NEWS

This file describes changes in recent versions of Slurm. It primarily
documents those changes that are of interest to users and administrators.

* Changes in Slurm 18.08.0pre1
==============================
 -- Add new burst buffer state of "teardown-fail" to indicate the burst buffer
    teardown operation is failing on specific buffers. This changes the numeric
    value of the BB_STATE_COMPLETE type. Any Slurm version 17.02 or 17.11 tool
    used to report burst buffer state information will report a state of "66"
    rather than "complete" for burst buffers which have been deleted, but still
    exist in the slurmctld daemon's tables (a very short-lived situation).
 -- Multiple backup slurmctld daemons can be configured:
    * Specify "BackupController#=<hostname> and "BackupAddr#=<address>" to
      identify up to 9 backup servers.
    * Output format of "scontrol ping" and the daemon status at the end of
      "scontrol status" is modified to report up status of the primary and all
      backup servers.
    * "scontrol takeover [#]" command can now identify the BackupController
      index number. Default value is "1" (the configured "BackupController" or
      "BackupController1" node).
 -- Enable jobs with zero node count for creation and/or deletion of persistent
    burst buffers.
    * The partition default MinNodes configuration parameter is now 0
      (previously 1 node).
    * Zero size jobs disabled for job arrays and heterogeneous jobs, but
      supported for salloc, sbatch and srun commands.
 -- Add "scontrol show dwstat" command to display Cray burst buffer status.
 -- Add "GetSysStatus" option to burst_buffer.conf file. For burst_buffer/cray
    this would indicate the location of the "dwstat" command.
 -- Add node and partition configuration options of "CpuBind" to control default
    task binding. Modify the scontrol to report and modify these parameters.
 -- Add "NumaCpuBind" option to knl.conf file to automatically change a node's
    CpuBind parameter based upon changes to a node's NUMA mode.
 -- Add sbatch "--batch" option to identify features required on batch node.
    For example "sbatch --batch=haswell ...".
 -- Add "BatchFeatures" field to output of "scontrol show job".
 -- Add support for "--bb" option to sbatch command.
 -- Add new SystemComment field to job data structure and database. Currently
    used for Burst Buffer error logs.
 -- Expand reservation "flags" field from 32 to 64 bits.
 -- Add job state flag of "SIGNALLING" to avoid race condition with multiple
    SIGSTOP/SIGCONT signals for the same job being active at the same time.
 -- Properly handle srun --will-run option when there are jobs in COMPLETING
    state.
 -- Properly report who is signaling a step.
 -- Don't combine updated reservation records in sreport's reservation report.
 -- node_features plugin - Add suport for XOR & XAND of job constraints (node
    feature specifications).
 -- Add support for parenthesis in a job's constraint specification to group
    like options together. For example
    --constraint="[(knl&snc4&flat)*4&haswell*1]" might be used to specify that
    four nodes with the features "knl", "snc4" and "flat" plus one node with
    the feature "haswell" are required.
 -- Improvements to how srun searches for the executible when using cwd.
 -- Now programs can be checked before execution if test_exec is set when using
    multi-prog option.
 -- Report NodeFeatures plugin configuration with scontrol and sview commands.
 -- Add acct_gather_profile/influxdb plugin.
 -- Add new job state of SO/STAGE_OUT indicating that burst buffer stage-out
    operation is in progress.

* Changes in Slurm 17.11.3
==========================
 -- Send SIG_UME correctly to a step.
 -- Sort sreport's reservation report by cluster, time_start, resv_name instead
    of cluster, resv_name, time_start.
 -- Avoid setting node in COMPLETING state indefinitely if the job initiating
    the node reboot is cancelled while the reboot in in progress.
 -- Scheduling fix for changing node features without any NodeFeatures plugins.
 -- Improve logic when summarizing job arrays mail notifications.
 -- Add scontrol -F/--future option to display nodes in FUTURE state.
 -- Fix REASONABLE_BUF_SIZE to actually be 3/4 of MAX_BUF_SIZE.
 -- When a job array is preempting make it so tasks in the array don't wait
    to preempt other possible jobs.
 -- Change free_buffer to FREE_NULL_BUFFER to prevent possible double free
    in slurmstepd.
 -- node_feature/knl_cray - Fix memory leaks that occur when slurmctld
    reconfigured.
 -- node_feature/knl_cray - Fix memory leak that can occur during normal
    operation.
 -- Fix srun environment variables for --prolog script.
 -- Fix job array dependency with "aftercorr" option and some task arrays in
    the first job fail. This fix lets all task array elements that can run
    proceed rather than stopping all subsequent task array elements.
 -- Fix potential deadlock in the slurmctld when using list_for_each.
 -- Fix for possible memory corruption in srun when running heterogeneous job
    steps.
 -- Fix job array dependency with "aftercorr" option and some task arrays in
    the first job fail. This fix lets all task array elements that can run
    proceed rather than stopping all subsequent task array elements.
 -- Fix output file containing "%t" (task ID) for heterogeneous job step to
    be based upon global task ID rather than task ID for that component of the
    heterogeneous job step.

* Changes in Slurm 17.11.2
==========================
 -- jobcomp/elasticsearch - append Content-Type to the HTTP header.
 -- MYSQL - Fix potential abort of slurmdbd when job has no TRES.
 -- Add advanced reservation flag of "REPLACE_DOWN" to replace DOWN or DRAINED
    nodes.
 -- slurm.spec-legacy - add missing libslurmfull.so to slurm.files.
 -- Fix squeue job ID filtering for pending job array records.
 -- Fix potential deadlock in _run_prog() in power save code.
 -- MYSQL - Add dynamic_offset in the database to force range for auto
    increment ids for the tres_table.
 -- MYSQL - Fix fallout from MySQL auto increment bug, see RELEASE_NOTES,
    only affects current 17.11 users tracking licenses or GRES in the database.
 -- Refactor logging logic to avoid possible memory corruption on non-x86
    architectures.
 -- Fix memory leak when getting jobs from the slurmdbd.
 -- Fix incorrect logic behind MemorySwappiness, and only set the value when
    specified in the configuration.

* Changes in Slurm 17.11.1-2
============================
 -- MYSQL - Make index for pack_job_id

* Changes in Slurm 17.11.1
==========================
 -- Fix --with-shared-libslurm option to work correctly.
 -- Make it so only daemons log errors on configuration option duplicates.
 -- Fix for ConstrainDevices=yes to work correctly.
 -- Fix to purge old jobs using burst buffer if slurmctld daemon restarted
    after the job's burst buffer work was already completed.
 -- Make logging prefix for slurmstepd to happen as soon as possible.
 -- mpi/pmix: Fix the job registration for the PMIx v2.1.
 -- Fix uid check for signaling a step with anything but SIGKILL.
 -- Fix uid check when requesting a jobid from a pid.
 -- Return ESLURM_TRANSITION_STATE_NO_UPDATE instead of EAGAIN when trying to
    signal a step that is still running a prolog.
 -- Update Cray slurm_playbook.yaml with latest recommended version.
 -- Only say a prolog is done running after the extern step is launched.
 -- Wait to start a batch step until the prolog and extern step are
    fully ran/launched.  Only matters if running with
    PrologFlags=[contain|alloc].
 -- Truncate a range for SlurmctldPort to FD_SETSIZE elements and throw an
    error, otherwise network traffic may be lost due to poll() not detecting
    traffic.
 -- Fix for srun --pack-group option that can reuse/corrupt memory.
 -- Fix handling ultra long hostlists in a hostfile.
 -- X11: fix xauth regex to handle '-' in hostnames again.
 -- Fix potential node reboot timeout problem for "scontrol reboot" command.
 -- Add ability for squeue to sort jobs by submit time.
 -- CRAY - Switch to standard pid files on Cray systems.
 -- Update jobcomp records on duplicate inserts.
 -- If unrecognized configuration file option found then print an appropriate
    fatal error message rather than relying upon random errno value.
 -- Initialize job_desc_msg_t's instead of just memset'ing them.
 -- Fix divide by zero when job requests no tasks and more memory than
    MaxMemPer{CPU|NODE}.
 -- Avoid changing Slurm internal errno on syslog() failures.
 -- BB - Only launch dependent jobs after the burst buffer is staged-out
    completely instead of right after the parent job finishes.
 -- node_features/knl_generic - If plugin can not fully load then do not spawn
    a background pthread (which will fail with invalid memory reference).
 -- Don't set the next jobid to give out to the highest jobid in the system on
    controller startup. Just use the checkpointed next use jobid.
 -- Docs - add Slurm/PMIx and OpenMPI build notes to the mpi_guide page.
 -- Add lustre_no_flush option to LaunchParameters for Native Cray systems.
 -- Fix rpmbuild issue with rpm 4.13+ / Fedora 25+.
 -- sacct - fix the display for the NNodes field when using the --units option.
 -- Prevent possible double-xfree on a buffer in stepd_completion.
 -- Fix for record job state on successful allocation but failed reply message.
 -- Fill in the user_name field for batch jobs if not sent by the slurmctld.
    (Which is the default behavior if PrologFlags=send_gids is not enabled.)
    This prevents job launch problems for sites using UsePAM=1.
 -- Handle syncing federated jobs that ran on non-origin clusters and were
    cancelled while the origin cluster was down.
 -- Fix accessing variable outside of lock.
 -- slurm.spec: move libpmi to a separate package to solve a conflict with the
    version provided by PMIx. This will require a separate change to PMIx as
    well.
 -- X11 forwarding: change xauth handling to use hostname/unix:display format,
    rather than localhost:display.
 -- mpi/pmix - Fix warning if not compiling with debug.

* Changes in Slurm 17.11.0
==========================
 -- Fix documentation for MaxQueryTimeRange option in slurmdbd.conf.
 -- Avoid srun abort trying to run on heterogeneous job component that has
    ended.
 -- Add SLURM_PACK_JOB_ID,SLURM_PACK_JOB_OFFSET to PrologSlurmctld and
    EpilogSlurmctld environment.
 -- Treat ":" in #SBATCH arguments as fatal error. The "#SBATCH packjob" syntax
    must be used instead.
 -- job_submit/lua plugin: expose pack_job fields to get.
 -- Prevent scheduling deadlock with multiple components of heterogeneous job
    in different partitions (i.e. one heterogeneous job component is higher
    priority in one partition and another component is lower priority in a
    different partition).
 -- Fix for heterogeneous job starvation bug.
 -- Fix some slurmctld memory leaks.
 -- Add SLURM_PACK_JOB_NODELIST to PrologSlurmctld and EpilogSlurmctld
    environment.
 -- If PrologSlurmctld fails for pack job leader then requeue or kill all
    components of the job.
 -- Fix for mulitple --pack-group srun arguments given out of order.
 -- Update slurm.conf(5) man page with updated example logrotate script.
 -- Add SchedulerParameters=whole_pack configuration parameter. If set, then
    hold, release and cancel operations on any component of a heterogeneous job