Add new SchedulerParameter (max_array_tasks) to limit the maximum number of
    tasks in a job array independently from the maximum task ID (MaxArraySize).
bug 2676

Add SbcastParameters configuration option to control default file
    destination directory and compression algorithm.
bug 2977

Replace sjstat, seff and sjobexit RPM packages with a single "contribs"
    package.

Remove separate slurm_blcr package. If Slurm is build with BLCR support,
    the files will now be part of the main Slurm packages.
bug 2061

Document that node Weight takes precedence over load with LLN scheduling.
bug 3204

The test cluster had jobs in the job table which prevented the cluster
from being deleted. This also caused problems for other tests because
accounts would be added to the test cluster and the accounts couldn't be
deleted because the cluster still had jobs.

This prevents it from being added to any stray clusters and allows it to
be cleaned up easier. test21.36 wasn't destroying the created test
cluster and this account was being added to the cluster and it couldn't
delete the account on subsequent runs. test21.36 is solved now too.

Follow on to commit c3266fca for 17.02+.

task/cray's _get_numa_nodes() function needs to run before task/cgroup
cleans up the cgroup hierarchies, otherwise ALPS memory compaction will
never run.

Also move task_p_add_pid() outside the #ifdef HAVE_NATIVE_CRAY
block so that the plugin will load (albeit without any functionality)
on non-Cray systems for testing purposes.

Revise documentation and provided slurm.conf templates as well.

Bug 3154.

Do not include SLURM_JOB_DERIVED_EC, SLURM_JOB_EXIT_CODE, or
    SLURM_JOB_EXIT_CODE in PrologSlurmctld environment (not available yet).
bug 1431

bug 2390

Many open Coverity issues point back to this. The results
of passing a negative value to strerror() are undefined,
return a static string instead.

the pro/epilog of jobs.

Even if several commands, e.g. sview, are not built and installed, the corresponding man pages are still installed.

The attached patch stop to install man pages when the corresponding commands are not built installed.
bug 2393

This reverts commit 428347cf.

Decided we didn't want a core dump on ever fatal, as fatal is used
in other programs instead of just the daemons.

There is a problem with gathering batch step statistics for jobs which are allocated on more than one node.

Sstat asks wrong node for batch step stats. It requests info from last node from hostlist while it should ask first host from hostlist (i.e. BatchHost), because only on the first node the batch step actually executes.

For example, when you have a job allocated on nodes n000[1-2] with BatchHost=p0001. You should be able to check its statistics by running sstat [ with -vv switch for more verbose output] (e.g. sstat -j 1234.batch -vv). Then you can see lines:

sstat: debug:  slurm_job_step_stat: getting pid information of job 1234.4294967294 on nodes n0002
sstat: debug:  job step 1234.4294967294 has already completed

The problem lays in sstat source code. For batch step a hostlist variable is taken from the hostlist_pop function, which returns last host from given hostlist. This should be replaced with the hostlist_shift function, which returns first host from the given hostlist. Patch attached.

bug 2975

burst_buffer/cray: Accept new jobs on backup slurmctld daemon without access
    to dw_wlm_cli command. No burst buffer actions will take place. Newly
    submitted jobs will be accepted and stay in pending state. Jobs depedent
    upon stage-in or stage-out will remain in their current state until the
    action can take place.

This commit fixes a bug in the multi-prog handling. When running
salloc -N 2 srun -O --multi-prog mp.conf where mp.conf reads

0-192 true

srun crashes can be observed. valgrind reports:

==6857== Invalid read of size 4
==6857==    at 0x45938D: bit_realloc (bitstring.c:189)
==6857==    by 0x5977A9: _update_task_mask (multi_prog.c:335)
==6857==    by 0x597A5E: _validate_ranks (multi_prog.c:403)
==6857==    by 0x597D1E: verify_multi_name (multi_prog.c:469)
==6857==    by 0x6E7B4BE: launch_p_handle_multi_prog_verify (launch_slurm.c:453)
==6857==    by 0x58A25D: launch_g_handle_multi_prog_verify (launch.c:493)
==6857==    by 0x58E556: _opt_args (opt.c:1927)
==6857==    by 0x58A3B9: initialize_and_process_args (opt.c:270)
==6857==    by 0x591F82: init_srun (srun_job.c:459)
==6857==    by 0x427E70: srun (srun.c:193)
==6857==    by 0x428E23: main (srun.wrapper.c:17)
==6857==  Address 0x5ace440 is 16 bytes inside a block of size 28 free'd
==6857==    at 0x4C2BB4A: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==6857==    by 0x446886: slurm_xrealloc (xmalloc.c:139)
==6857==    by 0x45944C: bit_realloc (bitstring.c:191)
==6857==    by 0x5977A9: _update_task_mask (multi_prog.c:335)
==6857==    by 0x597A5E: _validate_ranks (multi_prog.c:403)
==6857==    by 0x597D1E: verify_multi_name (multi_prog.c:469)
==6857==    by 0x6E7B4BE: launch_p_handle_multi_prog_verify (launch_slurm.c:453)
==6857==    by 0x58A25D: launch_g_handle_multi_prog_verify (launch.c:493)
==6857==    by 0x58E556: _opt_args (opt.c:1927)
==6857==    by 0x58A3B9: initialize_and_process_args (opt.c:270)
==6857==    by 0x591F82: init_srun (srun_job.c:459)
==6857==    by 0x427E70: srun (srun.c:193)

Do not process SALLOC_HINT, SBATCH_HINT or SLURM_HINT environment variables
    if any of the following salloc, sbatch or srun command line options are
    specified: -B, --cpu_bind, --hint, --ntasks-per-core, or --threads-per-core.bug 3118

Without this change, only the error was available, but no identification
  of the specific plugin that failed.

Coverity was complaining that the return value of s_p_get_* was
ignored. I added typecasting of the return value to (void) where
needed. No change in logic, just making Coverity happy ;)