Commits · 81a8178addf173d752411cecf11f668807e1ddc0 · Manuel G. Marciani / ces_slurm_simulator

17 Feb, 2017 1 commit
- Fix typos in acct_policy strick -> strict, no code change · 81a8178a
  Danny Auble authored Feb 17, 2017
  
  81a8178a
16 Feb, 2017 6 commits
- Fix correct state reason when job can't run 'safely' because of an · fad27852
  Josh Samuelson authored Feb 16, 2017
```
association GrpWall limit.
```
  fad27852
- Better debug output when a job is being held because of a GrpTRES[Run]Min · 325d674a
  Danny Auble authored Feb 16, 2017
```
limits.
```
  325d674a
- Continuation of last patch for associations. (same mistake) · 37be42ec
  Danny Auble authored Feb 16, 2017
```
Bug 3476
```
  37be42ec
- Fix correct variables when validating GrpTresMins on a QOS. · 92d2c645
  Josh Samuelson authored Feb 16, 2017
```
Bug 3476
```
  92d2c645
- Fix comments in acct_policy.c to reflect actual variables instead of · 4cfe6bde
  Danny Auble authored Feb 16, 2017
```
old ones.

This is cosmetic only, no code change.

Bug 3476
```
  4cfe6bde
- Doc - replace vestigial --shared by --oversubscribe. · c72a6497
  Alejandro Sanchez authored Feb 16, 2017
  
  c72a6497
15 Feb, 2017 2 commits
- Fix squeue when SLURM_BITSTR_LEN=0 is set in the user environment. · 0ea581a7
  Danny Auble authored Feb 15, 2017
```
Bug 3472
```
  0ea581a7
- Prevent deadlocked slurmstepd processes due to unsafe use of regcomp. · fe193906
  Tim Wickberg authored Feb 15, 2017
```
regcomp() is not safe to use across a fork in older glibc versions.
Reinitialize the keyvalue_re structure after the fork through an atfork()
handler.

Bug 3276.
```
  fe193906
14 Feb, 2017 5 commits
- Fix for jobs with constraint counts · 28a8dec1
  Morris Jette authored Feb 14, 2017
```
Honor --ntasks-per-node and --ntasks option when used with job constraints
    that contain node counts.
bug 3458
```
  28a8dec1
- Fix task/affinity to work correctly with --ntasks-per-socket. · ab9430d5
  Danny Auble authored Feb 14, 2017
  
  ab9430d5
- Docs - fix link to mcs_plugins page. · c825c0b3
  Tim Wickberg authored Feb 14, 2017
  
  c825c0b3
- Defer interactive job allocation until nodes ready · b988531d
  Morris Jette authored Feb 14, 2017
```
Defer interactive job allocation until ALL allocated nodes are ready
  rather than after PrologSlurmctld (if any) completes.
```
  b988531d
- scontrol - return error on unsupported +=/-= syntax when updating fields. · cf9844fe
  Dominik Bartkiewicz authored Feb 14, 2017
```
Bug 3467.
```
  cf9844fe
13 Feb, 2017 3 commits
- Fix burst_buffer/cray race condition · bd7504fb
  Morris Jette authored Feb 13, 2017
```
burst_buffer/cray - Do not execute "pre_run" operation until after all nodes
    are booted and ready for use.
bug 3461
```
  bd7504fb
- Fix minor memory leak in the slurmctld when removing a QOS. · 9984400a
  Danny Auble authored Feb 13, 2017
  
  9984400a
- Don't start job early · f6d42fdb
  Morris Jette authored Feb 13, 2017
```
Insure job does not start running before node is booted and PrologSlurmctld
    is complete.
bug 3446
```
  f6d42fdb
10 Feb, 2017 7 commits
- Protecting `volatile` variables to make Valgrind's drd happy. · 595b24f1
  Artem Polyakov authored Feb 10, 2017
```
Ported from 7a4aa7f2.
```
  595b24f1
- PMI2 - Make it possible to use %n or %h in a spool dir. · ef03f126
  Danny Auble authored Feb 10, 2017
  
  ef03f126
- Minor formatting changes to PMIX · d127b762
  Danny Auble authored Feb 10, 2017
  
  d127b762
- Make it so the pmix plugin sends the unformatted string when forwarding data, · bb123d94
  Danny Auble authored Feb 10, 2017
```
and make it so the slurmd formats it correctly when it comes in.

Continuation of commit 7db90c8a
```
  bb123d94
- Make it so tmpdir processes %n and %h correctly. · e8c2bfc5
  Danny Auble authored Feb 10, 2017
```
This is needed to make it so we can run pmix with multi-slurmd.
```
  e8c2bfc5
- [PATCH] Doc - Fix slurmd "-b" option in man page. · a87ee41a
  Alejandro Sanchez authored Feb 10, 2017
```
Bug 3446
```
  a87ee41a
- Improve power save documentation · e848a945
  Morris Jette authored Feb 10, 2017
```
bug 3446
```
  e848a945
09 Feb, 2017 6 commits
- burst_buffer/cray - DataWarp parsing fix · b9d05100
  Morris Jette authored Feb 09, 2017
```
burst_buffer/cray - Support default pool which is not the first pool
    reported by DataWarp and log in Slurm when pools that are added or removed
    from DataWarp.
bug 3453
```
  b9d05100
- Commit for last patch · 018314a6
  Danny Auble authored Feb 08, 2017
  
  018314a6
- PMIX - Make it possible to use %n or %h in a spool dir. · 7db90c8a
  Danny Auble authored Feb 08, 2017
  
  7db90c8a
- Revert "PMIX - Make it possible to use %n or %h in a spool dir." · 22ba56b9
  Danny Auble authored Feb 08, 2017
```
This reverts commit fd690a9c.
```
  22ba56b9
- PMIX - Make it possible to use %n or %h in a spool dir. · fd690a9c
  Danny Auble authored Feb 08, 2017
  
  fd690a9c
- Use volatile specifier to avoid flag caching. · 84e765d9
  Artem Polyakov authored Feb 08, 2017
  
  84e765d9
08 Feb, 2017 3 commits
- Update sstat man page. · 6287bbce
  Brian Christiansen authored Feb 07, 2017
```
sstat doesn't work on Cray ALPS but works on native Cray setups.
```
  6287bbce
- Fix record job state as PREEMPTED instead of REQUEUED. · 72c0b659
  Alejandro Sanchez authored Feb 07, 2017
```
Jobs preempted with PreemptMode=REQUEUE were incorrectly recorded as
REQUEUED in the accounting.

Bug 3444
```
  72c0b659
- power/cray - Disable power cap get and set operations on DOWN nodes · 08edc116
  Morris Jette authored Feb 07, 2017
```
bug 3448
```
  08edc116
07 Feb, 2017 1 commit
- Print message back to stderr when disabling affinity with --cpu_bind=verbose. · d4a286a4
  Dominik Bartkiewicz authored Feb 03, 2017
```
Bug 3447
```
  d4a286a4
03 Feb, 2017 1 commit
- Record job state as PREEMPTED instead of TIMEOUT when GraceTime is reached. · ba2c5584
  Alejandro Sanchez authored Feb 03, 2017
```
Bug 3444
```
  ba2c5584
31 Jan, 2017 3 commits
- NEWS update for next 16.05 version · b8aecfd0
  Danny Auble authored Jan 31, 2017
  
  b8aecfd0
- Update META for v16.05.9 tag · abfd9995
  Danny Auble authored Jan 31, 2017
  
  abfd9995
- Make sure acct policy limits imposed on a job are correct after requeue. · 5d941217
  Alejandro Sanchez authored Jan 31, 2017
  
  5d941217
30 Jan, 2017 2 commits

Danny Auble authored Jan 30, 2017

e3a7bdcc
f9804256
d72b13f2

Reference bug 3366

If you are running on a Bluegene system we rely on the prolog to take us out of configuring
state.  These commits work good for system rebooting the nodes where the prolog is running,
but in the case of Bluegene this is the opposite desire :).   These commits on a Bluegene
pretty much make it so a batch job never gets launched.

a4c51165

Clear job BeginTime reason · 0abbf727

Morris Jette authored Jan 30, 2017

Clear job's reason of "BeginTime" in a more timely fashion and/or prevents
    them from being stuck in a PENDING state. There are multiple ways of
    clearing the reason, especially on a lightly loaded system, but the
    state can persist indefinitely on a heavily loaded system.
bug 3368

0abbf727