Bug 4127

When --mail-type option isn't requested with ARRAY_TASKS, we need somehow
to summarize the different states each task finished in the array. We've
added a new ARRAY_TASK_REQUEUED flag to the array_flags to indicate that
at least one task was requeued. Also the logic now detects if at least
one task failed and/or if otherwise all finished successfully.

The patch also removes the RunTime from the the e-mail when summarizing
whole array, since it doesn't make sense to specify just the RunTime
of one of the tasks for this case.

Patch also fixes when ARRAY_TASKS is specified, previously the mail
notification for the master job task record included a range of
ExitCodes for all the tasks. Since this option is not for summarizing,
the patch makes it so only the range is shown when the option isn't
specified.

Bug 4539.

Scheduling fix for changing node features without any NodeFeatures plugins.
Bug 4577

Avoid setting node in COMPLETING state indefinitely if the job initiating
the node reboot is cancelled while the reboot in in progress. Bug
introduced in commit 7d246784

Bug 4536

Bug 4448

Logic was backwards before.

Bug 4459.

Refactor logging logic to avoid possible memory corruption on non-x86
    architectures.
bug 4469

only affects current 17.11 users tracking licenses or GRES in the database.

Bug 4553

increment ids for the tres_table.

Bug 4553

Turns out auto_increment=# doesn't work in MySQL at the time of writing
but does work correctly in MariaDB >=10.2.4+

See bug for more information.

This reverts commit 12794fb0.

increment ids for the tres_table.

Bug 4553

Turns out auto_increment=# doesn't work in MySQL at the time of writing
but does work correctly in MariaDB >=10.2.4+

See bug for more information.

Bug 4555

Bug 4558.

to replace DOWN or DRAINED nodes.

Bug 4537

Bug 4564

The ability to enforce strict content-type checking has existed since
Elasticsearch 5.3 via the http.content_type.required configuration setting.
In 5.x it is optional, and defaults to false, in Elasticsearch 6.0, that
setting defaults to true, and there is no way to disable it.

If required and not appended, Elasticsearch would response with status 406,
so "Content-Type: application/json" is appended from now on. Tested against
Elasticsearch 5.5.1 and 6.1.1.

Bug reported to slurm-users list from E.S. Rosenberg. Patch suggested didn't
check the return code of curl_slist_append() and didn't free the slist.

Bug 4551,4553

This mostly reverts commit 8325e9f8.

Bug 4478

We decided to not do this by default for the slurmd.  See bug 4478 for
more info/reasoning.

We also didn't want to rely on #include <linux/oom.h> since that might
not exist everywhere.

Format for inserted X11 cookies is now hostname/unix:display, rather
than localhost:display. This seems to resolve some issues for different
setups while not causing any issues in my test environments. This also
helps avoid a conflict between multiple nodes in a larger job if the
same display number is allocated on multiple nodes simultaneously.

Bug 4420.

Bug 4483

Bug 4541

The slurm and pmix RPMs both attempt to install a libpmi. While Slurm's
version should be preferred, the best way to handle this is to have both
pmix and slurm packages split this off into a separate RPM that can
conflict with each other, giving the admin a way to resolve the conflict
manually.

Bug 4511.

The cluster name could have been free'd before it would be strdup'ed.
Happens when a federated persisent connection is established and the
cluster needs to sync up with the remote cluster.

Bug 4503

Remove the origin job when a remote sibling job was cancelled while the
origin was down.

e.g.
job originated on cluster1
job runs on cluster2
cluster1 goes down
job is cancelled on cluster2
cluster1 comes backup. The origin job should be removed if the sibling
job was cancelled.

The user_name is only populated if LaunchParameters=send_gids is set.
Without this field set, the pam_setup call will get NULL instead of the
user_name, leading to batch job launch failures. (The step launch code
already handles this same issue separately.)

Bug 4412.

On a job [pack]allocation RPC request, if the allocation succeed but the
send response message back to the client failed (i.e. srun was killed before
it could receive the response), then modify the job_record pointer so that
the job_state is set to FAILED, the exit_code as if the job got a SIGTERM
signal and the state_reason to FAIL_LAUNCH. Then users when querying
the job with sacct can discern that something bad happend for this scenario,
instead of STATE being showed as COMPLETED and the ExitCode as 0:0.

Bug 4513.

Use FREE_NULL_BUFFER instead, otherwise we could attempt to
free_buffer this a second time if we jump to the rwfail label.

Bug 4484.

When printing fields in sacct with user specified units (--units), the
nnodes field showed an incorrect string. This commit reverts a65fa572
and avoids the unit conversion, which does not make sense outside the
context of a Blue Gene systems (deprecated) anyways.

Bug 4490.

Slurm may generate empty manifest files depending on configuration
and library availability. Disable the new empty manifest check to
allow builds to proceed with rpm 4.13+ / Fedora 25+.

Bug 4453.

before printing anything for a connection.