If the slurm.conf changes to remove nodes from a job step in the
process of running Node Health Check, the step record would be
purged. Then when the NHC completes, it's step record pointer
would be invalid.
The change here will prevent such a job step record from being
purged.
This was found while investigating bug 612, but would be a very
unusual condition and does not fix this bug.

always be set to SLURM_SUCCESS since it isn't used before.

If the switch/cray global state save/restore fails, avoid re-allocating
ports based upon the job steps that we can recover

Previously slurmctld would restart without information about
allocated ports. This add save/restore logic which will work
for graceful shutdowns. More work is needed in the case of
slurmctld crashes.

The merge from v2.6 to master in commit 18985643
seems to have gone horribly wrong. This should fix it.

bug 607

Make job array expressions more flexible to accept multiple step counts in
the expression (e.g. "--array=1-10:2,50-60:5,123").

If a partition was configured with Shared=yes, then submitting a
job with --core-spec option failed to allocate the node exclusively
to that job.

No changes in logic

placed.

Conflicts:
	src/plugins/select/bluegene/select_bluegene.c
	src/plugins/select/linear/select_linear.c

The select_p_resv_test() functions for the select/linear and
select/bluegene plugins were not updated for specifying core
counts per node. The additional argument was not used, but
now the functions are consistent across all of the select
plugins.

Add resrvation "flags" field to select_p_resv_test() and
select_g_resv_test() functions.
Also found errors in how those functions are referenced in
select/linear and select/bluegene

Underlying code incomplete. Only added the flag.