Newer
Older
This file describes changes in recent versions of SLURM. It primarily
documents those changes that are of interest to users and admins.
* Changes in Slurm 14.03.0pre7
==============================
-- Run job scheduling logic immediately when nodes enter service.
-- Added sbatch '--parsable' option to output only the job id number and the
cluster name separated by a semicolon. Errors will still be displayed.
-- Added failure management "slurmctld/nonstop" plugin.
-- Prevent jobs being killed when a checkpoint plugin is enabled or disabled.
-- Update the documentation about SLURM_PMI_KVS_NO_DUP_KEYS environment
variable.
-- select/cons_res bug fix for range of node counts with --cpus-per-task
option (e.g. "srun -N2-3 -c2 hostname" would allocate 2 CPUs on the first
node and 0 CPUs on the second node).
* Changes in Slurm 14.03.0pre6
==============================
-- Modify slurmstepd to log messages according to the LogTimeFormat
parameter in slurm.conf.
-- Insure that overlapping reservations do not oversubscribe available
licenses.
-- Added core specialization logic to select/cons_res plugin.
-- Added whole_node field to job_resources structure and enable gang scheduling
for jobs with core specialization.
-- When using FastSchedule = 1 the nodes with less than configured resources
are not longer set DOWN, they are set to DRAIN instead.
-- Modified 'sacctmgr show associations' command to show GrpCPURunMins
by default.
-- Replace the hostlist_push() function with a more efficient
hostlist_push_host().
-- Modify the reading of lustre file system statistics to print more
information when debug and when io error occur.
-- Add specialized core count field to job credential data.
NOTE: This changes the communications protocol from other pre-releases of
version 14.03. All programs must be cancelled and daemons upgraded from
previous pre-releases of version 14.03. Upgrades from version 2.6 or earlier
can take place without loss of jobs
-- Add version number to node and front-end configuration information visible
using the scontrol tool.
-- Add idea of a RESERVED flag for node state so idle resources are marked
not "idle" when in a reservation.
-- Added core specialization plugin infrastructure.
-- Added new job_submit/trottle plugin to control the rate at which a user
can submit jobs.
-- CRAY - added network performance counters option.
-- Allow scontrol suspend/resume to accept jobid in the format jobid_taskid
to suspend/resume array elements.
-- In the slurmctld job record, split "shared" variable into "share_res" (share
resource) and "whole_node" fields.
-- Fix the format of SLURM_STEP_RESV_PORTS. It was generated incorrectly
when using the hostlist_push_host function and input surrounded by [].
-- Modify the srun --slurmd-debug option to accept debug string tags
(quiet, fatal, error, info verbose) beside the numerical values.
-- Fix the bug where --cpu_bind=map_cpu is interpreted as mask_cpu.
-- Update the documentation egarding the state of cpu frequencies after
a step using --cpu-freq completes.
-- CRAY - Fix issue when a job is requeued and nhc is still running as it is
being scheduled to run again. This would erase the previous job info
that was still needed to clean up the nodes from the previous job run.
(Bug 526).
-- Set SLURM_JOB_PARTITION environment variable set for all job allocations.
-- Set SLURM_JOB_PARTITION environment variable for Prolog program.
-- Added SchedulerParameters option of partition_job_depth to limit scheduling
logic depth by partition.
-- Handle the case in which errno is not reset to 0 after calling
getgrent_r(), which causes the controller to core dump.
-- Added squeue format option of "%X" (core specialization count).
-- Added core specialization web page (just a start for now).
-- Added the SLURM_ARRAY_JOB_ID and SLURM_ARRAY_TASK_ID
-- Fix bug in job step allocation failing due to memory limit.
-- Modify the pbsnodes script to reflect its output on a TORQUE system.
-- Add ability to clear a node's DRAIN flag using scontrol or sview by setting
it's state to "UNDRAIN". The node's base state (e.g. "DOWN" or "IDLE") will
not be changed.
-- Modify the output of 'scontrol show partition' by displaying
DefMemPerCPU=UNLIMITED and MaxMemPerCPU=UNLIMITED when these limits are
configured as 0.
-- mpirun-mic - Major re-write of the command wrapper for Xeon Phi use.
-- Add new configuration parameter of AuthInfo to specify port used by
authentication plugin.
-- Corrected slurmstepd ident name when logging to syslog.
-- Fixed sh5util loop when there are no node-step files.
-- Add SLURM_CLUSTER_NAME to environment variables passed to PrologSlurmctld,
Prolog, EpilogSlurmctld, and Epilog
-- Add the idea of running a prolog right when an allocation happens
instead of when running on the node for the first time.
-- If user runs 'scontrol reconfig' but hostnames or the host count changes
the slurmctld throws a fatal error.
-- gres.conf - Add "NodeName" specification so that a single gres.conf file
can be used for a heterogeneous cluster.
-- Add flag to accounting RPC to indicate if job data is packed or not.
-- After all srun tasks have terminated on a node close the stdout/stderr
channel with the slurmstepd on that node.
-- In case of i/o error with slurmstepd log an error message and abort the
job.
-- Add --test-only option to sbatch command to validate the script and options.
The response includes expected start time and resources to be allocated.
==============================
-- Remove the ThreadID documentation from slurm.conf. This functionality has
been obsoleted by the LogTimeFormat.
-- Sched plugins - rename global and plugin functions names for consistency
with other plugin types.
-- BGQ - Added RebootQOSList option to bluegene.conf to allow an implicate
reboot of a block if only jobs in the list are running on it when cnodes
go into a failure state.
-- Correct task count of pending job steps.
-- Improve limit enforcement for jobs, set RLIMIT_RSS, RLIMIT_AS and/or
RLIMIT_DATA to enforce memory limit.
-- Pending job steps will have step_id of INFINITE rather than NO_VAL and
will be reported as "TBD" by scontrol and squeue commands.
-- Add logic so PMI_Abort or PMI2_Abort can propagate an exit code.
-- Added SlurmdPlugstack configuration parameter.
-- Added PriorityFlag DEPTH_OBLIVIOUS to have the depth of an association
not effect it's priorty.
-- Multi-thread the sinfo command (one thread per partition).
-- Added sgather tool to gather files from a job's compute nodes into a
central location.
-- Added configuration parameter FairShareDampeningFactor to offer a greater
priority range based upon utilization.
-- Change MaxArraySize and job's array_task_id from 16-bit to 32-bit field.
Additional Slurm enhancements are be required to support larger job arrays.
-- Added -S/--core-spec option to salloc, sbatch and srun commands to reserve
specialized cores for system use. Modify scontrol and sview to get/set
the new field. No enforcement exists yet for these new options.
struct job_info / slurm_job_info_t: Added core_spec
struct job_descriptorjob_desc_msg_t: Added core_spec
-- Do not set SLURM_NODEID environment variable on front-end systems.
-- Convert bitmap functions to use int32_t instead of int in data structures
and function arguments. This is to reliably enable use of bitmaps containing
up to 4 billion elements. Several data structures containing index values
were also changed from data type int to int32_t:
- Struct job_info / slurm_job_info_t: Changed exc_node_inx, node_inx, and
req_node_inx from type int to type int32_t
- job_step_info_t: Changed node_inx from type int to type int32_t
- Struct partition_info / partition_info_t: Changed node_inx from type int
to type int32_t
- block_job_info_t: Changed cnode_inx from type int to type int32_t
- block_info_t: Changed ionode_inx and mp_inx from type int to type int32_t
- Struct reserve_info / reserve_info_t: Changed node_inx from type int to
type int32_t
-- Modify qsub wrapper output to match torque command output, just print the
job ID rather than "Submitted batch job #"
-- Change Slurm error string for ESLURM_MISSING_TIME_LIMIT from
"Missing time limit" to
"Time limit specification required, but not provided"
-- Change salloc job_allocate error message header from
"Failed to allocate resources" to
"Job submit/allocate failed"
-- Modify slurmctld message retry logic to support Cray cold-standby SDB.
-- Added "JobAcctGatherParams" configuration parameter. Value of "NoShare"
disables accounting for shared memory.
-- Added fields to "scontrol show job" output: boards_per_node,
sockets_per_board, ntasks_per_node, ntasks_per_board, ntasks_per_socket,
ntasks_per_core, and nice.
-- Add squeue output format options for job command and working directory
(%o and %Z respectively).
-- Add stdin/out/err to sview job output.
-- Add new job_state of JOB_BOOT_FAIL for job terminations due to failure to
boot it's allocated nodes or BlueGene block.
-- CRAY - Add SelectTypeParameters NHC_NO_STEPS and NHC_NO which will disable
the node health check script for steps and allocations respectfully.
-- Reservation with CoreCnt: Avoid possible invalid memory reference.
-- Add new error code for attempt to create a reservation with duplicate name.
-- Validate that a hostlist file contains text (i.e. not a binary).
-- switch/generic - propagate switch information from srun down to slurmd and
slurmstepd.
-- CRAY - Do not package Slurm's libpmi or libpmi2 libraries. The Cray version
of those libraries must be used.
-- Added a new option to the scontrol command to view licenses that are
configured in use and avalable. 'scontrol show licenses'.
-- MySQL - Made Slurm compatible with 5.6
==============================
-- Add task pointer to the task_post_term() function in task plugins. The
terminating task's PID is available in task->pid.
-- Defer sending SIGKILL signal to processes while core dump in progress.
-- Added JobContainerPlugin configuration parameter and plugin infrastructure.
-- Added partition configuration parameters AllowAccounts, AllowQOS,
DenyAccounts and DenyQOS.
-- The rpmbuild option for a cray system with ALPS has changed from
%_with_cray to %_with_cray_alps.
-- The log file timestamp format can now be selected at runtime via the
LogTimeFormat configuration option. See the slurm.conf and slurmdbd.conf
man pages for details.
-- Added switch/generic plugin to a job's convey network topology.
-- BLUEGENE - If block is in 'D' state or has more cnodes in error than
MaxBlockInError set the job wait reason appropriately.
-- API use: Generate an error return rather than fatal error and exit if the
configuraiton file is absent or invalid. This will permit Slurm APIs to be
more reliably used by other programs.
-- Add support for load-based scheduling, allocate jobs to nodes with the
largest number of available CPUs. Added SchedulingParameters paramter of
"CR_LLN" and partition parameter of "LLN=yes|no".
-- Added job_info() and step_info() functions to the gres plugins to extract
plugin specific fields from the job's or step's GRES data structure.
-- Added sbatch --signal option of "B:" to signal the batch shell rather than
only the spawned job steps.
-- Added sinfo and squeue format option of "%all" to print all fields available
for the data type with a vertical bar separating each field.
-- Add mechanism for job_submit plugin to generate error message for srun,
salloc or sbatch to stderr. New argument added to job_submit function in
the plugin.
-- Add StdIn, StdOut, and StdErr paths to job information dumped with
"scontrol show job".
-- Permit Slurm administrator to submit a batch job as any user.
-- Set a job's RLIMIT_AS limit based upon it's memory limit and VsizeFactor
configuration value.
-- Remove Postgres plugins
-- Make jobacct_gather/cgroup work correctly and also make all jobacct_gather
plugins more maintainable.
-- Proctrack/pgid - Add support for proctrack_p_plugin_get_pids() function.
-- Sched/backfill - Change default max_job_bf parameter from 50 to 100.
-- Added -I|--item-extract option to sh5util to extract data item from series.
* Changes in Slurm 2.6.7
========================
-- Properly enforce a job's cpus-per-task option when a job's allocation is
constrained on some nodes by the mem-per-cpu option.
-- Correct the slurm.conf man pages and checkpoint_blcr.html page
describing that jobs must be drained from cluster before deploying
-- Fix issue where if using munge and munge wasn't running and a slurmd
needed to forward a message, the slurmd would core dump.
-- Update srun.1 man page documenting the PMI2 support.
-- Fix slurmctld core dump when a jobs gets its QOS updated but there
is not a corresponding association.
-- If a job requires specific nodes and can not run due to those nodes being
busy, the main scheduling loop will block those specific nodes rather than
the entire queue/partition.
* Changes in Slurm 2.6.6
========================
-- sched/backfill - Fix bug that could result in failing to reserve resources
for high priority jobs.
-- Correct job RunTime if requeued from suspended state.
-- Reset job priority from zero (held) on manual resume from suspend state.
-- If FastSchedule=0 then do not DOWN a node with low memory or disk size.
-- Update sshare.1 man page making it consistent with sacctmgr.1.
-- Do not reset a job's priority when the slurmctld restarts if previously
set to some specific value.
-- sview - Fix regression where the Node tab wasn't able to add/remove columns.
-- Fix slurmstepd lock when job terminates inside the infiniband
network traffic accounting plugin.
-- Correct the documentation to read filesystem instead of Lustre. Update
the srun help.
-- Fix the acct_gather_filesystem_lustre.c to compute the Lustre accounting
data correctly accumulating differences between sampling intervals.
Fix the data structure mismatch between acct_gather_filesystem_lustre.c
and slurm_jobacct_gather.h which caused the hdf5 plugin to log incorrect
data.
-- Don't allow PMI_TIME to be zero which will cause floating exception.
-- Fix purging of old reservation errors in database.
-- MYSQL - If starting the plugin and the database isn't up attempt to
connect in a loop instead of producing a fatal.
-- BLUEGENE - If IONodesPerMP changes in bluegene.conf recalculate bitmaps
based on ionode count correctly on slurmctld restart.
-- Fix step allocation when some CPUs are not available due to memory limits.
This happens when one step is active and using memory that blocks the
scheduling of another step on a portion of the CPUs needed. The new step
is now delayed rather than aborting with "Requested node configuration is
not available".
-- Make sure node limits get assessed if no node count was given in request.
-- Removed obsolete slurm_terminate_job() API.
-- Update documentation about QOS limits
-- Retry task exit message from slurmstepd to srun on message timeout.
-- Correction to logic reserving all nodes in a specified partition.
-- Added support for selecting AMD GPU by setting GPU_DEVICE_ORDINAL env var.
-- Properly enforce GrpSubmit limit for job arrays.
-- CRAY - fix issue with using CR_ONE_TASK_PER_CORE
-- CRAY - fix memory leak when using accelerators
* Changes in Slurm 2.6.5
========================
-- Correction to hostlist parsing bug introduced in v2.6.4 for hostlists with
more than one numeric range in brackets (e.g. rack[0-3]_blade[0-63]").
-- Add notification if using proctrack/cgroup and task/cgroup when oom hits.
-- Corrections to advanced reservation logic with overlapping jobs.
-- job_submit/lua - add cpus_per_task field to those available.
-- Add cpu_load to the node information available using the Perl API.
-- Correct a job's GRES allocation data in accounting records for non-Cray
systems.
-- Substantial performance improvement for systems with Shared=YES or FORCE
and large numbers of running jobs (replace bubble sort with quick sort).
-- proctrack/cgroup - Add locking to prevent race condition where one job step
is ending for a user or job at the same time another job stepsis starting
and the user or job container is deleted from under the starting job step.
-- Fixed sh5util loop when there are no node-step files.
-- Fix race condition on batch job termination that could result in a job exit
code of 0xfffffffe if the slurmd on node zero registers its active jobs at
the same time that slurmstepd is recording the job's exit code.
-- Correct logic returning remaining job dependencies in job information
reported by scontrol and squeue. Eliminates vestigial descriptors with
no job ID values (e.g. "afterany").
-- Improve performance of REQUEST_JOB_INFO_SINGLE RPC by removing unnecessary
locks and use hash function to find the desired job.
-- jobcomp/filetxt - Reopen the file when slurmctld daemon is reconfigured
or gets SIGHUP.
-- Remove notice of CVE with very old/deprecated versions of Slurm in
news.html.
-- Fix if hwloc_get_nbobjs_by_type() returns zero core count (set to 1).
-- Added ApbasilTimeout parameter to the cray.conf configuration file.
-- Handle in the API if parts of the node structure are NULL.
-- Fix srun hang when IO fails to start at launch.
-- Fix for GRES bitmap not matching the GRES count resulting in abort
(requires manual resetting of GRES count, changes to gres.conf file,
and slurmd restarts).
-- Modify sview to better support job arrays.
-- Modify squeue to support longer job ID values (for many job array tasks).
-- Fix race condition in authentication credential creation that could corrupt
memory. (NOTE: This race condition has existed since 2003 and would be
exceedingly rare.)
-- Slurmstepd variable initialization - Without this patch, free() is called
on a random memory location (i.e. whatever is on the stack), which can
result in slurmstepd dying and a completed job not being purged in a
timely fashion.
-- Fix slurmstepd race condition when separate threads are reading and
modifying the job's environment, which can result in the slurmstepd failing
with an invalid memory reference.
-- Fix erroneous error messages when running gang scheduling.
-- Fix minor memory leak.
-- scontrol modified to suspend, resume, hold, uhold, or release multiple
jobs in a space separated list.
-- Minor debug error when a connection goes away at the end of a job.
-- Validate return code from calls to slurm_get_peer_addr
-- BGQ - Fix issues with making sure all cnodes are accounted for when mulitple
steps cause multiple cnodes in one allocation to go into error at the
same time.
-- scontrol show job - Correct NumNodes value calculated based upon job
specifications.
-- BGQ - Fix issue if user runs multiple sub-block jobs inside a multiple
midplane block that starts on a higher coordinate than it ends (i.e if a
block has midplanes [0010,0013] 0013 is the start even though it is
listed second in the hostlist).
-- BGQ - Add midplane to the total_cnodes used in the runjob_mux plugin
for better debug.
-- Update AllocNodes paragraph in slurm.conf.5.
* Changes in Slurm 2.6.4
========================
-- Honor ntasks-per-node option with exclusive node allocations.
-- sched/backfill - Prevent invalid memory reference if bf_continue option is
configured and slurm is reconfigured during one of the sleep cycles or if
there are any changes to the partition configuration or if the normal
scheduler runs and starts a job that the backfill scheduler is actively
working on.
-- Update man pages information about acct-freq and JobAcctGatherFrequency
to reflect only the latest supported format.
-- Minor document update to include note about PrivateData=Usage for the
slurm.conf when using the DBD.
-- Expand information reported with DebugFlags=backfill.
-- Initiate jobs pending to run in a reservation as soon as the reservation
becomes active.
-- Purged expired reservation even if it has pending jobs.
-- Corrections to calculation of a pending job's expected start time.
-- Remove some vestigial logic treating job priority of 1 as a special case.
-- Memory freeing up to avoid minor memory leaks at close of daemons
-- Updated documentation to give correct units being displayed.
-- Report AccountingStorageBackupHost with "scontrol show config".
-- init scripts ignore quotes around Pid file name specifications.
-- Fixed typo about command case in quickstart.html.
-- task/cgroup - handle new cpuset files, similar to commit c4223940.
-- Replace the tempname() function call with mkstemp().
-- Fix for --cpu_bind=map_cpu/mask_cpu/map_ldom/mask_ldom plus
--mem_bind=map_mem/mask_mem options, broken in 2.6.2.
-- Restore default behavior of allocating cores to jobs on a cyclic basis
across the sockets unless SelectTypeParameters=CR_CORE_DEFAULT_DIST_BLOCK
or user specifies other distribution options.
-- Enforce JobRequeue configuration parameter on node failure. Previously
always requeued the job.
-- acct_gather_energy/ipmi - Add delay before retry on read error.
-- select/cons_res with GRES and multiple threads per core, fix possible
infinite loop.
-- proctrack/cgroup - Add cgroup create retry logic in case one step is
starting at the same time as another step is ending and the logic to create
and delete cgroups overlaps.
-- Improve setting of job wait "Reason" field.
-- Correct sbatch documentation and job_submit/pbs plugin "%j" is job ID,
not "%J" (which is job_id.step_id).
-- Improvements to sinfo performance, especially for large numbers of
partitions.
-- SlurmdDebug - Permit changes to slurmd debug level with "scontrol reconfig"
-- smap - Avoid invalid memory reference with hidden nodes.
-- Fix sacctmgr modify qos set preempt+/-=.
-- BLUEGENE - fix issue where node count wasn't set up correctly when srun
preforms the allocation, regression in 2.6.3.
-- Add support for dependencies of job array elements (e.g.
"sbatch --depend=afterok:123_4 ...") or all elements of a job array (e.g.
"sbatch --depend=afterok:123 ...").
-- Add support for new options in sbatch qsub wrapper:
-W block=true (wait for job completion)
Clear PBS_NODEFILE environment variable
-- Fixed the MaxSubmitJobsPerUser limit in QOS which limited submissions
a job too early.
-- sched/wiki, sched/wiki2 - Fix to work with change logic introduced in
version 2.6.3 preventing Maui/Moab from starting jobs.
-- Updated the QOS limits documentation and man page.
* Changes in Slurm 2.6.3
========================
-- Add support for some new #PBS options in sbatch scripts and qsub wrapper:
-l accelerator=true|false (GPU use)
-l mpiprocs=# (processors per node)
-l naccelerators=# (GPU count)
-l select=# (node count)
-l ncpus=# (task count)
-v key=value (environment variable)
-W depend=opts (job dependencies, including "on" and "before" options)
-W umask=# (set job's umask)
-- Added qalter and qrerun commands to torque package.
-- Corrections to qstat logic: job CPU count and partition time format.
-- Add job_submit/pbs plugin to translate PBS job dependency options to the
extend possible (no support for PBS "before" options) and set some PBS
environment variables.
-- Add spank/pbs plugin to set a bunch of PBS environment variables.
-- Backported sh5util from master to 2.6 as there are some important
bugfixes and the new item extraction feature.
-- select/cons_res - Correct MacCPUsPerNode partition constraint for CR_Socket.
-- scontrol - for setdebugflags command, avoid parsing "-flagname" as an
scontrol command line option.
-- Fix issue with step accounting if a job is requeued.
-- Close file descriptors on exec of prolog, epilog, etc.
-- Fix issue when a user has held a job and then sets the begin time
into the future.
-- Scontrol - Enable changing a job's stdout file.
-- Fix issues where memory or node count of a srun job is altered while the
srun is pending. The step creation would use the old values and possibly
hang srun since the step wouldn't be able to be created in the modified
allocation.
-- Add support for new SchedulerParameters value of "bf_max_job_part", the
maximum depth the backfill scheduler should go in any single partition.
-- acct_gather/infiniband plugin - Correct packets_in/out values.
-- BLUEGENE - Don't ignore a conn-type request from the user.
-- BGQ - Force a request on a Q for a MESH to be a TORUS in a dimension that
can only be a TORUS (1).
-- Change max message length from 100MB to 1GB before generating "Insane
message length" error.
-- sched/backfill - Prevent possible memory corruption due to use of
bf_continue option and long running scheduling cycle (pending jobs could
have been cancelled and purged).
-- CRAY - fix AcceleratorAllocation depth correctly for basil 1.3
-- Created the environment variable SLURM_JOB_NUM_NODES for srun jobs and
updated the srun man page.
-- BLUEGENE/CRAY - Don't set env variables that pertain to a node when Slurm
isn't doing the launching.
-- gres/gpu and gres/mic - Do not treat the existence of an empty gres.conf
file as a fatal error.
-- Fixed for if hours are specified as 0 the time days-0:min specification
is not parsed correctly.
-- Subtract the PMII_COMMANDLEN_SIZE in contribs/pmi2/pmi2_api.c to prevent
certain implementation of snprintf() to segfault.
* Changes in Slurm 2.6.2
========================
-- Fix issue with reconfig and GrpCPURunMins
-- Fix of wrong node/job state problem after reconfig
-- Allow users who are coordinators update their own limits in the accounts
they are coordinators over.
-- BackupController - Make sure we have a connection to the DBD first thing
to avoid it thinking we don't have a cluster name.
-- Correct value of min_nodes returned by loading job information to consider
the job's task count and maximum CPUs per node.
-- If running jobacct_gather/none fix issue on unpacking step completion.
-- Reservation with CoreCnt: Avoid possible invalid memory reference.
-- sjstat - Add man page when generating rpms.
-- Make sure GrpCPURunMins is added when creating a user, account or QOS with
sacctmgr.
-- Fix for invalid memory reference due to multiple free calls caused by
job arrays submitted to multiple partitions.
-- Enforce --ntasks-per-socket=1 job option when allocating by socket.
-- Validate permissions of key directories at slurmctld startup. Report
anything that is world writable.
-- Improve GRES support for CPU topology. Previous logic would pick CPUs then
reject jobs that can not match GRES to the allocated CPUs. New logic first
filters out CPUs that can not use the GRES, next picks CPUs for the job,
and finally picks the GRES that best match those CPUs.
-- Switch/nrt - Prevent invalid memory reference when allocating single adapter
per node of specific adapter type
-- CRAY - Make Slurm work with CLE 5.1.1
-- Fix segfault if submitting to multiple partitions and holding the job.
-- Use MAXPATHLEN instead of the hardcoded value 1024 for maximum file path
lengths.
-- If OverTimeLimit is defined do not declare failed those jobs that ended
in the OverTimeLimit interval.
* Changes in Slurm 2.6.1
========================
-- slurmdbd - Allow job derived ec and comments to be modified by non-root
users.
-- Fix issue with job name being truncated to 24 chars when sending a mail
message.
-- Fix minor issues with spec file, missing files and including files
erroneously on a bluegene system.
-- sacct - fix --name and --partition options when using
accounting_storage/filetxt.
-- squeue - Remove extra whitespace of default printout.
-- BGQ - added head ppcfloor as an include dir when building.
-- BGQ - Better debug messages in runjob_mux plugin.
-- PMI2 Updated the Makefile.am to build a versioned library.
-- CRAY - Fix srun --mem_bind=local option with launch/aprun.
-- PMI2 Corrected buffer size computation in the pmi2_api.c module.
-- GRES accounting data wrong in database: gres_alloc, gres_req, and gres_used
fields were empty if the job was not started immediately.
-- Fix sbatch and srun task count logic when --ntasks-per-node specified,
but no explicit task count.
-- Corrected the hdf5 profile user guide and the acct_gather.conf
documentation.
-- IPMI - Fix Math bug getting new wattage.
-- Corrected the AcctGatherProfileType documentation in slurm.conf
-- Corrected the sh5util program to print the header in the csv file
only once, set the debug messages at debug() level, make the argument
check case insensitive and avoid printing duplicate \n.
-- If cannot collect energy values send message to the controller
to drain the node and log error slurmd log file.
-- Handle complete removal of CPURunMins time at the end of the job instead
of at multifactor poll.
-- sview - Add missing debug_flag options.
-- PGSQL - Notes about Postgres functionality being removed in the next
version of Slurm.
-- MYSQL - fix issue when rolling up usage and events happened when a cluster
was down (slurmctld not running) during that time period.
-- sched/wiki2 - Insure that Moab gets current CPU load information.
-- Prevent infinite loop in parsing configuration if including file containing
one blank line.
-- Fix pack and unpack between 2.6 and 2.5.
-- Fix job state recovery logic in which a job's accounting frequency was
not set. This would result in a value of 65534 seconds being used (the
equivalent of NO_VAL in uint16_t), which could result in the job being
requeued or aborted.
-- Validate a job's accounting frequency at submission time rather than
waiting for it's initiation to possibly fail.
-- Fix CPURunMins if a job is requeued from a failed launch.
-- Fix in accounting_storage/filetxt to correct start times which sometimes
could end up before the job started.
-- Fix issue with potentially referencing past an array in parse_time()
-- CRAY - fix issue with accelerators on a cray when parsing BASIL 1.3 XML.
-- Fix issue with a 2.5 slurmstepd locking up when talking to a 2.6 slurmd.
-- Add argument to priority plugin's priority_p_reconfig function to note
when the association and QOS used_cpu_run_secs field has been reset.
* Changes in Slurm 2.6.0
========================
-- Fix it so bluegene and serial systems don't get warnings over new NODEDATA
enum.
-- When a job is aborted send a message for any tasks that have completed.
-- Correction to memory per CPU calculation on system with threads and
allocating cores or sockets.
-- Requeue batch job if it's node reboots (used to abort the job).
-- Enlarge maximum size of srun's hostlist file.
-- IPMI - Fix first poll to get correct consumed_energy for a step.
-- Correction to job state recovery logic that could result in assert failure.
-- Record partial step accounting record if allocated nodes fail abnormally.
Francois Diakhate
committed
-- Accounting - fix issue where PrivateData=jobs or users could potentially
show information to users that had no associations on the system.
-- Make PrivateData in slurmdbd.conf case insensitive.
-- sacct/sstat - Add format option ConsumedEnergyRaw to print full energy
values.
* Changes in Slurm 2.6.0rc2
===========================
-- HDF5 - Fix issue with Ubuntu where HDF5 development headers are
overwritten by the parallel versions thus making it so we need handle
both cases.
-- ACCT_GATHER - handle suspending correctly for polling threads.
-- Make SLURM_DISTRIBUTION env var hold both types of distribution if
specified.
-- Remove hardcoded /usr/local from slurm.spec.
-- Modify slurmctld locking to improve performance under heavy load with
very large numbers of batch job submissions or job cancellations.
-- sstat - Fix issue where if -j wasn't given allow last argument to be checked
for as the job/step id.
-- IPMI - fix adjustment on poll when using EnergyIPMICalcAdjustment.
* Changes in Slurm 2.6.0rc1
===========================
-- Added helper script for launching symmetric and MIC-only MPI tasks within
SLURM (in contribs/mic/mpirun-mic).
-- Change maximum delay for state save from 2 secs to 5 secs. Make timeout
configurable at build time by defining SAVE_MAX_WAIT.
-- Modify slurmctld data structure locking to interleave read and write
locks rather than always favor write locks over read locks.
-- Added sacct format option of "ALL" to print all fields.
-- Deprecate the SchedulerParameters value of "interval" use "bf_interval"
instead as documented.
-- Add acct_gather_profile/hdf5 to profile jobs with hdf5
-- Added MaxCPUsPerNode partition configuration parameter. This can be
especially useful to schedule systems with GPUs.
-- Permit "scontrol reboot_node" for nodes in MAINT reservation.
-- Added "PriorityFlags" value of "SMALL_RELATIVE_TO_TIME". If set, the job's
size component will be based upon not the job size alone, but the job's
size divided by it's time limit.
Morris Jette
committed
-- Added sbatch option "--ignore-pbs" to ignore "#PBS" options in the batch
script.
-- Rename slurm_step_ctx_params_t field from "mem_per_cpu" to "pn_min_memory".
Job step now accepts memory specification in either per-cpu or per-node
basis.
-- Add ability to specify host repitition count in the srun hostfile (e.g.
"host1*2" is equivalent to "host1,host1").
* Changes in Slurm 2.6.0pre3
============================
-- Add milliseconds to default log message header (both RFC 5424 and ISO 8601
time formats). Disable milliseconds logging using the configure
parameter "--disable-log-time-msec". Default time format changes to
ISO 8601 (without time zone information). Specify "--enable-rfc5424time"
to restore the time zone information.
-- Add username (%u) to the filename pattern in the batch script.
-- Added options for front end nodes of AllowGroups, AllowUsers, DenyGroups,
and DenyUsers.
-- Fix sched/backfill logic to initiate jobs with maximum time limit over the
partition limit, but the minimum time limit permits it to start.
-- gres/gpu - Fix for gres.conf file with multiple files on a single line
using a slurm expression (e.g. "File=/dev/nvidia[0-1]").
-- Replaced ipmi.conf with generic acct_gather.conf file for all acct_gather
plugins. For those doing development to use this follow the model set
forth in the acct_gather_energy_ipmi plugin.
-- Added more options to update a step's information
-- Add DebugFlags=ThreadID which will print the thread id of the calling
thread.
-- CRAY - Allocate whole node (CPUs) in reservation despite what the
user requests. We have found any srun/aprun afterwards will work on a
subset of resources.
-- Do not purge inactive interactive jobs that lack a port to ping (added
for MR+ operation).
-- Advanced reservations with hostname and core counts now supports asymetric
reservations (e.g. specific different core count for each node).
-- Added slurmctld/dynalloc plugin for MapReduce+ support.
-- Added "DynAllocPort" configuration parameter.
-- Added partition paramter of SelectTypeParameters to override system-wide
value.
-- Added cr_type to partition_info data structure.
-- Added allocated memory to node information available (within the existing
select_nodeinfo field of the node_info_t data structure). Added Allocated
Memory to node information displayed by sview and scontrol commands.
-- Make sched/backfill the default scheduling plugin rather than sched/builtin
(FIFO).
Alejandro Lucero Palau
committed
-- Added support for a job having different priorities in different partitions.
-- Added new SchedulerParameters configuration parameter of "bf_continue"
which permits the backfill scheduler to continue considering jobs for
backfill scheduling after yielding locks even if new jobs have been
submitted. This can result in lower priority jobs from being backfill
scheduled instead of newly arrived higher priority jobs, but will permit
more queued jobs to be considered for backfill scheduling.
-- Added support to purge reservation records from accounting.
-- Cray - Add support for Basil 1.3
* Changes in SLURM 2.6.0pre1
============================
-- Add "state" field to job step information reported by scontrol.
-- Notify srun to retry step creation upon completion of other job steps
rather than polling. This results in much faster throughput for job step
execution with --exclusive option.
-- Added "ResvEpilog" and "ResvProlog" configuration parameters to execute a
program at the beginning and end of each reservation.
-- Added "slurm_load_job_user" function. This is a variation of
"slurm_load_jobs", but accepts a user ID argument, potentially resulting
in substantial performance improvement for "squeue --user=ID"
-- Added "slurm_load_node_single" function. This is a variation of
"slurm_load_nodes", but accepts a node name argument, potentially resulting
in substantial performance improvement for "sinfo --nodes=NAME".
-- Added "HealthCheckNodeState" configuration parameter identify node states
on which HealthCheckProgram should be executed.
-- Remove sacct --dump --formatted-dump options which were deprecated in
2.5.
-- Added support for job arrays (phase 1 of effort). See "man sbatch" option
-a/--array for details.
-- Add new AccountStorageEnforce options of 'nojobs' and 'nosteps' which will
allow the use of accounting features like associations, qos and limits but
not keep track of jobs or steps in accounting.
-- Cray - Add new cray.conf parameter of "AlpsEngine" to specify the
communication protocol to be used for ALPS/BASIL.
-- select/cons_res plugin: Correction to CPU allocation count logic in for
cores without hyperthreading.
-- Added new SelectTypeParameter value of "CR_ALLOCATE_FULL_SOCKET".
-- Added PriorityFlags value of "TICKET_BASED" and merged priority/multifactor2
plugin into priority/multifactor plugin.
Morris Jette
committed
-- Add "KeepAliveTime" configuration parameter controlling how long sockets
used for srun/slurmstepd communications are kept alive after disconnect.
-- Added SLURM_SUBMIT_HOST to salloc, sbatch and srun job environment.
-- Added SLURM_ARRAY_TASK_ID to environment of job array.
-- Added squeue --array/-r option to optimize output for job arrays.
-- Added "SlurmctldPlugstack" configuration parameter for generic stack of
slurmctld daemon plugins.
-- Removed contribs/arrayrun tool. Use native support for job arrays.
-- Modify default installation locations for RPMs to match "make install":
_prefix /usr/local
_slurm_sysconfdir %{_prefix}/etc/slurm
_mandir %{_prefix}/share/man
_infodir %{_prefix}/share/info
-- Add acct_gather_energy/ipmi which works off freeipmi for energy gathering
* Changes in Slurm 2.5.8
========================
-- Fix for slurmctld segfault on NULL front-end reason field.
-- Avoid gres step allocation errors when a job shrinks in size due to either
down nodes or explicit resizing. Generated slurmctld errors of this type:
"step_test ... gres_bit_alloc is NULL"
-- Fix bug that would leak memory and over-write the AllowGroups field if on
"scontrol reconfig" when AllowNodes is manually changed using scontrol.
-- Get html/man files to install in correct places with rpms.
-- Remove --program-prefix from spec file since it appears to be added by
default and appeared to break other things.
-- Updated the automake min version in autogen.sh to be correct.
-- Select/cons_res - Correct total CPU count allocated to a job with
--exclusive and --cpus-per-task options
-- switch/nrt - Don't allocate network resources unless job step has 2+ nodes.
-- select/cons_res - Avoid extraneous "oversubscribe" error messages.
-- Reorder get config logic to avoid deadlock.
-- Enforce QOS MaxCPUsMin limit when job submission contains no user-specified
time limit.
-- EpilogSlurmctld pthread is passed required arguments rather than a pointer
to the job record, which under some conditions could be purged and result
in an invalid memory reference.
* Changes in Slurm 2.5.7
========================
-- Fix for linking to the select/cray plugin to not give warning about
undefined variable.
-- Add missing symbols to the xlator.h
-- Avoid placing pending jobs in AdminHold state due to backfill scheduler
interactions with advanced reservation.
-- Accounting - make average by task not cpu.
-- CRAY - Change logging of transient ALPS errors from error() to debug().
-- POE - Correct logic to support poe option "-euidevice sn_all" and
"-euidevice sn_single".
-- Accounting - Fix minor initialization error.
-- POE - Correct logic to support srun network instances count with POE.
-- POE - With the srun --launch-cmd option, report proper task count when
the --cpus-per-task option is used without the --ntasks option.
-- POE - Fix logic binding tasks to CPUs.
-- sview - Fix race condition where new information could of slipped past
the node tab and we didn't notice.
-- Accounting - Fix an invalid memory read when slurmctld sends data about
-- If a prolog or epilog failure occurs, drain the node rather than setting it
down and killing all of its jobs.
-- Priority/multifactor - Avoid underflow in half-life calculation.
-- POE - pack missing variable to allow fanout (more than 32 nodes)
-- Prevent clearing reason field for pending jobs. This bug was introduced in
v2.5.5 (see "Reject job at submit time ...").
-- BGQ - Fix issue with preemption on sub-block jobs where a job would kill
all preemptable jobs on the midplane instead of just the ones it needed to.
-- switch/nrt - Validate dynamic window allocation size.
-- BGQ - When --geo is requested do not impose the default conn_types.
-- RebootNode logic - Defers (rather than forgets) reboot request with job
running on the node within a reservation.
-- switch/nrt - Correct network_id use logic. Correct support for user sn_all
and sn_single options.
-- sched/backfill - Modify logic to reduce overhead under heavy load.
-- Fix job step allocation with --exclusive and --hostlist option.
-- Select/cons_res - Fix bug resulting in error of "cons_res: sync loop not
progressing, holding job #"
-- checkpoint/blcr - Reset max_nodes from zero to NO_VAL on job restart.
-- launch/poe - Fix for hostlist file support with repeated host names.
-- priority/multifactor2 - Prevent possible divide by zero.
-- srun - Don't check for executable if --test-only flag is used.
-- energy - On a single node only use the last task for gathering energy.
Since we don't currently track energy usage per task (only per step).
Otherwise we get double the energy.
-- Gres accounting - Fix regression in 2.5.5 for keeping track of gres
requested and allocated.
* Changes in Slurm 2.5.5
========================
-- Fix for sacctmgr add qos to handle the 'flags' option.
-- Export SLURM_ environment variables from sbatch, even if "--export"
option does not explicitly list them.
-- If node is in more than one partition, correct counting of allocated CPUs.
-- If step requests more CPUs than possible in specified node count of job
allocation then return ESLURM_TOO_MANY_REQUESTED_CPUS rather than
ESLURM_NODES_BUSY and retrying.
-- CRAY - Fix SLURM_TASKS_PER_NODE to be set correctly.
-- Accounting - more checks for strings with a possible `'` in it.
-- sreport - Fix by adding planned down time to utilization reports.
-- Do not report an error when sstat identifies job steps terminated during
its execution, but log using debug type message.
-- Select/cons_res - Permit node removed from job by going down to be returned
to service and re-used by another job.
-- Select/cons_res - Tighter packing of job allocations on sockets.
-- SlurmDBD - fix to allow user root along with the slurm user to register a
cluster.
-- Select/cons_res - Fix for support of consecutive node option.
-- Select/cray - Modify build to enable direct use of libslurm library.
-- Bug fixes related to job step allocation logic.
-- Cray - Disable enforcement of MaxTasksPerNode, which is not applicable
with launch/aprun.
-- Accounting - When rolling up data from past usage ignore "idle" time from
a reservation when it has the "Ignore_Jobs" flag set. Since jobs could run
outside of the reservation in it's nodes without this you could have
double time.
-- Accounting - Minor fix to avoid reuse of variable erroneously.
-- Reject job at submit time if the node count is invalid. Previously such a
job submitted to a DOWN partition would be queued.
-- Purge vestigial job scripts when the slurmd cold starts or slurmstepd
terminates abnormally.
-- Add sanity check for NULL cluster names trying to register.
-- BGQ - Push action 'D' info to scontrol for admins.
-- Reset a job's reason from PartitionDown when the partition is set up.
-- BGQ - Handle issue where blocks would have a pending job on them and
while it was free cnodes would go into software error and kill the job.
-- BGQ - Fix issue where if for some reason we are freeing a block with
a pending job on it we don't kill the job.
-- BGQ - Fix race condition were a job could of been removed from a block
without it still existing there. This is extremely rare.
-- BGQ - Fix for when a step completes in Slurm before the runjob_mux notifies
the slurmctld there were software errors on some nodes.
-- BGQ - Fix issue on state recover if block states are not around
and when reading in state from DB2 we find a block that can't be created.
You can now do a clean start to rid the bad block.
-- Modify slurmdbd to retransmit to slurmctld daemon if it is not responding.
-- BLUEGENE - Fix issue where when doing backfill preemptable jobs were
never looked at to determine eligibility of backfillable job.
-- Cray/BlueGene - Disable srun --pty option unless LaunchType=launch/slurm.
-- CRAY - Fix sanity check for systems with more than 32 cores per node.
-- CRAY - Remove other objects from MySQL query that are available from
the XML.
-- BLUEGENE - Set the geometry of a job when a block is picked and the job
isn't a sub-block job.
-- Cray - avoid check of macro versions of CLE for version 5.0.
-- CRAY - Fix memory issue with reading in the cray.conf file.
-- CRAY - If hostlist is given with srun make sure the node count is the same
as the hosts given.
-- CRAY - If task count specified, but no tasks-per-node, then set the tasks
per node in the BASIL reservation request.
-- CRAY - fix issue with --mem option not giving correct amount of memory
per cpu.
-- CRAY - Fix if srun --mem is given outside an allocation to set the
APRUN_DEFAULT_MEMORY env var for aprun. This scenario will not display
the option when used with --launch-cmd.
-- Change sview to use GMutex instead of GStaticMutex
-- CRAY - set APRUN_DEFAULT_MEMROY instead of CRAY_AUTO_APRUN_OPTIONS
-- sview - fix issue where if a partition was completely in one state the
cpu count would be reflected correctly.
-- BGQ - fix for handling half rack system in STATIC of OVERLAP mode to
implicitly create full system block.
-- CRAY - Dynamically create BASIL XML buffer to resize as needed.
-- Fix checking if QOS limit MaxCPUMinsPJ is set along with DenyOnLimit to
deny the job instead of holding it.
-- Make sure on systems that use a different launcher than launch/slurm not
to attempt to signal tasks on the frontend node.
-- Cray - when a step is requested count other steps running on nodes in the
allocation as taking up the entire node instead of just part of the node
allocated. And always enforce exclusive on a step request.
-- Cray - display correct nodelist, node/cpu count on steps.
* Changes in Slurm 2.5.4
========================
-- Fix bug in PrologSlurmctld use that would block job steps until node
responds.
Danny Auble
committed
-- CRAY - If a partition has MinNodes=0 and a batch job doesn't request nodes
put the allocation to 1 instead of 0 which prevents the allocation to
happen.
-- Better debug when the database is down and using the --cluster option in
the user commands.
-- When asking for job states with sacct, default to 'now' instead of midnight
of the current day.
-- Fix for handling a test-only job or immediate job that fails while being
built.
-- Comment out all of the logic in the job_submit/defaults plugin. The logic
is only an example and not meant for actual use.
-- Eliminate configuration file 4096 character line limitation.
-- More robust logic for tree message forward
-- BGQ - When cnodes fail in a timeout fashion correctly look up parent
midplane.
-- Correct sinfo "%c" (node's CPU count) output value for Bluegene systems.
-- Backfill - Responsive improvements for systems with large numbers of jobs
(>5000) and using the SchedulerParameters option bf_max_job_user.
-- slurmstepd: ensure that IO redirection openings from/to files correctly
handle interruption
-- BGQ - Able to handle when midplanes go into Hardware::SoftwareFailure
-- GRES - Correct tracking of specific resources used after slurmctld restart.
Counts would previously go negative as jobs terminate and decrement from
a base value of zero.
-- Fix for priority/multifactor2 plugin to not assert when configured with
--enable-debug.
-- Select/cons_res - If the job request specified --ntasks-per-socket and the
allocation using is cores, then pack the tasks onto the sockets up to the
specified value.
-- BGQ - If a cnode goes into an 'error' state and the block containing the
cnode does not have a job running on it do not resume the block.
-- BGQ - Handle blocks that don't free themselves in a reasonable time better.
-- BGQ - Fix for signaling steps when allocation ends before step.
-- Fix for backfill scheduling logic with job preemption; starts more jobs.
-- xcgroup - remove bugs with EINTR management in write calls
-- jobacct_gather - fix total values to not always == the max values.
-- Fix for handling node registration messages from older versions without
energy data.
-- BGQ - Allow user to request full dimensional mesh.
-- sdiag command - Correction to jobs started value reported.
-- Prevent slurmctld assert when invalid change to reservation with running
jobs is made.
-- BGQ - If signal is NODE_FAIL allow forward even if job is completing
and timeout in the runjob_mux trying to send in this situation.
-- BGQ - More robust checking for correct node, task, and ntasks-per-node
options in srun, and push that logic to salloc and sbatch.
-- GRES topology bug in core selection logic fixed.
-- Fix to handle init.d script for querying status and not return 1 on
success.
* Changes in SLURM 2.5.3
========================
-- Gres/gpu plugin - If no GPUs requested, set CUDA_VISIBLE_DEVICES=NoDevFiles.
This bug was introduced in 2.5.2 for the case where a GPU count was
configured, but without device files.
-- task/affinity plugin - Fix bug in CPU masks for some processors.
-- Modify sacct command to get format from SACCT_FORMAT environment variable.
-- BGQ - Changed order of library inclusions and fixed incorrect declaration
to compile correctly on newer compilers
-- Fix for not building sview if glib exists on a system but not the gtk libs.
-- BGQ - Fix for handling a job cleanup on a small block if the job has long
since left the system.
-- Fix race condition in job dependency logic which can result in invalid
memory reference.
* Changes in SLURM 2.5.2
========================
-- Fix advanced reservation recovery logic when upgrading from version 2.4.
-- BLUEGENE - fix for QOS/Association node limits.
-- Add missing "safe" flag from print of AccountStorageEnforce option.
-- Fix logic to optimize GRES topology with respect to allocated CPUs.
-- Add job_submit/all_partitions plugin to set a job's default partition
to ALL available partitions in the cluster.
-- Modify switch/nrt logic to permit build without libnrt.so library.
-- Handle srun task launch failure without duplicate error messages or abort.
Matthieu Hautreux
committed
-- Fix bug in QoS limits enforcement when slurmctld restarts and user not yet
added to the QOS list.
-- Fix issue where sjstat and sjobexitmod was installed in 2 different RPMs.
-- Fix for job request of multiple partitions in which some partitions lack
nodes with required features.
-- Permit a job to use a QOS they do not have access to if an administrator
manually set the job's QOS (previously the job would be rejected).
-- Make more variables available to job_submit/lua plugin: slurm.MEM_PER_CPU,
slurm.NO_VAL, etc.
-- Fix topology/tree logic when nodes defined in slurm.conf get re-ordered.
-- In select/cons_res, correct logic to allocate whole sockets to jobs. Work
by Magnus Jonsson, Umea University.
-- In select/cons_res, correct logic when job removed from only some nodes.
-- Avoid apparent kernel bug in 2.6.32 which apparently is solved in
at least 3.5.0. This avoids a stack overflow when running jobs on
more than 120k nodes.
-- BLUEGENE - If we made a block that isn't runnable because of a overlapping
block, destroy it correctly.
-- Switch/nrt - Dynamically load libnrt.so from within the plugin as needed.
This eliminates the need for libnrt.so on the head node.
-- BLUEGENE - Fix in reservation logic that could cause abort.
* Changes in SLURM 2.5.1
========================
-- Correction to hostlist sorting for hostnames that contain two numeric
components and the first numeric component has various sizes (e.g.
"rack9blade1" should come before "rack10blade1")
-- BGQ - Only poll on initialized blocks instead of calling getBlocks on
each block independently.
-- Fix of task/affinity plugin logic for Power7 processors having hyper-
threading disabled (cpu mask has gaps).
-- Fix of job priority ordering with sched/builtin and priority/multifactor.
-- CRAY - Fix for setting up the aprun for a large job (+2000 nodes).
Morris Jette
committed
-- Fix for race condition related to compute node boot resulting in node being
set down with reason of "Node <name> unexpectedly rebooted"
-- RAPL - Fix for handling errors when opening msr files.
-- BGQ - Fix for salloc/sbatch to do the correct allocation when asking for
-N1 -n#.
-- BGQ - in emulation make it so we can pretend to run large jobs (>64k nodes)
-- BLUEGENE - Correct method to update conn_type of a job.
-- BLUEGENE - Fix issue with preemption when needing to preempt multiple jobs
to make one job run.
-- Fixed issue where if an srun dies inside of an allocation abnormally it