processing job dependencies

    get purged before their trigger can be processed (e.g. MinJobAge=1).

due to very low MinJobAge configuration).

    RLIMIT_STACK.

    job at submission time, but to attempt scheduling many jobs at once for
    improved performance under heavy load.

    than one port for improved scalability.

    reservations.

Fixed issue where if the slurmdbd was down past the MinJobAge and the start of the job hadn't been sent yet to throw it on the queue to be sent when the slurmdbd comes back up.

preemptive strike before 2.1 merge which will do this exact thing, but the code has changed locations so many conflicts will happen

    that of the slurmctld daemon: the same directory used for log files,
    SlurmdLogFile (if specified with an absolute pathname) otherwise the
    directory used to save state, SlurmdSpoolDir.

    affinity is enabled, keep track of the individual CPUs allocated to jobs
    rather than just the count of CPUs allocated (which could overcommit
    specific CPUs for running jobs).

meaningful if StateSaveLocation=/dev/null

Fix for handling correctly jobs that resize, and also reporting correct stats on a job after it finishes.

cleaner and tighter code

being freed by sched/backfill.