What this does is set the state earlier to match a normal set.

Remove the unneeded _send_pending_exit_msgs.  There is only one task and
we have the message for it, so don't worry about that one.

Most important, wait for the other slurmstepd's to send their message,
otherwise they could be lost on the other end.

Fix gang scheduling and license release logic if single node job killed on
    bad node. Notifying gang and releasing licences is normally done when
    the epilog completion happens, but if the node(s) assigned to a job are
    all down, that does not happen. This results in the licenses being
    reserved indefinitely and the gang scheduler being left with a bad
    (old) job pointer that can result in various failure modes
bug 2867

Add hotels. Other minor changes.

anyway to attempt to log the backtraces of the potential
unkillable processes.

667f1105.

We have decided to go back to the way 15.08 called NHC instead of calling
it first before sending a SIGKILL to the steps tasks.  With this patch we
only start the NHC early when we have to resend the SIGKILL for unkillable
processes.  This will hopefully get us the backtrace of the unkillable
processes which was the reason we did it this way in the first place :).

processes.

Bug 2892.

Bug 2874

We will most likely redo this logic (as it appears to be duplicated) in
a following patch.

Don't generate an error when a batch job is submitted that must wait
  for stage-in before starting.

Bug 2886

Conflicts:
	src/sstat/options.c

Was incorrectly translating request to job.extern if part of a
comma-separate list.

Bug 2890.

(regression in 16.05.2).

related commit 5d3e5e1e

Bug 2612 and 2886

'-' without a '\' in front of it.

Document limitations in burst buffer use by the salloc command (possible
    access problems from a login node).
bug 2883

task/cgroup plugin is configured with ConstrainRAMSpace=yes, then set soft
    memory limit to allocated memory limit (previously no soft limit was set).
bug 2679

of 0.

This might be the cause of run away jobs.  I couldn't see how an end_time
could be 0, but if it was it would just exit and never set time_end to
anything.  At least if it happens now we can have an idea that it is
possible and we will have an idea this is the place it happens.

This will keep from referencing the task array that might not be set up
correctly in src/common/plugstack.c _spank_handle_init().

Prevent possible incorrect counting of GRES of a given type if a node has
    the multiple "types" of a given GRES "name", which could over-subscribe
    GRES of a given type.
bug 2836

The wrong variable was being used in gres.c to print a message of
this sort:
error: gres/gpu: job 7135 dealloc node ### type ### gres count underflow (# #)
The first number in the parenthesis was based upon the wrong index and it's
contents garbage

cleaning up on a restart.

sleep is killed.

for steps.

for a step.

hostlists when brackets == 0.