- 24 Jul, 2018 5 commits
-
-
Tim Wickberg authored
Leads to deadlock in 'scontrol reconfigure': 0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51 1 0x00007f61b1880801 in __GI_abort () at abort.c:79 2 0x00007f61b1fe2bd7 in __xassert_failed (expr=expr@entry=0x7f61b1ff2989 "_store_locks(locks)", file=file@entry=0x7f61b1ff2e80 "../../../../slurm/src/common/assoc_mgr.c", line=line@entry=2114, func=func@entry=0x7f61b1ff4bf8 <__func__.17903> "assoc_mgr_lock") at ../../../../slurm/src/common/xassert.c:57 3 0x00007f61b1ead017 in assoc_mgr_lock (locks=locks@entry=0x7f61ad6da610) at ../../../../slurm/src/common/assoc_mgr.c:2114 4 0x00005632675e4509 in _adjust_limit_usage (type=type@entry=0, job_ptr=job_ptr@entry=0x7f6190015000) at ../../../../slurm/src/slurmctld/acct_policy.c:719 5 0x00005632675e4b49 in acct_policy_add_job_submit (job_ptr=job_ptr@entry=0x7f6190015000) at ../../../../slurm/src/slurmctld/acct_policy.c:2515 6 0x00005632676767c7 in _restore_job_dependencies () at ../../../../slurm/src/slurmctld/read_config.c:2612 7 read_slurm_conf (recover=recover@entry=1, reconfig=reconfig@entry=true) at ../../../../slurm/src/slurmctld/read_config.c:1310 8 0x0000563267668d0f in _slurm_rpc_reconfigure_controller (msg=msg@entry=0x7f61ad6dade0) at ../../../../slurm/src/slurmctld/proc_req.c:3645 9 0x000056326766f646 in slurmctld_req (msg=0x7f61ad6dade0, arg=0x7f619c000dc0) at ../../../../slurm/src/slurmctld/proc_req.c:425 10 0x00005632675efbc9 in _service_connection (arg=<optimized out>) at ../../../../slurm/src/slurmctld/controller.c:1285 11 0x00007f61b1c386db in start_thread (arg=0x7f61ad6db700) at pthread_create.c:463 12 0x00007f61b196188f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95 Bug 5469. This reverts commit 4b7ad3b6.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Broderick Gardner authored
Bug 5248.
-
- 23 Jul, 2018 5 commits
-
-
Tim Wickberg authored
CID 187110.
-
Tim Wickberg authored
Avoid a double-locking issue by moving the locks previously required by assoc_mgr_clear_used_info() up in the calling path to _restore_job_dependencies(). Add commented out lock annotations to use later.
-
Tim Wickberg authored
Nothing ever checks this return code anyways.
-
Tim Wickberg authored
This also cleans up several locations that could try to repeatedly call close(). See prior commit for further details on why that is best avoided.
-
Tim Wickberg authored
Quoting part of the close() man page: Retrying the close() after a failure return is the wrong thing to do, since this may cause a reused file descriptor from another thread to be closed. This can occur because the Linux kernel always releases the file descriptor early in the close operation, freeing it for reuse; the steps that may return an error, such as flushing data to the filesystem or device, occur only later in the close operation.
-
- 21 Jul, 2018 4 commits
-
-
Tim Wickberg authored
Set DISPLAY to SLURM_X11_SETUP_FAILED to make it clear that the tunnel setup has failed. This at least gives the user a hint as to why their X11 apps aren't working, although further refinement should be done later: tim@zoidberg:~$ srun --x11 xclock Error: Can't open display: SLURM_X11_SETUP_FAILED srun: error: node001: task 0: Exited with exit code 1
-
Tim Wickberg authored
Creates a local XAUTHORITY file in TmpFS on the node, and deletes it upon job termination. This avoids file locking contention on ~/.Xauthority in the users home directory. Bug 3647.
-
Tim Wickberg authored
-
Tim Wickberg authored
Build out sufficient plumbing such that a temporary XAUTHORITY file can be used that is local to the compute node, thus avoiding lock contention on ~/.Xauthority on parallel filesystems. This commit only includes the requisite plumbing to pass this around. If this is not used, a null string results, and the XAUTHORITY env var will not be forced into the user environment. Add support and fix the modified API call in pam_slurm_adopt while here. Bug 3647.
-
- 20 Jul, 2018 3 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Tim Wickberg authored
-
- 19 Jul, 2018 12 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
Update slurm.spec and slurm.spec-legacy as well.
-
Tim Wickberg authored
-
Tim Wickberg authored
The lower limit of 1024 may be too short for srun with large-scale jobs, and lead to problems processing task completion messages in a timely fashion. Rather than adjust that, unify the two separate macros into SLURM_DEFAULT_LISTEN_BACKLOG with the higer 4096 value. Bug 5164.
-
Tim Wickberg authored
Without Delegate=yes, systemd will "fix" the cgroup hierarchies whenever 'systemctl daemon-reload' is called, which will then remove any restrictions placed on memory or device access for a given job. This is a problem especially since 'systemctl daemon-reload' may be called automatically by rpm/yum or a variety of config file mangers, leading to jobs escaping from slurmd/slurmstepd's control. This setting should work for systemd versions >= 205. https://www.freedesktop.org/wiki/Software/systemd/ControlGroupInterface/ Bug 5292.
-
Morris Jette authored
-
Morris Jette authored
addresses problem reported by clang
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Morris Jette authored
bug introduced in commit a7d9313d
-
Felip Moll authored
When a job with time_end=0 and TRES null exists from an association that is currently inside a reservation, the hourly rollup segfaults. Bug 5143
-
Tim Wickberg authored
And from underlying slurm_msg_sendto_timeout call as well.
-
- 18 Jul, 2018 11 commits
-
-
Morris Jette authored
Add function to clear total_gres at start of scheduling cycle Modify logic to avoid overflow on gpu counter
-
Dominik Bartkiewicz authored
As reported by Avalon Johnson on slurm-users https://groups.google.com/forum/#!topic/slurm-users/BsMQ7Uk1PLw Bug 5287.
-
Alejandro Sanchez authored
bug 4373, comment #24
-
Felip Moll authored
Removed the sentence which incorrectly stated that when not using the gres flag enforce-binding option, cpus other than the ones defined in gres.conf could be used for a gpu. Bug 5189
-
Brian Christiansen authored
-
Brian Christiansen authored
srun was already fixed in b7053bda (Bug 3294). Bug 5126
-
Brian Christiansen authored
-
Felip Moll authored
bug 5189
-
Morris Jette authored
Add salloc/sbatch/srun option of --gres-flags=disable-binding to disable filtering of CPUs with respect to generic resource locality. This option is currently required to use more CPUs than are bound to a GRES (i.e. if a GPU is bound to the CPUs on one socket, but resources on more than one socket are required to run the job). This option may permit a job to be allocated resources sooner than otherwise possible, but may result in lower job performance. bug 5189
-
Tim Wickberg authored
-
Broderick Gardner authored
'have_innodb' is deprecated. Bug 5317.
-