-
Notifications
You must be signed in to change notification settings - Fork 656
/
RELEASE_NOTES_LLNL
37 lines (33 loc) · 1.91 KB
/
RELEASE_NOTES_LLNL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
LLNL CHAOS-SPECIFIC RELEASE NOTES FOR SLURM VERSION 2.1
16 October 2009
This lists only the most significant changes from SLURM v2.0 to v2.1
with respect to Chaos systems. See the file RELEASE_NOTES for other
changes.
For system administrators:
* The pam_slurm Pluggable Authentication Module for SLURM previously
distributed separately has been moved within the main SLURM distribution
and is packaged as a separate RPM.
* Added command "sacctmgr show problems" to display problems in the accounting
database (e.g. accounts with no users, users with no UID, etc.).
* Completely disable logging of sched/wiki and sched/wiki2 (Maui & Moab)
message traffic unless DebugFlag=Wiki is configured.
Mostly for users:
* Added -"-signal=<int>@<time>" option to salloc, sbatch and srun commands to
notify programs before reaching the end of their time limit.
* Added a --detail option to "scontrol show job" to display the cpu/memory
allocation informaton on a node-by-node basis.
* Add new job wait reason, ReqNodeNotAvail: Required node is not available
(down or drained).
* Added environment variable support to sattach, salloc, sbatch and srun
to permit user control over exit codes so application exit codes can be
distiguished from those generated by SLURM. SLURM_EXIT_ERROR specifies the
exit code when a SLURM error occurs. SLURM_EXIT_IMMEDIATE specifies the
exit code when the --immediate option is specified and resources are not
available. Any other non-zero exit code would be that of the application
run by SLURM.
SLURM state files in version 2.1 are different from those of version 2.1.
After installing SLURM version 2.1, plan to restart without preserving
jobs or other state information. While SLURM version 2.0 is still running,
cancel all pending and running jobs (e.g.
"scancel --state=pending; scancel --state=running"). Then stop and restart
daemons with the "-c" option or use "/etc/init.d/slurm startclean".