- 08 Dec, 2016 9 commits
-
-
Morris Jette authored
-
Morris Jette authored
Revise commit e6ef1f0c The root cause of the failure is a bug in HWLOC that has since been fixed in HWLOC version 1.11.5.
-
Tim Wickberg authored
-
Tim Wickberg authored
If the second call to getgrouplist() found additional groups, ngroups will be overwritten with this new larger value, while the gids list would be truncated. (ngroups is a value-result arg.) This will then lead to _gids_cache_lookup() returning the wrong number of groups including invalid parts of memory, which are likely to include some zeros. Those zeros could then make it to the setgroups() call and thus give the user access to the root group. Especially as setgroups will succeed as long as the array does not contain -1 as a gid. Bug 3320.
-
Tim Wickberg authored
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
task/cgroup - Change error message if CPU binding can not take place to better identify the root cause of the problem. Specifically, if the hwloc_get_obj_below_by_type() function call completely fails that is likely due to task/affinity not being configured, so cpusets are not configured. Previous message was "task/cgroup: task[%u] infinite loop broken while trying to provision compute elements using %s (bitmap:%s)" The new message is "task/cgroup: hwloc_get_obj_below_by_type() failing, task/affinity plugin also required"
-
Dominik Bartkiewicz authored
uint32_t needs %u on 32-bit platforms. Noticed by clang/travisci.
-
- 07 Dec, 2016 9 commits
-
-
Morris Jette authored
Test would fail if more than one task per core and SelectTypeParamers included CR_ONE_TASK_PER_CORE
-
Danny Auble authored
Bug 3258
-
Danny Auble authored
This reverts commit 55cb7973.
-
Danny Auble authored
This reverts commit 817c2ca4. # Conflicts: # NEWS
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Danny Auble authored
-
- 06 Dec, 2016 22 commits
-
-
Morris Jette authored
-
Morris Jette authored
The test is still failing for me (along with a bunch of others), but this change at least gets the test started.
-
Morris Jette authored
-
Danny Auble authored
a slurmctld restart or reconfig, as they aren't really error messages. Bug 3258
-
Danny Auble authored
Bug 3258
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
Done jost to run "git push" again after internal github error on previous push: remote: Resolving deltas: 100% (4/4), completed with 4 local objects. remote: Unexpected system error after push was received. remote: These changes may not be reflected on github.com! remote: Your unique error code: bdecb7b0f321368fe1f037a81a6e9c2c
-
Morris Jette authored
This restores the socket count check at node registration for non-KNL systems (at least systems without NodeFeaturesPlugins type that includes "knl"). This is a refinement of commit 1ce9a7c4
-
Tim Wickberg authored
-
Tim Wickberg authored
Note that this does not protect against all possible problems here. The setgroups() call in Linux at least is willing to set any gid_t value except -1 on a group, so calls will not always fail on corrupted group lists. Bug 3320.
-
Tim Wickberg authored
-
Tim Wickberg authored
Remove uncached _get_grouplist() call which was only used here. Bug 3315.
-
Morris Jette authored
-
Morris Jette authored
test12.2 was consistently failing on smd# cluster with tiny differences in the disk read and written. This change permits those tiny discrepancies to exist without failing the test. Here are the numbers, which are consistent: sacct --noheader -p --job=763.0 --format MaxDiskWrite,AveDiskWrite,MaxDiskRead,AveDiskRead 10.00M|10.00M|10.03M|10.03M| (i.e. 0.3% discrepancy, up to 0.5% allowed with current code)
-
Morris Jette authored
Test was failing due to hitting the memory limit (at least on smd1).
-
Morris Jette authored
There were already several configurations that could cause this test to fail. I just added another.
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Morris Jette authored
Fix parsing in regression test1.92 for some prompts. bug 2792
-
Morris Jette authored
Recognize a KNL's proper NUMA count (rather than setting it to the value in slurm.conf) when using FastSchedule=0. Previous logic would change the NUMA count on the node to match what was in slurm.conf, which would mess up task layout with respect to the sockets. bug 3306
-