- 06 Sep, 2017 3 commits
-
-
Brian Christiansen authored
-
Marshall Garey authored
to match the parameter ordering.
-
Marshall Garey authored
Bug 4052
-
- 05 Sep, 2017 8 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Brian Christiansen authored
From ea621ec4
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Reported by Clang
-
Morris Jette authored
Add retry on file open fail and remove some unnecessary sleeps
-
Morris Jette authored
-
- 04 Sep, 2017 5 commits
-
-
Alejandro Sanchez authored
** CID 175193: (FORWARD_NULL) Theoretically we shouldn't have a job_desc_msg_t without an associated part_record, but just in case let's harden the code. Introduced in previous commit: 24365514.
-
Morris Jette authored
-
Morris Jette authored
If portions of a pack job launch fail, then release the full job allocation rather than having srun exit with outstanding job allocations.
-
Morris Jette authored
Bug 4132
-
Alejandro Sanchez authored
Initially job mem limits were tested at submission time through _validate_min_mem_partition() -> _valid_pn_min_mem(), but not tested again at scheduling time, thus leading to jobs incorrectly being scheduled against partitions where the job exceeded their MaxMemPer* limit (which can in turn be inherited from the system wide limit too). NOTE: New WAIT_PN_MEM_LIMIT job_state_reason enum component added to support this new waiting reason. Bug 2291.
-
- 02 Sep, 2017 2 commits
-
-
Morris Jette authored
Modify step create logic so that call components of a heterogeneous job launched by a single srun command have the same step ID value.
-
Morris Jette authored
-
- 01 Sep, 2017 14 commits
-
-
Danny Auble authored
-
Danny Auble authored
checked on submit. This only mattered when submitting a job to multiple partitions. Bug 4066
-
Danny Auble authored
on node 0. Bug 4035
-
Morris Jette authored
Prevent a heterogeneous job allocation from including the same nodes in multiple components (required by MPI jobs spanning components).
-
Tim Wickberg authored
Add the xassert() that the inline version has to _remove_job_hash().
-
Tim Wickberg authored
To avoid a reliance on the slurm.conf file (as this command is designed to help you construct said file) the path to check the space was hard-coded as /tmp. But, if TmpFS is meant to be elsewhere we'd emit the wrong value here. Bug 3272.
-
Tim Wickberg authored
A lot of slurmdbd operations are authenticated by the accounting_storage plugin, rather than in slurmdbd. To allow the drop_priv flag to work it must be checked in is_user_min_admin_level() in addition to the various functions in proc_req.c .
-
Tim Wickberg authored
If SLURM_TESTSUITE_DROP_PRIV is defined (regardless of the value), set SLURM_DROP_PRIV in the message header so the slurmctld/slurmdbd will treat the request as if issued from an unprivileged user account. Bug 4080.
-
Tim Wickberg authored
See prior commit da1d3858. Bug 4080.
-
Tim Wickberg authored
This reverts commit 46a46b91.
-
Tim Wickberg authored
This reverts commit a2134b14.
-
Tim Wickberg authored
This reverts commit 29fe710d.
-
Tim Wickberg authored
This reverts commit 47dad9e8.
-
Tim Wickberg authored
On second thought, we should be using these or quite similar functions. This reverts commit a44ef130.
-
- 31 Aug, 2017 8 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Morris Jette authored
Make error specify the same option name as on srun input
-
Morris Jette authored
Check for an invalid pack group value before checking for nothing to start
-
Tim Wickberg authored
-
Tim Wickberg authored
Will be removed in 17.11, and 'sgj_job' is a typo anyways.
-
Tim Wickberg authored
-
Tim Wickberg authored
A lot of slurmdbd operations are authenticated by the accounting_storage plugin, rather than in slurmdbd. To allow the drop_priv flag to work it must be checked in is_user_min_admin_level() in addition to the various functions in proc_req.c .
-