-> 2.6 slurmd

This makes it consistent with the value of default_queue_depth.
The backfill scheduler should be able to easily handle this value
(or much higher for pretty much any configuration).

Conflicts:
	src/common/stepd_api.c

Conflicts:
	src/common/read_config.c
	src/plugins/accounting_storage/pgsql/accounting_storage_pgsql.c
	src/plugins/jobcomp/pgsql/jobcomp_pgsql.c
	src/slurmctld/job_mgr.c

could end up before the job started. Bug 371

This function can now be called to test for processes which are
dumping in order to avoid sending them a SIGKILL until dump
completes. Change in logic required for job_container/cray.

Handle invalid process ID better (probably never happens)
Re-arrange logic to eliminate duplicate file close on error

plugins more maintainable.

We now reject jobs with an invalid accounting frequency at
submit time rather than launch time, so the error is slightly
different and the test needs to change for that.

This avoids waiting for the job's initiation to fail.

Only cancel the job.

Fix job state recovery logic in which a job's accounting frequency was
not set. This would result in a value of 65534 seconds being used (the
equivalent of NO_VAL in uint16_t), which could result in the job being
requeued or aborted.

Problem reported by BYU. slurm.conf included a file one byte in
length. Logic created a buffer one byte long and used fgets()
to read the file. fgets() reads one byte less than the buffer
size to include a trailing '\0', so it fails to read the file.

Basically the system size has to be set up before you call the
priority/multifactor plugin.  If a job is finishing while the slurmctld
is starting then it would fatal on the init if it wasn't set up.

core reservations and reservation prolog/epilog