changes

it happen correctly on an L/P system

the block

One when select/cray plugin is loaded (for any system)
One when "scontrol show topology" is executed (leak is slurmctld daemon)

The original logic had a problem if you shrank a job and later grew it.
Nodes previously released would reappear when the job grows, but have
zero CPUs associated with them. The problem was due to the original nodes
list of a job being preserved in the job_resources data structure. The
new logic confirms that those nodes are still in the job's allocation
before rebuilding the job_resources data structure.

file.

more maintainable.

the slurm rpm

UMBC.

Prevent background salloc disconnecting terminal at termination. Patch by Don Albert, Bull.

Update list of authors and Copyright holders.

Conflicts:
	META
	NEWS

In sview, display a node's architecture (arch) field

More than one TaskPlugin can be configured in a comma separated list.
Patch from Andriy Grytsenko (Massive Solutions Limited).

Fix error handling bug in mpi/mvapich plugin that could result in srun going into an infinite loop.

Conflicts:
	NEWS

Avoid clearing a node's Arch, OS, BootTime and SlurmdStartTime when
    "scontrol reconfig" is run. Patch from Martin Perry, Bull.

this adds another instruction to the grant all on slurm_acct_db.*  TO 'slurm'@'system0';  where system0 is the hostname.  Our users have add errors from mysql  as shown below where it needs to be applied for both localhost and the hostname
From Nancy Kritkausky, Bull

Correct several man pages describing what a dpendency of afterok means.

This fixes a segfault that can happen when the user supplies an ill-formatted
SLURM_TIME_FORMAT string. Fixed by always initializing the display_fmt string.

Instead of segfaulting, the response now uses the fallback:
01_BUG-Fix_parse_time_segfault.diff from Gerrit Renker, CSCS.