- 16 May, 2017 1 commit
-
-
Morris Jette authored
Refactor the salloc option parsing logic to build a separate set of job descriptor data structures for each portion of the heterogeneous job. Only the last job record is current submitted to slurmctld.
-
- 15 May, 2017 1 commit
-
-
Morris Jette authored
-
- 13 May, 2017 3 commits
-
-
Morris Jette authored
-
Isaac Hartung authored
Bug 3695
-
Morris Jette authored
bug 3779
-
- 12 May, 2017 4 commits
-
-
Morris Jette authored
If capmc reports a node name, but not mcdram_cfg for the node, then log the missing data rather than assume the value is zero and report a value mismatch with cnselect.
-
Alejandro Sanchez authored
When requesting an operation on jobs, where the operation permits to specify more than one job in the same request, and a job array appears before a regular job (no-array job) in the list of jobs to operate with, the job_array_resp_msg_t pointer was not properly NULL'ed and thus incorrectly accessed when processing the no-array job. This fix prevents the crash from happening in the following scontrol operations: uhold, hold, suspend, requeue, requeuehold, update, release when the same request has <array_jobid>,<non-array_jobid> in this order in the job list to process. Bug 3759
-
Morris Jette authored
Job expansion example in FAQ enhanced to demonstrate operation in heterogeneous environments. bug 2979
-
Alejandro Sanchez authored
Do not attempt to schedule jobs after changing the power cap if there are already many active threads.
-
- 11 May, 2017 2 commits
-
-
Danny Auble authored
# Conflicts: # META # NEWS
-
Danny Auble authored
-
- 10 May, 2017 3 commits
-
-
Danny Auble authored
-
Dominik Bartkiewicz authored
Bug 3760
-
Danny Auble authored
didn't work at all. Bug 3712.
-
- 09 May, 2017 9 commits
-
-
Danny Auble authored
This reverts commit ecfd007f.
-
Dominik Bartkiewicz authored
-
Brian Christiansen authored
Continuation of 9a1370e3 CID 168995
-
Danny Auble authored
It was noticed that while doing any update to a job the admin comment would be blown away. This patch fixes that.
-
Dominik Bartkiewicz authored
Bug 3789
-
Danny Auble authored
run multiple tasks on multiple nodes. Changing the max nodes setting from 3 to 6 fixes the issue without apparent compromise to the test.
-
Danny Auble authored
destroying a mutex.
-
Brian Christiansen authored
When running sacct from a federated client, the db returns jobs for each cluster with duplicate jobs removed on each cluster. A federated job could have ran on a different cluster when the before the jobid's rolled. This patch filters out past old federated jobs and leaves the newest ones. Reverted d31965 which was too slow.
-
Brian Christiansen authored
This reverts commit d31965f3.
-
- 08 May, 2017 5 commits
-
-
Danny Auble authored
clear it as started.
-
Danny Auble authored
This reverts commit e87edf8d. Per Moe's suggestion, we revert this sense it most likely isn't totally correct.
-
Morris Jette authored
-
Morris Jette authored
- 06 May, 2017 3 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
-
- 05 May, 2017 9 commits
-
-
Danny Auble authored
-
Morris Jette authored
Work is incomplete, but for now we get all jobs in federation and highlight the local nodes associated with each job.
-
Danny Auble authored
select_g_select_nodeinfo_set(). This is a continuation of commit 80443cc1. Bug 3690. Test 1.62 will fail otherwise.
-
Danny Auble authored
we failed. Bug 3690 continuation of commit bf0429d1.
-
Morris Jette authored
Add "cluster_name" field to node_info_t and partition_info_t data structure. It is filled in only when the cluster is part of a federation and SHOW_GLOBAL flag used. Functions slurm_load_node() slurm_load_partitions() modified to show all nodes/partitions in a federation when the SHOW_GLOBAL flag is used.
-
Danny Auble authored
bf0429d1 this is no longer needed to wait as long. Bug 3690 continuation of commit bf0429d1.
-
Danny Auble authored
If this failed the job was 3/4's allocated and not in the database. This would cause a segfault if done where it happened before. Bug 3690 continuation of commit bf0429d1
-
Brian Christiansen authored
The decay thread could signal the cond before the init thread waits for it causing deadlock.
-
Morris Jette authored
-