diff --git a/RELEASE_NOTES b/RELEASE_NOTES index 7cdd9eeaf69550090f6f7bc26c7726084b872246..b1e55593f7564b77ec8f3ad362eb754d8c3b55fc 100644 --- a/RELEASE_NOTES +++ b/RELEASE_NOTES @@ -1,5 +1,5 @@ RELEASE NOTES FOR SLURM VERSION 16.05 -29 March 2016 (after 16.05.0-pre2) +27 April 2016 IMPORTANT NOTES: ANY JOBS WITH A JOB ID ABOVE 2,147,463,647 WILL BE PURGED WHEN SLURM IS @@ -91,6 +91,26 @@ CONFIGURATION FILE CHANGES (see man appropriate man page for details) connections. -- Added SchedulingParameters option of "bf_min_prio_reserve". Jobs below the specified threshold will not have resources reserved for them. + -- Add SchedulerParameter "no_env_cache", if set no environment cache will be + used when launching a job, instead the job will fail and drain the node if + the environment isn't loaded normally. + -- Remove the SchedulerParameters option of "assoc_limit_continue", making it + the default value. Add option of "assoc_limit_stop". If "assoc_limit_stop" + is set and a job cannot start due to association limits, then do not attempt + to initiate any lower priority jobs in that partition. Setting this can + decrease system throughput and utlization, but avoid potentially starving + larger jobs by preventing them from launching indefinitely. + -- Rename partition configuration from "Shared" to "OverSubscribe". Rename + salloc, sbatch, srun option from "--shared" to "--oversubscribe". The old + options will continue to function. Output field names also changed in + scontrol, sinfo, squeue and sview. + -- Add TopologyParam option of "TopoOptional" to optimize network topology + only for jobs requesting it. + -- Configuration parameter "CpuFreqDef" used to set default governor for job + step not specifying --cpu-freq (previously the parameter was unused) + -- Use TaskPluginParam for default task binding if no user specified CPU + binding. User --cpu_bind option takes precident over default. No longer + any error if user --cpu_bind option does not match TaskPluginParam. COMMAND CHANGES (see man pages for details) =========================================== @@ -117,6 +137,7 @@ COMMAND CHANGES (see man pages for details) -- Added "sacctmgr show lostjobs" to report any orphaned jobs in the database. -- Add reservation flag of "purge_comp" which will purge an advanced reservation once it has no more active (pending, suspended or running) jobs. + -- Add ARRAY_TASKS mail option to send emails to each task in a job array. OTHER CHANGES ============= @@ -136,6 +157,9 @@ OTHER CHANGES of Intel Knights Landing processor. -- If NodeHealthCheckProgram configured HealthCheckInterval is non-zero, then modify slurmd to run it before registering with slurmctld. + -- select/cray - Initiate step node health check at start of step termination + rather than after application completely ends so that NHC can capture + information about hung (non-killable) processes. API CHANGES ===========