Bug 7680.

pthread_create is not a async-signal-safe function, since it can't be
used in a fork (according to POSIX). Calling it may end-up with deadlock
in glibc.

Bug 7666.

Bug 7779

Bug 7779

Add some slurm.conf validation to make sure that typed GRES and
untyped GRES don't mix (e.g. do not allow gres=gpu:1,gpu:tesla:1).

Bug 7517

If a configured GPU matches a GPU on the system, match them together.
If a configured GPU has mismatched Cores or Links with the system GPU,
then omit that configured GPU from the final list.

Bug 7517

Before, data from slurm.conf was only partially merged if there was
no corresponding entry in gres.conf.
Now, Slurm tries to match up gres.conf records to what is defined in
slurm.conf.

Bug 7517

Bug 7169

Increase the maximum array len large to be packed/unpacked with one order of
magnitude, since the current value demonstrated it is not enough when an MPI
program spawns a considerable amount of tasks over a big set of nodes.

This limit was introduced in 627928f4.

Bug 7495

There was never any security to allow for this,
so we are just removing it.

Bug 7765

Admin/Operator users were not able to skip MaxQueryTimeRange when trying
to show/fix runaway jobs.
This commit uses _validate_operator() instead of
_validate_slurm_user() in _get_jobs_cond() as well as check for operators
in _fix_runaway_jobs().

Bug 7765

Bug 7708

Don't remove jobs from preemptee_candidates List.

Bug 7708

Since this happens inside a the user process, this can inadvertently
cause the user's job to die by running out of file descriptors.

Bug 7814.

Co-authored-by: William Arndt <warndt@lbl.gov>

Bug 7499

Regression introduced in fb26b706.

Bug 7675

Now the signaling of the batch step and the handeling of the flags is totally
handled in _kill_all_active_steps() in slurmd, and _handle_signal_container()
in stepd to ensure that:
- if KILL_JOB_BATCH then only batch container is signaled
- if KILL_FULL_JOB then batch script and its children are also signaled
- if both of the above then only the batch script and its children are signaled

We do not relay anymore on proctrack_g_signal() to handle the batch step
signaling anymore, therefore it works the same for all proctrack plugins.

This commit also includes minor related fixes in other code handling such
signaling flags, and documentation improvement.

Bug 7282

Signed-off-by: Tim Wickberg <tim@schedmd.com>

Bug 7697

1cd43fce

Bug 7630

Bug 7727

This was missed in commit 6ac4ce84.

An incorrect logic with the variables holding available cores in the
gres_plugin_job_core_filter3() function lead to a potential infinite
"while (avail_cores_tot > req_cores)" loop, leaving slurmctld unresponsive.

Bug 7685.

Bug 7719

Signed-off-by: Danny Auble <da@schedmd.com>

Regression caused by 72736af2.

Bug 7708.

FastSchedule will be removed in 20.02.
FastSchedule=2 functionality has been moved to
SlurmdParameters=config_overrides.

Bug 7496.


Signed-off-by: Tim Wickberg <tim@schedmd.com>

Bug 7699

Continuation of 64876087

Bug 7698

Bug 4781

Otherwise, there could be time frames where printed schednodes
information could be obsolete.

Bug 7676.

exclusively to that job.

Bug 7510

use correct start_time for TIME_FLOAT resevation
in _job_overlap()

Bug 7458

Bug 7458

Bug 7458

Move _validate_node_choice() before prolog/epilog check

Bug 7458

Free the gres_devices list to avoid a valgrind warning on exit.

Bug 7644.

Continuation of 30bbc11d



Bug 7445

Signed-off-by: Dominik Bartkiewicz <bart@schedmd.com>

When --batch=<feature> is used, the batch_host isn't chosen until the
job is being launched -- because the features could be different on boot
(e.g. KNL nodes). Thus if the job is allocated nodes that need to be
booted, it needs to wait till they are all booted so it can make a
decision at launch time.

Bug 7445

Signed-off-by: Dominik Bartkiewicz <bart@schedmd.com>

This is a continuation to 7da439b4

Bug 7445

Only do so once the task actually finishes. Otherwise, a requeued task
could set an incorrect max_exit_code even if completed with exit code 0
after re-running again, leading to problems with i.e. other jobs with an
afterok type of dependency on such array relying on the incorrectly set
max_exit_code.

Bug 7552.

In case of features like cpu&fastio&[knl|westmere] additional bit_or
resulted in returning something like (cpu&fastio)|knl|westmere, which
is obviously wrong. XOR/XAND features are handled properly in
_get_req_features.

Bug 7378