On second thought, since we enable debug mode automatically this
won't be useful for what I plan to use it for, and since nothing
uses it let's just remove it entirely.

Doesn't really need to be directly in configure.ac, so move it.

Note that this has been subtly broken for quite a while:
- The correct variable to test is x_ac_debug, not ac_debug.
- The correct value when debugging is enabled is "yes", not "true".

Update slurm.spec and slurm.spec-legacy as well.

Update slurm.spec and slurm.spec-legacy as well.

Caused by b1ff4342 when moving the slurmdbd agent internals.

Bug 5882.

Regression caused in 18.08.1; can happen under any circumstance that
leads to the job array elements being split before the backfill/sched
has had a chance to populate these.

Bug 5876.

# Conflicts:
#	src/common/slurmdbd_defs.c

up when the dbd was down.

This reverts commit c07784b4.

Bug 5875 and 5613

Change to true/false while here.

CID 188810.

The industry's settling on a term different than what we'd originally
gone with, so make a concession to the newer term and make it easier
to search for and discover.

Just return an xmalloc()'d string rather than handle all this
manual buffer length tracking. The only calling path into here
is trivially updated to work with this instead.

Note that gres_plugin_init() always returns 0 at present, so
take the opportunity to simplify that handling as well.

The safe_read() and safe_write() macros are the better option, and
these were only used in one location each. (Converted in previous
patch.)

xmalloc cannot return NULL here - on error it would fatal() already.

Fix function declaration and code to match style guide.

Change the debug3 message to print raw timestamps, rather than
convert, same as the other debug3 messages. This also then removes
the last calls to timestr(), so remove that static function as well.

strcpy would not have worked for the described purpose anyways.

I was going to rework these to remove the strcpy() calls, but
found that each of these have equivalent debug3() statements
just without the timestamps rendered as time strings.

Given that these are debug3(), and that you'd need to manually
inject a DEBUG_TIME definition somewhere to expose them, I don't
see the value of keeping them around. Especially as if you're digging
this far in to the source, it seems easier to troubleshoot off the raw
timestamp anyways IMNSHO.

Replace strcpy() with strlcpy().

Replace strcpy() with strlcpy(), and refactor the code. Fix the
function prototype to match the style guide while here.

Replace strlen(), xmalloc(), strcpy() sequence with xstrdup().
Add const annotation and fix function declaration style while here.