Corrected the restart order for nodes and slurmctld when adding nodes.

Bug 7135

Commit 2f571559 added internal calls to mysql_db_commit() in
_archive_purge_table(). Previous sacctmgr logic triggered the operation
and then asked for confirmation through commit_check(), making it
impossible to rollback even when the user discarded the changes.

Bug 7130.

Previous logic initialized a static variable once and forever. This
variable was used as a test to call or not log_job_resources and
dynamic changes through 'scontrol setdebugflags [+|-]SelectType' would
have no effect. New logic uses the standard select_debug_flags as
everywhere else and aligns with cons_tres logic.

Bug 7137.

Commit 2194edf3 was intended to change the default fairshare algorithm
to Fair Tree. There was an oversight with the conversion where Fair Tree
wouldn't be the default if PriorityFlags was commented out.

Bug 7154.

Reflow lines at <= 80 columns when modifying them, and use lower-case
for HTML tags.

Bug 7275.

fix/improve example logrotate configuration in man slurm.conf

pkill without -x option treats argument as a pattern in this case
slurmd matches not only slurmd but also slurmdbd.

Bug 6962

Test rendered invalid by changes made for bug 7128.
bug 7128

bug 7128

bug 7128

bug 7066

bug 7066

i.e.

gres=gpu:k80:4,gpu:tesla:4

the code would put the k80 as one number 4, and then add the tesla and
not remove the 4 but add the 8 so if all your other nodes only have 1
gpu and you are asking for 2 nodes it would return 4 instead of 1.

Bug 7066

This is a continuation of commit edd09b55.

Bug 7071.

Documentation for "-d" was missing from the man page.
Documentation for "-i" was missing from "slurmctld -h".
Sync the description for "-D".

Bug 7245.

Instead of scontrol show daemon for logrotate we can just use list
of all daemons. This should be less confusing for users since slurmdbd
is not listed by scontrol show daemons and required separate logrotate
configuration.

Bug 6962

The init() function creates a separate thread that tries to index the
list of unindexed jobs (loaded from the state or newly finished) into
the elasticsearch server. This fix prevents this thread from trying to
do so until after slurm_jobcomp_set_location sets the log_url.

Bug 7153.

Also added TimeEligible as a field that can be requested.
Bug 7074

If all runaway jobs have submit_time=0, something is wrong, and we
don't know from when to reroll the usage tables.
Bug 7074

When fixing runaway jobs, submit time should always be set in the
database, so use that when rerolling usage. Avoids rerolling from 0.
Bug 7074

Bug 7074

Bug 7251.

If bf_min_prio_reserve was not the last option listed, the setting would
not have been applied correctly, and an error message printed instead.

Bug 7242.

Bug 6221

Bug 6221

Put comments in the code to help figure out what is being packed.
No real code change.

Bug 7204

Bug 7204

Bug 6221

Return ESLURM_INVALID_TIME_MIN_LIMIT instead of ESLURM_INVALID_TIME_LIMIT
to provide a better error message back to the user.

Bug 6221

Bug 6221

Bug 7179.

bug 7203

Bug 6847
Bug 7190

For heterogeneous jobs, do not count the each component against the QOS or
association job limit multiple times.

bug 7190

bug 7197

bug 7197

bug 7197