- 22 Aug, 2018 7 commits
-
-
Danny Auble authored
# Conflicts: # NEWS
-
Danny Auble authored
Bug 5608 Tim approved
-
Brian Christiansen authored
If the dbd comes up after a job array has been submitted to the controller, the controller calls _update_job_tres() which calls assoc_mgr_set_tres_cnt_array() which allocates memory for the job's tres_alloc_cnt. The job array gets scheduled, but job_array_split() doesn't NULL out the pending job's tres_alloc_cnt, so both the array task and the pending array job are pointing to the same memory. The array task calls job_set_alloc_tres() which free's the running job's tres_alloc_cnt and now the pending array job is pointing to bad memory and when the array splits again the new array task tries to free tres_alloc_cnt in job_set_alloc_tres() and segfaults. Bug 5604
-
Tim Wickberg authored
-
Tim Wickberg authored
Update slurm.spec and slurm.spec-legacy as well.
-
Danny Auble authored
-
Danny Auble authored
This will need to be revisited perhaps link to libslurmfull.so instead since we will need symbols not exported. See bug 5605
-
- 21 Aug, 2018 19 commits
-
-
Tim Wickberg authored
-
Danny Auble authored
The previous commit causes erroneous errors to be printed. test2.14 was made a victim of it.
-
Danny Auble authored
# Conflicts: # src/plugins/jobacct_gather/common/common_jag.c
-
Danny Auble authored
-
Alejandro Sanchez authored
Bug 5561.
-
Felip Moll authored
Bug 5538
-
Danny Auble authored
Bug 5538 This will be handled further in bug 5602
-
Isaac Hartung authored
Previous behavior was confusing, and the code path hard to follow. Rework, and clarify the established behavior in the appropriate man pages. Bug 4514.
-
Broderick Gardner authored
X11 implies Contain, and Contain is incompatible with NoHold as the Extern step creation will race with the step launches. Bug 5500.
-
Tim Wickberg authored
This reverts commit dcc93022.
-
Tim Wickberg authored
This reverts commit b9fb67d8.
-
Morris Jette authored
CID 187777. Bug 5593.
-
Tim Wickberg authored
-
Tim Wickberg authored
scan-build complains about this possibly being used uninitialized with pthread_join(), but on review the path it identifies as causing that is impossible. Initialize this anyways in the hopes of surppressing that warning.
-
Tim Wickberg authored
-
Tim Wickberg authored
This also randomizes the backup controller port numbers which were not being handled here for some reason.
-
Dominik Bartkiewicz authored
Bug 5166
-
Tim Wickberg authored
Bug 5426
-
Felip Moll authored
Bug 5072
-
- 20 Aug, 2018 14 commits
-
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
Change _build_path to return the xmalloc()'d string, and make sure to free it where appropriate. Add comment about how the processing of the PATH environment variable here is backwards from convention.
-
Tim Wickberg authored
-
Tim Wickberg authored
Fixes GCC 8.2.0 warnings.
-
Tim Wickberg authored
Fix some -Werror=restrict and -Werror=format= issues.
-
Tim Wickberg authored
-
Tim Wickberg authored
Silence GCC 8.2.0 warning.
-
Tim Wickberg authored
-
Tim Wickberg authored
-
Tim Wickberg authored
Nothing modifies this, so don't bother making a copy of it. Related to GCC 8.2.0 compilation fixes.
-
Tim Wickberg authored
Fixes GCC 8.2.0 compilation warnings. Bug 5465.
-
Tim Wickberg authored
-
Michael Hinton authored
MySQL permits up to 64-character database names, but Slurm was truncating at 33-characters. If we exceed this limit, let the mysql_query fail and give the admin a chance to sort it out, rather than truncating and then failing to query against the un-truncated name later on. While here correct the fatal() message. Bug 5586.
-