This also adds the ability to display allocated memory on a node
bug 1804

Conflicts:
	slurm/slurm.h.in
	src/slurmctld/job_scheduler.c
	src/slurmd/slurmstepd/task.c

Rather than generating loads of log messages about too much time
being used to build the job queue every few seconds, log it only
every 10 minutes.
bug 1827

Log the number of job-partition pairs added to the queue for job
scheduling.
bug 1827

Minor updates to web page document
Minor improvements to error message logic

Improve logging
Disable stage-in/out if only persistent buffers used
End-to-end with persistent buffer works
Improve user error information

Also, use first pool as default if none configured
Several changes for updated API calls

On Cray we were seeing an srun error reading slurmstepd message header.
  This was due to a shutdown race condition and the error message has
  been removed.
Cray: Disable LDAP references from slurmstepd on job launch due for
  improved scalability.
Document EioTimeout configuration parameter for large system
bug 1786

If an RPC is being sent to a node and the destination job no longer
exists, log the message "job not running" rather than looking up
the error and reporting "invalid job id".

No change in logic, just renamed some variable and function names
  for greater clarity

Previously only batch job completions were being captured.
bug 1820

If a job was running on the node when slurmctld restarted. The slurmd
would notify slurmctld when the job ended and slurmctld would change
its state from UNKNOWN to IDLE, at least if the job termination happened
prior to the slurmd being asked for configuration information. The
configuration information might then not be collected for some time.
I've modified the code to address this problem and try to collect
configuration information from every node after slurmctld startup,
eliminating this race condition.
bug 1805

Bug 1208

This patch provides a rewrite of how /proc/cpuinfo is parsed in common_jag.c, as the original code made the incorrect assumption that cpuinfo follows a sane format across architectures ;-)

The motivation for this patch is that the original code was producing stack smashing on a POWER7 running RHEL6.4 Red Hat adds -fstack-protector along with a lot of other protective CFLAGS when building RPMs. The code ran okay with -fno-stack-protector, but that is not the best work-around.

So, the relevant /proc/cpuinfo line on an Intel (Xeon X5675) system looks like:

cpu MHz                : 3066.915

Whereas the relevant line in a POWER7 system is

clock                : 3550.000000MHz

My patch replaces the assumption that the relevant number starts 11 characters into the string with another assumption: That the relevant number starts two characters after a colon in a string that matches (M|G)Hz.

All in all, the function has a few more calls, which may be a performance issue if it has to be called multiple times, but since the section I edited only gets evaluated if we don't know the cpu frequency, getting it right will actually result in fewer string operations and unnecessary opens of /proc/cpuinfo for systems likewise affected.

Finally, I also read the actual value into a double and multiply it up to the size indicated by the suffix, so we end up with KHz? It was unclear what the original code intended, since it matched both MHz and GHz, replaced the decimal point with a zero, and read the result as an int.

--
Chandler Wilkerson
Center for Research Computing
Rice University

This reverts commit 2c95e2d2.

Conflicts:
	src/plugins/select/alps/basil_interface.c

This is related to bug 1822.  It isn't clear why the code was taken out in
this commit in the first place and based off of commit 2e2de6a4 (which is
the reason for the conflict) we tried unsuccessfully to put it back.

It appears the only difference here is the addition of
always setting mppnppn = 1 instead of always to
job_ptr->details->ntasks_per_node when no ntasks is set.

This appears to only be an issue with salloc or sbatch as ntasks
is always set for srun.

I've attached a patch to add a new script and templates that create slurm.conf and gres.conf on a Cray SMW.

The script parses the output of xthwinv, combines nodes with the same hardware, and writes the output to slurm.conf. It also uses NodeName in gres.conf to allow having the same gres.conf file on all nodes.

I removed the settings in slurm.conf that we had commented out or set to the default.

Also change units to bytes everywhere

Bug 1783