- 07 May, 2015 21 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
"operator" will break the build. So renaming it makes things work correctly.
-
Danny Auble authored
-
Morris Jette authored
Original logic lacked check for NULL list pointer
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
-
=Veronique Legrand authored
-
Nicolas Joly authored
-
Morris Jette authored
The previous logic could deadlock with the log functions pthead_atfork() functions depending upon the call order.
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
Apply the limit to SPANK plugins. Avoid invalid memory reference is slurmdbd calls slurm_get_prolog_timeout() Updates to commit 659ae598
-
Danny Auble authored
-
- 06 May, 2015 10 commits
-
-
Morris Jette authored
There were many calls to slurm_list_iterator_create() without the matching slurm_list_iterator_destroy(), which would result in memory leaks.
-
Morris Jette authored
-
Morris Jette authored
cancelling job ID that could not be found was treated like an error when it should just return success.
-
Morris Jette authored
Add re-entrant versions of glibc time functions (e.g. localtime) to Slurm in order to eliminate rare deadlock of slurmstepd fork and exec calls. bug 1638
-
Morris Jette authored
-
Danny Auble authored
14.11 worked fine, but there were changes in 15.08 that weren't addressed in the original patch.
-
Morris Jette authored
-
Danny Auble authored
utilization.
-
Danny Auble authored
random crashing in db2 when the slurmctld is exiting. Signed-off-by:
Danny Auble <da@schedmd.com>
-
David Bigagli authored
-
- 05 May, 2015 9 commits
-
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Dead assignments, possible zero divide, and wrong data types
-
David Bigagli authored
Minor typo fixes
-
Christopher Bottoms authored
-
Morris Jette authored
Modify all tests to use cancel_job() rather than "exec scancel ..." so that we can do better error handling all in one place. There were several places in the globals module that printed "FAILURE" and set a local "exit_code" variable. I added "global exit_code" to those functions in hopes of those failures being reflected in the test's exit code.
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Conflicts: src/plugins/select/cons_res/job_test.c
-