steps.

bug 3610

Bug 3097

Fix correct macro this time, Tim Wickberg's fault for prior failed patch.

Bug 3605.

This reverts commit 53523c54.

Bug 3605.

This reverts commit 979db782.

Partial backport from 17.11 commit 4874f988.

Bug 2484

Association.

This reverts commits 92d2c645 and 37be42ec.  This caused incorrect
behavior, original code was correct.  This also corrects documentation additions
in commit 4cfe6bde.

This code caused the first clause never to be correct and if you were
over the limit you would get the third clause reporting a huge number available
where it should be a negative number.  The reality is the first clause should
had been triggered and handled correctly.

This reverts commit af52111c.

Association.

This reverts commits 92d2c645 and 37be42ec.  This caused incorrect
behavior, original code was correct.  This also corrects documentation additions
in commit 4cfe6bde.

This code caused the first clause never to be correct and if you were
over the limit you would get the third clause reporting a huge number available
where it should be a negative number.  The reality is the first clause should
had been triggered and handled correctly.

system.

What this patch does is looks at the partition the job is running in to get
the cpus per node off the first node in the partition.  Before it would
just look at the first node on the system which isn't that great on a
heterogeneous system.

Bug 3486

Bug 3577.

other half go to libslurmdb.so.

Turns out in 17.11 if you are using --with-shared-libslurm sacct will link to
both libslurmfull.so and libslurmdb.so.  In the case of linking to the
accounting_storage/slurmdbd plugin it will callback to slurmdbd_defs_init
which would be in libslurmdb.so but the call to acct_storage_p_get_connection
would call slurm_open_slurmdbd_conn which is in libslurmfull.so and it would
xassert on slurmdbd_defs_inited which was set in libslurmdb.so but not in
libslurmfull.so.

So the moral of the story is don't export half a file in one lib and the other
half in a different lib.

Perhaps we should only have 1 lib all together, but that isn't the way it is
done today.  This fixes the issue and has the entire file exported to
libslurmfull.so so we should be good.  But just a note for the future.

This was an unexpected regression caused by commit 5a5347c7.

Bug 3530

NOTE: there does appear to be other issues here, but we didn't fell comfortable
changing this many things in 17.02 for fear of breaking something.

This only fixes a bit of the issue as it appears node_scheduler.c has a fuller
test for these.  In 17.11 I plan to make this a function that will fill in
the min, req, max nodes and use it in both places in the code to prevent this
from happening again.

Code calls list_find_first to search in resv_list whether the requested
name for the new reservation already exists. If it exists, resv_ptr is
set with the pointer to the existing reservation. Then the code goto
bad_parse label and xfreed that resv_ptr, thus corrupting the list data
by freeing the existing reservation. This is fixed by only freeing memory on the
new local resv_ptr instead of always freeing memory.  xfree is also not
sufficient for freeing the memory, we needed to call _del_resv_rec() or we would
leak the memory we had transferred from the resv_desc_ptr.  This also involved
NULLing out the other variables freed after bad_parse, or you would get
double frees.

Bug 3558.

assoc_mgr_make_tres_str_from_array.

array task id.

This is combination with the previous commit.

Bug 3563

I also made seff work on a non-standard perl install.

I would had loved to do this in multiple commits, but it wasn't that possible.

moment the dbd is assigning a db_index.

Bug 3512

This is a definite culprit of creating runaway jobs.

because the dbd is setting a db_index.

in the middle of setting db_indexes.

multifactor.  Almost impossible to make happen, but still a bug.

The stepd->protocol_version field is not set at this point,
if not handled here then the following call to stepd_get_uid
will fail as a protocol_version of zero is rejected.

This appeared to work in 16.05 and older as the stepd_get_uid()
API call used to ignore the invalid protocol_version and
return the correct result anyways, but that was fixed in c177ff95.

Bug .

correctly with Slurm tools.

What this does is replace the incoming string with the gres actually chosen.

Continuation of last commit.

Bug 3521

the database.

Symbol required for read_slurm_cgroup_conf() call to work.

Bug 3550.

burst_buffer/cray - Fix parsing for discontinuous allocated nodes. A job
    allocation of "20,22" must be expressed as "20\n22" when passed to
    the dw_wlm_cli program.
bugs 3540 and 3544

capmc_resume (Cray resume node script) - Do not disable changing a node's
    active features if SyscfgPath is configured in the knl.conf file.
bug 3533

If a job is cancelled by the user while it's allocated nodes are being
    reconfigured (i.e. the capmc_resume program is rebooting nodes for the job)
    and the node reconfiguration fails (i.e. the reboot fails), then don't
    requeue the job but leave it in a cancelled state. Note the JOB_RECONFIG_FAIL
    state flag is currently only used by capmc_resume, but could be used for
    other programs responsible for node reboots.
bug 3392