license of a certain type.

node_features/knl_cray plugin: Increase default CapmcTimeout parameter from
    10 to 60 seconds.
bug 3100

capmc_suspend/resume - If a request modify NUMA or MCDRAM state on a set of
    nodes or reboot a set of nodes fails then just requeue the job and abort the
    entire operation rather than trying to operate on individual nodes.
bug 3100

Allow a node's PowerUp state flag to be cleared using update_node RPC.
bug 3100

When powering up a node to change it's state (e.g. KNL NUMA or MCDRAM mode)
    then pass to the ResumeProgram the job ID assigned to the nodes in the
    SLURM_JOB_ID environment variable.
bug 3100

Don't log error for job end_time being zero if node health check is still
    running.
bug 3053

Restore ability to manually power down nodes, broken in 15.08.12
in commit b4904661

The patch introduced in commit b4904661 (not powering down dead node) has a bad side effect.  Adding the "(node_ptr->last_idle != 0)" condition prevents from powering down nodes with the following command:

scontrol update nodename=nX state=power_down

because the state update function relies on zeroing the "last_idle" variable when a power_down is requested (see src/slurmctld/node_mgr.c, line 1589).

Reverting this commit should solve the problem...but I let you decide...

Didier GAZEN

bug 3093

node_features/knl_cray: If a node is rebooted outside of Slurm's direction,
    update it's active features with current MCDRAM and NUMA mode information.
bug 3071

Fix race condition that could result in MCDRAM state information coming
  from capmc rather than cnselect (used state for next boot rather than
  latest boot).
bug 3080

No functional change, just silencing the warning message in this instance.

Bug 3079.

Fix some whitespace on line endings while here.

Bug 3073.

bug 3065

Add Slurm overview by Alex on first day, move KNL to second morning,
shift roadmap to right after lunch.

bug 3065

Previous cap was 2 sec (default TCP timeout) times the node count
  and divided by 1000. A 9000 node job would have the messages
  spread out over 18 seconds. This change caps the spread at
  5 seconds and assumes the normal TCP logic can handle the rest
bug 3044

If the overhead of determining the hostlist for a given task list
is too high, then report a hostlist of "Unknown" instead. If the
overhead is too high, then srun will become unresponsive and communications
will timeout/fail.
bug 3044

Modify srun task completion handling to only build the task/node string for
    logging purposes if it is needed. Modified for performance purposes.
bug 3044

Add get_log_level() function to return the highest LOG_LEVEL_* used
for any logging mechanism.

This reverts commit 1ec2a4ae.

Bug 3063.

Restructure srun command locking for task_exit processing logic for improved
  parallelism. This change decreases the amount of time consumed by serial
  logic by 2 orders of magnitude.
bug 3044

Preserve node "RESERVATION" state when one of multiple overlapping
    reservations ends. Previous logic would clear the node's
    RESERVATION state flag when any one of the reservations on the
    node ended rather than keeping the node in RESERVATION state
    until the last reservation ended.
bug 3057

The logic is now heavier weight, so increase interval between tests
  from 2 to 5 seconds

Handle case when slurmctld daemon restart while compute node reboot in
    progress. Return node to service rather than setting DOWN.
bug 3042