GitLab

HBMCACHEPCT explicitly named and no longer operating on a node list,
goiong straight to nid list.  Should save memory/time.

Adding in a better validtor to prevent issues with users typing bad
NUMA/MCDRAM modes.  Also, adding a default setting if none provided.

The original ticket for this task was too broad.  This only handles
adding the Task Start, Task End messages and migrating the run timer
for determining when walltime runs out.

Resolve "Enhancement: Log extra events from Cobalt"

Closes #65

See merge request !30

We need to log and end the task timers on both __rc_unknown and
__rc_success on __task_finalize.  The only outcome that doesn't
officially end a task is __rc_retry, as it will go into communication
retries at that point.

Cobalt now gets updates on NUMA and MCDRAM mode changes.

Closes #66

See merge request !31

Only set the timers/emit messages once, despite retry attempts.

start.

Resolve "Integrate gronkd into Cobalt"

Closes #49

See merge request !28

Macros are a bit different and SLES 12 can leverage systemd

Resolve "Increase size of job_data.exit_status to integer"

Closes #61

See merge request !27

Resolve "Flip bgqsystem backfiller sign"

See merge request !23

There was a change in the call and a behavior change to not try to
redrain blocks.  This was what the originally intended behavior was.

Uncovered another bug while working on this, old behavior masked it, but
there was a way to get a block to ignore the scheduled flag as well.

Fix for nodes getting hung up in cleanup-pending state

A well (or poorly depending on how you look at it) qdel could cause
Cobalt to put a node into cleanup but never complete the cleanup due to
there being no ALPS backend reservation to clean up.  This would clear
if there were no jobs currently running, however, it would hang nodes
otherwise.

Closes #56

See merge request !26

A well (or poorly depending on how you look at it) qdel could cause
Cobalt to put a node into cleanup but never complete the cleanup due to
there being no ALPS backend reservation to clean up.  This would clear
if there were no jobs currently running, however, it would hang nodes
otherwise.

Adding in build files for the RESERVATION_SUMMARY view for CDB

Adding in templates and build script for the reservation summary view from Gabe West.
Closes #52

See merge request !24

Qsub path can now be specified for eLogin qsubs.

This prevents us from getting a unwrapped qsub.py or the qsub being in a
different location on the mom from the eLogin host when using qsub -I
from a eLogin node on Cray systems.

Closes #54

See merge request !25

This prevents us from getting a unwrapped qsub.py or the qsub being in a
different location on the mom from the eLogin host when using qsub -I
from a eLogin node on Cray systems.