8a18001ad0, min_nodes=1, max_nodes=1, req_nodes=1, mode=2, preemptee_candidates=0x0, preemptee_job_list=0x7f8a2f910c40, exc_core_bitmap=0x0) at node_select.c:588 avail_bitmap=0x7f8a2f910d38, min_nodes=1, max_nodes=1, req_nodes=1, exc_core_bitmap=0x0) at backfill.c:367 The cause of this problem is that _node_state_dup() in gres.c does not duplicate the no_consume flag. The cr_ptr passed to _rm_job_from_nodes() is created with _dup_cr() which calls _node_state_dup(). Below is a simple patch to fix the problem. A "future-proof" alternative might be to memcpy() from gres_ptr to new_gres and only handle pointers separately.
-
Morris Jette authored
-
- 19 Feb, 2015 4 commits
-
-
Brian Christiansen authored
Bug 1471
-
Morris Jette authored
-
Morris Jette authored
"If you specify a maximum node count and the host list contains more nodes, the extra node names will be silently ignored." Not so.
-
Danny Auble authored
runs certain sreport reports.
-
- 18 Feb, 2015 11 commits
-
-
Morris Jette authored
-
Morris Jette authored
For srun command with the --no-alloc option, the dummy credential created did not have two new fields (job_constraints and job_gres_list) initialized, resulting in invalid memory references. Bug introduced earlier today.
-
Morris Jette authored
Added "--mail=stage_out" option to job submission commands to notify user when burst buffer state out is complete.
-
Morris Jette authored
Add new job_descriptor fields to the job_submit/lua interface: clusters, power_flags, and sicp_mode
-
Morris Jette authored
Add SLURM_JOB_CONSTAINTS to environment variables available to the Prolog. bug 1458
-
Morris Jette authored
Add job credential to "Run Prolog" RPC used with a configuration of PrologFlags=alloc. This allows the Prolog to be passed identification of GPUs allocated to the job.
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Add SLURM_JOB_GPUS environment variable to those available in Prolog. Also add list of environment variables available in the various prologs and epilogs on the web page. bug 1458
-
Danny Auble authored
-
Brian Christiansen authored
-
- 17 Feb, 2015 11 commits
-
-
Danny Auble authored
runjob happened, and the step was part of an array. This is an addition to commit 49e0f5f2
-
Danny Auble authored
in the runjob_mux plugin.
-
Brian Christiansen authored
Bug 1461 Commit: 2e2d924e
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
This completes work on the JSON parser for the capmc get_power_cap_capabilities command, which reports specifics about each node's power caps and ranges.
-
Morris Jette authored
Add a find_node_record2() function that differs from find_node_record() in that it does not print an error if a node name is not found. This is important for use with the Cray capmc command, which generates information about not only compute nodes, but also login and service nodes. Using this function prevents a bunch of errors from the power/cray plugin when trying to find these non-compute nodes in the slurm configuration.
-
Morris Jette authored
-
Morris Jette authored
See bug 1461
-
- 14 Feb, 2015 2 commits
-
-
Morris Jette authored
-
Danny Auble authored
-
- 13 Feb, 2015 3 commits
-
-
David Bigagli authored
-
Morris Jette authored
-
Morris Jette authored
If call was made to change a node's state to the same state it was already in and set its reason to the same value it already had, then an accounting record was generated. If a script, say NodeHealthCheck is repeatedly setting a node state (say DRAIN), it could generate a huge number of redundant accounting records. This eliminates these redundant records. related to bug 1437
-
- 12 Feb, 2015 6 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Conflicts: META NEWS
-
Morris Jette authored
-
Morris Jette authored
-
Brian Christiansen authored
-