The thread logic still needs to be fleshed out

There was a double lock (deadlock) in the power management plugin driver
The power and burst buffer plugins needed to be restarted whenever the primary slurmctld resumed control rather than once at daemon start time. Otherwise the backup and primary slurmctld could both try to manage power and burst buffer resources.

Add module for inter-cluster job development work

Was using "%d" rather than "%u" to print job ID. printing negative
job id value for inter-cluster jobs.

Conditionally restart daemons after RPM package upgrade.

uid's are 0 when associations are loaded.

The backfill scheduler build a queue of eligible job/partition
information and then proceeds to determine when and where those
jobs will start. The backfill scheduler can be configured to
periodically release locks in order to let other operations
take place. If the partition(s) associated with one of those
jobs changes during one of those periods, the job will still
be considered for scheduling in the old partition until the
backfill scheduler starts over with a new job/partition list.
This change to the backfill scheduler validates each job's
partition in from the list based upon current information
(considering any partition changes).
See bug 1436

If bitmap size is initially NO_VAL 0xfffffffe, then a tiny buffer
is allocated and accessing it can go off the end of the buffer.
This has not been observed in production, but only in the investigation
of another problem.

 Fix segfault in controller when deleting a user association of a user which had been previously removed from the system.

 Bug 1238

Fix slurmctld initialization problem which could cause requeue of the last
task in a job array to fail if executed prior to the slurmctld loading
the maximum size of a job array into a variable in the job_mgr.c module.

Fix slurmctld job recovery logic which could cause the last task in a job
array to be lost on restart.

Only supported with task/affinity plugin.

When CgroupMountpoint was not defined in cgroup.conf the mount point
got undefined. This resulted in cgroups not being released.

In order to support inter-cluster job dependencies, the MaxJobID
configuration parameter default value has been reduced from 4,294,901,760
to 2,147,418,112 and it's maximum value is now 2,147,463,647.
ANY JOBS WITH A JOB ID ABOVE 2,147,463,647 WILL BE PURGED WHEN SLURM IS
UPGRADED FROM AN OLDER VERSION!

Add PowerParameter and user options to level power cap for all
 jobs associated with a parallel job. Per-job option can be
 disabled on a global basis or job power leveling can be forced
 for all jobs.

Add the sicp_mode and power_flags to the slurmctld job record.
Save and restore the new fields.
Pack them in the job_info structure and view them using "scontrol show job"

 -- Add job submission command options: --sicp (available for inter-cluster
    dependencies) and --power (specify power management options) to salloc,
    sbatch, and srun commands.
 -- Add DebugFlags option of SICP (inter-cluster option logging).
 -- Added DebugFlags value "SICP".
 -- Added job_descriptor field of "cluster".

footprint of a job. Add an option not to kill jobs that is over memory
limit.

Use %__make in slum.spec

Fix a couple of dead assignments and uninitialized variable issues reported by CLANG

event  REQUEUED to slurmdbd.

Related to bug 1429

Improve "Prolog and Epilog Scripts" in slurm.conf(5)

The environment variable name SLURM_JOB_CLUSTER_NAME should be
SLURM_CLUSTER_NAME. This is also available in Prolog and Epilog, so
remove note about it only being available in PrologSlurmctld and
EpilogSlurmctld.

Previously it was not possible to distinguish between a job needing
exclusive nodes and the default job/partition configuration.