Conflicts:
	src/slurmctld/job_mgr.c

lock was deemed not necessary because the information (db_index) was only
internal and was only modified in the same function later which is
protected by the write lock.

qos on the cluster to fill that in correctly.

incorrectly when getting fairshare from the parent.

Patches from Andriy Grytsenko (Massive Solutions Limited).

Cross cluster communication doesn't work with srun.

steps in the allocation if --no-kill is used on the allocation.

partial of each node.

ignoring all nodeboards in error state.

Moe,

I found that I need the following lines added in
order to allow LLONG_MIN and LLONG_MAX to be used
in the range routines.

Phil

work.

doesn't fill up all the cpus on the nodes requested.

sub block.

Add ReconfigFlags configuration option to control actions of
"scontrol reconfig". Patch from Don Albert, Bull.

Conflicts:
	NEWS

wasn't started at the time the eligible message was sent but started
before the db_index was returned information like start time would be lost.

cpu count instead of the block count.  Needed when running multiple jobs
per block.

which may not exist in all cases.