Configuring slurm with munge manually installed in /usr/local, with the
library in /usr/local/lib but an empty /usr/local/lib64 directory
will cause the munge plugins to look for libmunge.so in the wrong place.

The munge.spec file has historically provided libmunge.so as part of
munge-devel, which Slurm depends on already.

Add SchedulerParameters configuration parameter of "default_gbytes", which
    treats numeric only (no suffix) value for memory and tmp disk space as being
    in units of Gigabytes. Mostly for compatability with LSF.

Move BatchScript to end of each job's information when using
    "scontrol -dd show job" to make it more readable.

Can cause random assertion failures and core dumps due to differences
between definition in slurm.h and bitstring.h.

Inadverently introduced in 8967a4e7.

17.02 API has been changed so that node Port parameter is now packed and
unpacked on REQUEST_NODE_INFO RPC. Some client requests such as 'scontrol
write config', 'scontrol show node' will display the port if different
to SlurmdPort. Port parameter is also available now to 'sinfo' if
explicitly requested through '-O port' and to the 'sview' full node info.

Always send SlurmdPort in RPC even when in multiple-slurmd mode.

Bug 3240.

Fix security issue caused by insecure file path handling triggered by
the failure of a Prolog script. To exploit this a user needs to
anticipate or cause the Prolog to fail for their job.

(This commit is slightly different from the fix to the 15.08 branch.)

CVE-2016-10030.

Fix security issue caused by insecure file path handling triggered by
the failure of a Prolog script. To exploit this a user needs to
anticipate or cause the Prolog to fail for their job.

CVE-2016-10030.

Prevent "stray" jobs from using resources when the srun/salloc
will never launch the actual compute tasks.

Bug 3344.

PluginDir is allowed to be a PATH-style list of directories; remove
incorrect test of the variable as if it were a single directory and
comment that the check for that is elsewhere.

Bug 3361.

Add SchedulerParameters option of spec_cores_first to select specialized
    cores from the lowest rather than highest number cores and sockets.
bug 3349

Null terminate before strchr().

This is a new message when "PrologFlags=contain" or
  "PrologFlags=alloc" is configured.
bug 3351

Cancel interactive job if Prolog failure with "PrologFlags=contain"
    configured.
bug 3351

Do not allocate specialized cores to jobs using the --exclusive option.
bug 3349

Fix memory and file descriptor leaks in slurmd daemon's sbcast logic.

The part_ptr is sent into the function, there is no reason to look it up
again.  Coverity reported this.

Remove req_node_layout, which was only used with wiki/wiki2.
This leads to removal of _get_cpu_cnt() as well.

Remove SchedulerPort, only used for communication to Moab/Maui.
Remove slurm_get_sched_port() from API.
Remove schedport from slurm_ctl_conf struct.

bug 2161

burst_buffer/cray - Remove leading zeros from node ID lists passed to
    dw_wlm_cli program.
bug 3008

version is lower than the min version, set it to the min.

Bug 3050

bug 1752

sched/backfill - Fix logic to reserve resources for jobs that require a
    node reboot (i.e. to change KNL mode) in order to start.
bug 3346

go into JobAdminHeld.

Bug 3201

Fix for possible infinite loop in select/cons_res plugin when trying to
    satisfy a job's ntasks_per_core or socket specification.
bug 3329

Bug 2992.

Modify regression test1.89 to avoid leaving vestigial job. Also reduce
    logging to reduce likelyhood of Expect buffer overflow.
bug 3273

node_features/knl_generic - Add capability to detected Uncorrectable Memory
    Errors (UME) and if detected then log the event in all job and step stderr
    with a message of the form:
    error: *** STEP 1.2 ON tux1 UNCORRECTABLE MEMORY ERROR AT 2016-12-14T09:09:37 ***
bug 3341

Reverts most of commit 84023f27.

Searching the PATH in slurmd can fail due to root_squash'd NFS
filesystems, leading to the "wrong" program being launched.

If you'd like the performance benefit from avoiding this lookup
during each separate task launch, set SLURM_TEST_EXEC=1 instead
which will perform the lookup once within srun, which then ensures
the lookup happens under the users own environment and not that
of the slurmd.

Bug 2992.

level.

Provide limited support for the MemSpecLimit configuration parameter without
    the task/cgroup plugin.

If the second call to getgrouplist() found additional groups,
ngroups will be overwritten with this new larger value, while
the gids list would be truncated. (ngroups is a value-result arg.)
This will then lead to _gids_cache_lookup() returning the wrong
number of groups including invalid parts of memory, which are likely
to include some zeros.

Those zeros could then make it to the setgroups() call and thus
give the user access to the root group. Especially as setgroups
will succeed as long as the array does not contain -1 as a gid.

Bug 3320.

task/cgroup - Change error message if CPU binding can not take place to
    better identify the root cause of the problem. Specifically, if
    the hwloc_get_obj_below_by_type() function call completely fails
    that is likely due to task/affinity not being configured, so
    cpusets are not configured. Previous message was
    "task/cgroup: task[%u] infinite loop broken while trying to provision compute elements using %s (bitmap:%s)"
    The new message is
    "task/cgroup: hwloc_get_obj_below_by_type() failing, task/affinity plugin also required"