Commits · 056e0bb820f8e30942eef09e4d63b4dad4471820 · Manuel G. Marciani / ces_slurm_simulator

26 Apr, 2019 7 commits
- Hide archive load query log message also behind DB_QUERY. · 056e0bb8
  Alejandro Sanchez authored Apr 26, 2019
```
Bug 6832.
```
  056e0bb8
- Add debug logging at the end of an archive load attempt. · 380e935e
  Nate Rini authored Apr 26, 2019
```
Bug 6832.
```
  380e935e
- Assert that archive loading is always done with autocommit=0. · a10fc71c
  Nate Rini authored Apr 26, 2019
```
Bug 6832.
```
  a10fc71c
- Use goto cleanup instead of repeating cleanup code. · 650a91a9
  Nate Rini authored Apr 26, 2019
```
No functional change.

Bug 6832.
```
  650a91a9
- Limit records per single SQL statement when loading archived data. · 34e9d41b
  Nate Rini authored Apr 26, 2019
```
Otherwise, we could send communication packets bigger than max_allowed_packet.

Bug 6832.

Co-authored-by: Tim Wickberg <tim@schedmd.com>
```
  34e9d41b
- accounting_storage/mysql - fix memory leak in the archive load logic. · e8567e06
  Alejandro Sanchez authored Apr 26, 2019
```
Regression introduced in 8d643e79.

Bug 6832.
```
  e8567e06
- accounting_storage/mysql - fix SIGABRT in the archive load logic. · e174e135
  Nate Rini authored Apr 26, 2019
```
The problem was freeing an interior pointer to buffer contents before
the call to FREE_NULL_BUFFER. The issue was only triggered when loading
an archived data with protocol version < 17.11.

Regression introduced in 8d643e79.

Bug 6832.
```
  e174e135
24 Apr, 2019 4 commits
- Fix issue with backfill scheduler scheduling tasks of an array · 70d12f07
  Moe Jette authored Apr 24, 2019
```
when not the head job.

Bug 6837

For more in depth explanation see comment 24
```
  70d12f07
- Improve test12.10 for testing non-eligible jobs · 6015bffe
  Albert Gil authored Apr 22, 2019
```
Bug 6873
```
  6015bffe
- Fix non-eligible jobs with sacct -j and not -s. · b3e46057
  Albert Gil authored Apr 22, 2019
```
When specifying -j and not -s then non-eligible jobs will be shown
by sacct.
Time windows can also be used with -S and -E.
If --state is also used then non-eligible jobs won't be shown,
because non-eligible are not actually PD.

Bug 6873

# Conflicts:
#	NEWS
```
  b3e46057
- Fixed issue with jobs not appearing in sacct after dependency satisfied. · 5e23ae2b
  Ben Roberts authored Apr 23, 2019
```
Bug 6805
```
  5e23ae2b
23 Apr, 2019 5 commits
- Fix potential deadlock with backup slurmctld. · c04a488d
  Danny Auble authored Apr 22, 2019
```
Bug 6898
```
  c04a488d
- Add missing information for test12.10 · 316f0305
  Danny Auble authored Apr 23, 2019
```
Continuation of commit cc153e03
```
  316f0305
- Fix sacct -PD to avoid CA before start jobs. · fadd6763
  Albert Gil authored Apr 22, 2019
```
Cancelled jobs before start have time_start=0 in the DB, but their
time_end!=0 (the cancel time).
Query fixed to handle properly these cases for -s PD.

Bug 6894
```
  fadd6763
- Continuation of last commit, always remove copy of config file. · c4900135
  Danny Auble authored Apr 23, 2019
```
Bug 6878
```
  c4900135
- avoid leaving vestigial config file · 536dd7a5
  Morris Jette authored Apr 23, 2019
```
Test was always leaving around a file named "slurm.conf.orig".
Now delete the file on test success.

bug 6878
```
  536dd7a5
22 Apr, 2019 2 commits
- Improve test21.39 to be more robust. · d4982cd1
  Danny Auble authored Apr 22, 2019
```
Bug 6878
```
  d4982cd1
- Sync "numtask" squeue format option for jobs and steps to "numtasks" · 6f686863
  Ben Roberts authored Apr 16, 2019
```
Brought the option in line with the "numtasks" option you use when
not specifying --step. Preserved backwards compatbility for "numtask"
for steps. Also upated docs and expect test.

Bug 6861
```
  6f686863
21 Apr, 2019 1 commit
- Docs - add dates for SLUG'19. · b8d63454
  Tim Wickberg authored Apr 21, 2019
  
  b8d63454
18 Apr, 2019 5 commits
- Continuation of last commit... · 31e297b9
  Dominik Bartkiewicz authored Apr 13, 2019
```
Properly initialize structures throughout Slurm.

Bug 6613
```
  31e297b9
- Properly initialize a few structures in the api. · 4728d0e8
  Danny Auble authored Mar 27, 2019
```
Bug 6613
```
  4728d0e8
- Correctly typecast signals being sent through the api. · 7fa46453
  Dominik Bartkiewicz authored Mar 27, 2019
```
Bug 6613
```
  7fa46453
- Restore '-T ALL' functionality in sreport. · 4b7113da
  Tim Wickberg authored Apr 18, 2019
```
Regression from aca37654

.

Bug 6826.

Co-authored-by: Chad Vizino <chad@schedmd.com>
```
  4b7113da
- Fix 'combinations' typo. · 60c84299
  Tim Wickberg authored Apr 17, 2019
  
  60c84299
17 Apr, 2019 3 commits
- Continuation of 4c48a84a. · 37bb5897
  Danny Auble authored Apr 17, 2019
```
Wrong variable was used.  This works fine for 18.08, but in 19.05 this
code doesn't work correctly.

Bug 6739
```
  37bb5897
- Fixed typos in documentation · 35c83f34
  Ben Roberts authored Apr 16, 2019
```
Bug 6865
```
  35c83f34
- Update multi-cluster html · 4d6af92d
  Brian Christiansen authored Apr 16, 2019
```
Bug 6313
```
  4d6af92d
16 Apr, 2019 11 commits
- Add test21.39 · 0bd1ee9b
  Nate Rini authored Mar 17, 2019
```
Bug 6625.
```
  0bd1ee9b
- Better debug when uncomments code · 01998895
  Danny Auble authored Apr 04, 2019
  
  01998895
- Don't reset a FAIL_QOS or FAIL_ACCOUNT job reason until the qos or account changed. · 39a6d267
  Danny Auble authored Mar 11, 2019
```
These are conditions that need to remain constant until something
changes on the job to reevaluate.

Bug 6625
```
  39a6d267
- When changing a jobs account/qos always make sure you remove the old limits. · a54c5edb
  Danny Auble authored Mar 11, 2019
```
What was happening here is you had to not be >= operator to have the old
limits removed.  This makes it so it always happens.

Bug 6625
```
  a54c5edb
- If QOS is removed on a pending job set that job as ineligible. · fbdd1ccb
  Brian Christiansen authored Mar 15, 2019
```
Bug 6625
```
  fbdd1ccb
- If Association is removed on a pending job set that job as ineligible. · 6fbc4809
  Danny Auble authored Mar 11, 2019
```
Before we went up the tree to the next assoc_ptr.  As we validate an
association on the id as well as the uid the assoc_ptr was eventually
going to become invalid.  Setting it to NULL here solves a bunch of
issues with things later on.

Bug 6625
```
  6fbc4809
- Avoid seg_fault on referencing association without a valid_qos bitmap. · 02a0a901
  Danny Auble authored Mar 11, 2019
```
Bug 6625
```
  02a0a901
- Set state_reason if select_nodes() fails job for QOS or Account. · 133d9076
  Nathan Rini authored Mar 11, 2019
```
Bug 6625.
```
  133d9076
- Don't abort when the job doesn't have an association that was removed before... · ecab8d35
  Danny Auble authored Mar 11, 2019
```
Don't abort when the job doesn't have an association that was removed before the job was able to make it to the database.

Bug 6625
```
  ecab8d35
- Verify job is pending add/removing accrual time · 4d946468
  Brian Christiansen authored Mar 15, 2019
```
Bug 6625
```
  4d946468
- This appears to be a missing xfree. · c8356cbf
  Danny Auble authored Mar 26, 2019
  
  c8356cbf
13 Apr, 2019 2 commits

Don't purge jobs if backfill is running. · 426abc7f

Marshall Garey authored Apr 03, 2019

The backfill scheduler keeps a local list of job pointers. Since the
backfill scheduler yields locks, it's possible for pending jobs to be
canceled and purged in these yield periods. The backfill scheduler then
has pointers to now invalid memory, and dereferencing those pointers is
undefined behavior and may result in a segfault.

This commit prevents purging jobs while the backfill scheduler is
running.

Bug 6621

426abc7f

Add mutex to protect slurmctld_diag_stats.bf_active. · e703ba97

Danny Auble authored Apr 03, 2019

The next patch will use slurmctld_diag_stats.bf_active to determine
if we can purge jobs or not.

Bug 6621

e703ba97