If a node or partition name is duplicated in slurm.conf, then log
it using error() and ignore the duplicate record. Update documentation
to be more clear

starts.

SLURM wouldn't create another overlapping block on a job requesting that
size of block.

Previously the code would come up with how much memory a PE should have
instead of the memory a node should have.

Mark Nelson.

incorrectly set in the job.

that go into error the block is freed.

start a job on the block.  Before there was room for a block boot to fail
if another thread was trying to free it while trying to start a job.

it is happening, just to keep tabs correctly on the matter.

dynamic mode.

job_ptr.

job on a block and it isn't being configured or booted we fixed a double
free issue.

catch, and fix, the problem when the next job comes through to use it.

larger than midplane jobs.

This makes it easier to handle failover, if a slurmctld keeps failing
(slurmdbd gets shutdown or something) this way we don't continually send
a message over and over again in this senario, at least it drops down
how many messages we send over and over again.

Add support for new SchedulerParameters of bf_max_job_user, maximum number
of jobs to attempt backfilling per user. Work by Bjørn-Helge Mevik,
University of Oslo.

Fix sched/wiki2 to support job account name, gres, partition name, wckey,
or working directory that contains "#" (a job record separator). Without
this patch, the parsing will probably stop once reaching the "#".