- slurm-users - lists.schedmd.com

getting slurm going
by Steven Jones 08 Dec '24

08 Dec '24

What tests can I do to prove that slurm is talking to the nodes pls? regards Steven

3 5

Non-Standard Mail Notification in Job
by Daniel Miliate 05 Dec '24

05 Dec '24

Newbie here, I'm trying to send myself an email notification with a custom body/subject for notification of important events. Is this possible? I've tried a few things in the script below. Thanks in advance and apologies if this is not how this listing works! - Daniel ---- Script ---- .... #SBATCH --mail-user= example(a)example.example #SBATCH --export=ALL #SBATCH --mail-type=ALL # Both of these work which mail which mailx EMAIL="example(a)example.example" # Attempt 1 echo "Test email body" | mail -s "Test Subject" "$EMAIL" # Attempt 2 mail -s "hello" "$EMAIL" <<EOF hello world EOF # Attempt 3 echo "Specific Information" | mail -s "SLURM Job information" $EMAIL ---- End ----

3 2

slurm not running on a warewulf node
by Steven Jones 04 Dec '24

04 Dec '24

Hi, I have set a log creation/location in slurm.conf as, SlurmdLogFile=/var/log/slurm/slurmd.log But it is 0 length. Slurm will not run, what else do I need to do to log why its failing pls? regards Steven

4 6

slurmd on a warwwulf node - not running
by Steven Jones 04 Dec '24

04 Dec '24

Hi, I have munge running on the controller and nodes fine, all tests passed. I have slurmctld running on the controller ok after checking the logs /var/spool/slurmctld was not created which I assume should have happened via the rpm install? Anyway I cant get slurmd to run on the warewulf nodes and cant find any log to check? How to fault find this? regards Steven

3 3

How can I make sure my user have only one job per node (Job array --exclusive=user,)
by Oren 03 Dec '24

03 Dec '24

Hi, I have a cluster of 20-nodes, and I want to run a jobarray on that cluster, but I want each node to get one job per node. When I do the following: #!/bin/bash #SBATCH --job-name=process_images_train # Job name #SBATCH --time=50:00:00 # Time limit hrs:min:sec #SBATCH --tasks=1 #SBATCH --cpus-per-task=4 #SBATCH --mem=50000 #SBATCH --array=0-19# 19 # Job array with 20 jobs (0 to 19) I get 10 jobs in node #1 and 10 jobs in node #20, I want a job in each node. I've tried: #SBATCH --exclusive=user Also #SBATCH --spread-job #SBATCH --distribution=cyclic Nothing changes, node#1 got 10 jobs and node#2 got 10 jobs. Thanks

2 6

multiple conf-server entries for sackd
by Brian Andrus 03 Dec '24

03 Dec '24

Not sure anyone would know, but... If you are running slurm in HA mode (multiple SlurmctldHost entries) is it possible to point sackd to more than one using the --conf-server option? So either specify --conf-server more than once, or have a comma-delimited list of them? The docs are a little light about that. Brian Andrus

1 0

sinfo not listing any partitions
by Kent L. Hanson 02 Dec '24

02 Dec '24

I am doing a new install of slurm 24.05.3 I have all the packages built and installed on headnode and compute node with the same munge.key, slurm.conf, and gres.conf file. I was able to run munge and unmunge commands to test munge successfully. Time is synced with chronyd. I can't seem to find any useful errors in the logs. For some reason when I run sinfo no nodes are listed. I just see the headers for each column. Has anyone seen this or know what a next step of troubleshooting would be? I'm new to this and not sure where to go from here. Thanks for any and all help! The odd output I am seeing [username@headnode ~] sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST (Nothing is output showing status of partition or nodes) Slurm.conf ClusterName=slurmkvasir SlurmctldHost=kadmin2 MpiDefault=none ProctrackType=proctrack/cgroup PrologFlags=contain ReturnToService=2 SlurmctldPidFile=/var/run/slurm/slurmctld.pid SlurmctldPort=6817 SlurmPidFile=/var/run/slurm/slurmd.pid SlurmdPort=6818 SlurmdSpoolDir=/var/spool/slurmd SlurmUser=slurm StateSaveLocation=/var/spool/slurmctld TaskPlugin=task/cgroup MinJobAge=600 SchedulerType=sched/backfill SelectType=select/cons_tres PriorityType=priority/multifactor AccountingStorageHost=localhost AccountingStoragePass=/var/run/munge/munge.socket.2 AccountingStorageType=accounting_storage/slurmdbd AccountingStorageTRES=gres/gpu,cpu,node JobCompType=jobcomp/none JobAcctGatherFrequency=30 JobAcctGatherType=jobacct_gather/cgroup SlurmctldDebug=info SlurmctldLogFile=/var/log/slurm/slurmctld.log SlurmdDebug=info SlurmLogFile=/var/log/slurm/slurmd.log nodeName=k[001-448] PartitionName=default Nodes=k[001-448] Default=YES MaxTime=INFINITE State=up Slurmctld.log Error: Configured MailProg is invalid Slurmctld version 24.05.3 started on cluster slurmkvasir Accounting_storage/slurmdbd: clusteracct_storage_p_register_ctld: Regisetering slurmctld at port 8617 Error: read_slurm_conf: default partition not set. Revovered state of 448 nodes Down nodes: k[002-448] Recovered information about 0 jobs Revovered state of 0 reservations Read_slurm_conf: backup_controller not specified Select/cons_tres; select_p_reconfigure: select/cons_tres: reconfigure Running as primary controller Slurmd.log Error: Node configuration differs from hardware: CPUS=1:40(hw) Boards=1:1(hw) SocketsPerBoard=1:2(hw) CoresPerSocket=1:20(hw) ThreadsPerCore:1:1(hw) CPU frequency setting not configured for this node Slurmd version 24.05.3started Slurmd started on Wed, 27 Nov 2024 06:51:03 -0700 CPUS=1 Boards=1 Cores=1 Threads=1 Memory=192030 TmpDisk=95201 uptime 166740 CPUSpecList=(null) FeaturesAvail=(null) FeaturesActive=(null) Error: _forward_thread: failed to k019 (10.142.0.119:6818): Connection timed out (Above line repeated 20 or so times for different nodes.) Thanks, Kent Hanson

6 9

Slurm version 24.11 is now available
by Tim Wickberg 29 Nov '24

29 Nov '24

We are pleased to announce the availability of the Slurm 24.11 release. To highlight some new features in 24.11: - New gpu/nvidia plugin. This does not rely on any NVIDIA libraries, and will build by default on all systems. It supports basic GPU detection and management, but cannot currently identify GPU-to-GPU links, or provide usage data as these are not exposed by the kernel driver. - Add autodetected GPUs to the output from "slurmd -C". - Added new QOS-based reports to "sreport". - Revamped network I/O with the "conmgr" thread-pool model. - Added new "hostlist function" syntax for management commands and configuration files. - switch/hpe_slingshot - Added support for hardware collectives setup through the fabric manager. (Requires SlurmctldParameters=enable_stepmgr) - Added SchedulerParameters=bf_allow_magnetic_slot configuration option to allow backfill planning for magnetic reservations. - Added new "scontrol listjobs" and "liststeps" commands to complement "listpids", and provide --json/--yaml output for all three subcommands. - Allow jobs to be submitted against multiple QOSes. - Added new experimental "oracle" backfill scheduling support, which permits jobs to be delayed if the oracle function determines the reduced fragmentation of the network topology is sufficiently advantageous. - Improved responsiveness of the controller when jobs are requeued by replacing the "db_index" identifier with a slurmctld-generated unique identifier. ("SLUID") - New options to job_container/tmpfs to permit site-specific scripts to modify the namespace before user steps are launched, and to ensure all steps are completely captured within that namespace. The Slurm documentation has also been updated to the 24.11 release. (Older versions can be found in the archive, linked from the main documentation page.) Slurm can be downloaded from https://www.schedmd.com/download-slurm/ . - Tim -- Tim Wickberg Chief Technology Officer, SchedMD LLC Commercial Slurm Development and Support

1 0

Non-intiutive rank placement/CPU masking
by Ohlerich, Martin 28 Nov '24

28 Nov '24

Dear * , I've some question for understanding. Essentially, I use the following job script: ---------------------------> #!/bin/bash #SBATCH -J srun_test #SBATCH --time=0:02:00 #SBATCH --export=NONE #SBATCH --partition=test #SBATCH --nodes=2 #SBATCH -o ./%x.%j.out #SBATCH -D . srun --export=all --mpi=pmi2 --verbose --cpu-bind=verbose -n 96 --ntasks-per-node=48 --ntasks-per-socket=24 -c 2 -m block:block -B 2:56:1 /lrz/sys/tools/placement_test_2021/bin/placement-test.intel_impi ---------------------------> On a sapphire rapid with 2 sockets and 56 physical CPUs. The executable is just some MPI dummy, build with Intel Compilers and Intel MPI (although I guess that's not essential here). The related MPI environment looks as show below (1). Slurm environment is shown below under (2). Now. what I would expect intuitively is that I place here 24 ranks per socket. And here also on the even CPUs in sequence starting from CPU 0 of each socket. What I get is srun: defined options srun: -------------------- -------------------- srun: (null) : i20r01c01s[06,08] srun: cpu-bind : verbose srun: cpus-per-task : 2 srun: distribution : block:block srun: export : all srun: extra-node-info : 2:56:1 srun: jobid : 395726 srun: job-name : MGLET_srun srun: mpi : pmi2 srun: nodes : 2 srun: ntasks : 96 srun: ntasks-per-node : 48 srun: ntasks-per-socket : 24 srun: verbose : 1 srun: -------------------- -------------------- srun: end of defined options srun: jobid 395726: nodes(2):`i20r01c01s[06,08]', cpu counts: 224(x2) srun: Implicitly setting --exact, because -c/--cpus-per-task given. srun: CpuBindType=verbose,threads srun: launching StepId=395726.0 on host i20r01c01s06, 48 tasks: [0-47] srun: launching StepId=395726.0 on host i20r01c01s08, 48 tasks: [48-95] srun: topology/tree: init: topology tree plugin loaded cpu-bind=MASK - i20r01c01s06, task 1 1 [98437]: mask 0xc set cpu-bind=MASK - i20r01c01s06, task 2 2 [98438]: mask 0x30 set cpu-bind=MASK - i20r01c01s06, task 3 3 [98439]: mask 0xc0 set cpu-bind=MASK - i20r01c01s06, task 4 4 [98440]: mask 0x300 set cpu-bind=MASK - i20r01c01s06, task 5 5 [98441]: mask 0xc00 set cpu-bind=MASK - i20r01c01s06, task 6 6 [98442]: mask 0x3000 set cpu-bind=MASK - i20r01c01s06, task 7 7 [98443]: mask 0xc000 set cpu-bind=MASK - i20r01c01s06, task 8 8 [98444]: mask 0x30000 set cpu-bind=MASK - i20r01c01s06, task 9 9 [98445]: mask 0xc0000 set cpu-bind=MASK - i20r01c01s06, task 10 10 [98446]: mask 0x300000 set cpu-bind=MASK - i20r01c01s06, task 11 11 [98447]: mask 0xc00000 set cpu-bind=MASK - i20r01c01s06, task 12 12 [98448]: mask 0x3000000 set cpu-bind=MASK - i20r01c01s06, task 13 13 [98449]: mask 0xc000000 set cpu-bind=MASK - i20r01c01s06, task 14 14 [98450]: mask 0x30000000 set cpu-bind=MASK - i20r01c01s06, task 15 15 [98451]: mask 0xc0000000 set cpu-bind=MASK - i20r01c01s06, task 16 16 [98452]: mask 0x300000000 set cpu-bind=MASK - i20r01c01s06, task 17 17 [98453]: mask 0xc00000000 set cpu-bind=MASK - i20r01c01s06, task 18 18 [98454]: mask 0x3000000000 set cpu-bind=MASK - i20r01c01s06, task 19 19 [98455]: mask 0xc000000000 set cpu-bind=MASK - i20r01c01s06, task 20 20 [98456]: mask 0x30000000000 set cpu-bind=MASK - i20r01c01s06, task 21 21 [98457]: mask 0xc0000000000 set cpu-bind=MASK - i20r01c01s06, task 22 22 [98458]: mask 0x300000000000 set cpu-bind=MASK - i20r01c01s06, task 23 23 [98459]: mask 0xc00000000000 set cpu-bind=MASK - i20r01c01s06, task 24 24 [98460]: mask 0x300000000000000 set cpu-bind=MASK - i20r01c01s06, task 25 25 [98461]: mask 0xc00000000000000 set cpu-bind=MASK - i20r01c01s06, task 26 26 [98462]: mask 0x3000000000000000 set cpu-bind=MASK - i20r01c01s06, task 27 27 [98463]: mask 0xc000000000000000 set cpu-bind=MASK - i20r01c01s06, task 28 28 [98464]: mask 0x30000000000000000 set cpu-bind=MASK - i20r01c01s06, task 29 29 [98465]: mask 0xc0000000000000000 set cpu-bind=MASK - i20r01c01s06, task 30 30 [98466]: mask 0x300000000000000000 set cpu-bind=MASK - i20r01c01s06, task 31 31 [98467]: mask 0xc00000000000000000 set cpu-bind=MASK - i20r01c01s06, task 32 32 [98468]: mask 0x3000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 33 33 [98469]: mask 0xc000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 34 34 [98470]: mask 0x30000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 35 35 [98471]: mask 0xc0000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 36 36 [98472]: mask 0x300000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 37 37 [98473]: mask 0xc00000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 38 38 [98474]: mask 0x3000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 39 39 [98475]: mask 0xc000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 40 40 [98476]: mask 0x30000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 41 41 [98477]: mask 0xc0000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 42 42 [98478]: mask 0x300000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 43 43 [98479]: mask 0xc00000000000000000000000 set cpu-bind=MASK - i20r01c01s06, task 44 44 [98480]: mask 0x3 set cpu-bind=MASK - i20r01c01s06, task 45 45 [98481]: mask 0xc set cpu-bind=MASK - i20r01c01s06, task 46 46 [98482]: mask 0x30 set srun: Node i20r01c01s06, 48 tasks started cpu-bind=MASK - i20r01c01s08, task 48 0 [179006]: mask 0x3 set cpu-bind=MASK - i20r01c01s08, task 49 1 [179007]: mask 0xc set cpu-bind=MASK - i20r01c01s08, task 50 2 [179008]: mask 0x30 set cpu-bind=MASK - i20r01c01s08, task 51 3 [179009]: mask 0xc0 set cpu-bind=MASK - i20r01c01s08, task 52 4 [179010]: mask 0x300 set cpu-bind=MASK - i20r01c01s08, task 53 5 [179011]: mask 0xc00 set cpu-bind=MASK - i20r01c01s08, task 54 6 [179012]: mask 0x3000 set cpu-bind=MASK - i20r01c01s08, task 55 7 [179013]: mask 0xc000 set cpu-bind=MASK - i20r01c01s08, task 56 8 [179014]: mask 0x30000 set cpu-bind=MASK - i20r01c01s08, task 57 9 [179015]: mask 0xc0000 set cpu-bind=MASK - i20r01c01s08, task 58 10 [179016]: mask 0x300000 set cpu-bind=MASK - i20r01c01s08, task 59 11 [179017]: mask 0xc00000 set cpu-bind=MASK - i20r01c01s08, task 60 12 [179018]: mask 0x3000000 set cpu-bind=MASK - i20r01c01s08, task 61 13 [179019]: mask 0xc000000 set cpu-bind=MASK - i20r01c01s08, task 62 14 [179020]: mask 0x30000000 set cpu-bind=MASK - i20r01c01s08, task 63 15 [179021]: mask 0xc0000000 set cpu-bind=MASK - i20r01c01s08, task 64 16 [179022]: mask 0x300000000 set cpu-bind=MASK - i20r01c01s08, task 65 17 [179023]: mask 0xc00000000 set cpu-bind=MASK - i20r01c01s08, task 66 18 [179024]: mask 0x3000000000 set cpu-bind=MASK - i20r01c01s08, task 67 19 [179025]: mask 0xc000000000 set cpu-bind=MASK - i20r01c01s08, task 68 20 [179026]: mask 0x30000000000 set cpu-bind=MASK - i20r01c01s08, task 69 21 [179027]: mask 0xc0000000000 set cpu-bind=MASK - i20r01c01s08, task 70 22 [179028]: mask 0x300000000000 set cpu-bind=MASK - i20r01c01s08, task 71 23 [179029]: mask 0xc00000000000 set cpu-bind=MASK - i20r01c01s08, task 72 24 [179030]: mask 0x300000000000000 set cpu-bind=MASK - i20r01c01s08, task 73 25 [179031]: mask 0xc00000000000000 set cpu-bind=MASK - i20r01c01s08, task 74 26 [179032]: mask 0x3000000000000000 set cpu-bind=MASK - i20r01c01s08, task 75 27 [179033]: mask 0xc000000000000000 set cpu-bind=MASK - i20r01c01s08, task 76 28 [179034]: mask 0x30000000000000000 set cpu-bind=MASK - i20r01c01s08, task 77 29 [179035]: mask 0xc0000000000000000 set cpu-bind=MASK - i20r01c01s08, task 78 30 [179036]: mask 0x300000000000000000 set cpu-bind=MASK - i20r01c01s08, task 79 31 [179037]: mask 0xc00000000000000000 set cpu-bind=MASK - i20r01c01s08, task 80 32 [179038]: mask 0x3000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 81 33 [179039]: mask 0xc000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 82 34 [179040]: mask 0x30000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 83 35 [179041]: mask 0xc0000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 84 36 [179042]: mask 0x300000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 85 37 [179043]: mask 0xc00000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 86 38 [179044]: mask 0x3000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 87 39 [179045]: mask 0xc000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 88 40 [179046]: mask 0x30000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 89 41 [179047]: mask 0xc0000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 90 42 [179048]: mask 0x300000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 91 43 [179049]: mask 0xc00000000000000000000000 set cpu-bind=MASK - i20r01c01s08, task 92 44 [179050]: mask 0x3 set cpu-bind=MASK - i20r01c01s08, task 93 45 [179051]: mask 0xc set cpu-bind=MASK - i20r01c01s08, task 94 46 [179052]: mask 0x30 set cpu-bind=MASK - i20r01c01s08, task 95 47 [179053]: mask 0xc0 set srun: Node i20r01c01s08, 48 tasks started cpu-bind=MASK - i20r01c01s06, task 47 47 [98483]: mask 0xc0 set In words: the last 4 ranks meant for the second socket of each node are actually placed on the first socket ... So, effectively --ntasks-per-socket is then ignored? I couldn't see an obvious reason for that. Maybe I miss some important point ... or neglect some interference with Intel MPI's runtime environment (there didn't change anything in the masks when unsetting KMP_AFFINITY). I'd like also to mention that explicit placement via cpu_map does the right thing. ---------------------------> #!/bin/bash #SBATCH -J srun_test #SBATCH --time=0:02:00 #SBATCH --export=NONE #SBATCH --partition=test #SBATCH --nodes=2 #SBATCH --ntasks-per-node=48 #SBATCH -o ./%x.%j.out #SBATCH -D . srun --export=all --mpi=pmi2 --cpu-bind=map_cpu:$(seq 0 2 46 | tr '\n' ',')$(seq 56 2 102 | tr '\n' ',') /lrz/sys/tools/placement_test_2021/bin/placement-test.intel_impi ---------------------------> So, a workaround is available for me. But if I could get some illuminating hint on where my intuition failed above, I'd be very grateful. Thank you! Cheers, Martin (1) I_MPI_FILESYSTEM=on I_MPI_HYDRA_BOOTSTRAP=slurm I_MPI_HYDRA_BOOTSTRAP_EXEC_EXTRA_ARGS=--external-launcher I_MPI_HYDRA_BRANCH_COUNT=128 I_MPI_OFFLOAD=0 I_MPI_OFFLOAD_FAST_MEMCPY_COLL=0 I_MPI_OFFLOAD_RDMA=0 I_MPI_OFI_PROVIDER=psm3 I_MPI_ROOT=/dss/lrzsys/sys/spack/release/24.1.0/opt/x86_64/intel-oneapi-mpi/2021.11.0-gcc-w56vuor/mpi/2021.11 KMP_AFFINITY=granularity=thread,compact,1,0 (2) SLURMD_NODENAME=i20r01c01s06 SLURM_CLUSTER_NAME=sng2 SLURM_CONF=/etc/slurm/slurm.conf SLURM_CPUS_ON_NODE=224 SLURM_ECLIBR=0 SLURM_ECPLUG=1 SLURM_ERLAST=sbatch SLURM_ERSBAC=1 SLURM_GET_USER_ENV=0 SLURM_GTIDS=0 SLURM_JOBID=395726 SLURM_JOB_ACCOUNT=pr28fa SLURM_JOB_CPUS_PER_NODE=224(x2) SLURM_JOB_END_TIME=1732781080 SLURM_JOB_GID=3000114 SLURM_JOB_ID=395726 SLURM_JOB_NAME=MGLET_srun SLURM_JOB_NODELIST=i20r01c01s[06,08] SLURM_JOB_NUM_NODES=2 SLURM_JOB_PARTITION=test SLURM_JOB_QOS=test SLURM_JOB_START_TIME=1732780959 SLURM_JOB_UID=3808660 SLURM_JOB_USER=di49zop SLURM_LOCALID=0 SLURM_NNODES=2 SLURM_NODEID=0 SLURM_NODELIST=i20r01c01s[06,08] SLURM_PRIO_PROCESS=0 SLURM_PROCID=0 SLURM_SCRIPT_CONTEXT=prolog_task SLURM_SETUP_LICENSE=open source - no access restrictions SLURM_SETUP_MAINTAINER_LIST=Bader(a)lrz.de SLURM_SUBMIT_DIR=/dss/dsshome1/00/di49zop/test_mglet SLURM_SUBMIT_HOST=login26 SLURM_TASKS_PER_NODE=224(x2) SLURM_TASK_PID=98399 SLURM_TOPOLOGY_ADDR=leaf.m02r05.i20r01c01s06 SLURM_TOPOLOGY_ADDR_PATTERN=switch.switch.node

1 0

I'm very confused about priority/fairshare calculation
by Cutts, Tim 27 Nov '24

27 Nov '24

I don't know how many times I've read the docs; I keep thinking I understand it, but something is really wrong with prioritisation on our cluster, and we're struggling to understand why. The setup: 1. We have a group who submit two types of work; production jobs and research jobs. 2. We have two sacctmgr accounts for this; let's call those 'prod' and 'research'. 3. We also have some dedicated hardware that they paid for which can be used only by users associated with the prod account. Desired behaviour: 1. Usage of their dedicated hardware by production jobs should not hugely decrease the fairshare priority for research jobs in other partitions. 2. Usage of shared hardware should decrease their fairshare priority (whether by production or research jobs) 3. Memory should make a relatively small contribution to TRES usage (it's not normally the constrained resource) Our approach: Set TRESBillingWeights for cpu, memory and gres/GPU usage on shared partitions. Typically these are set to: CPU=1.0,Mem=0.25G,GRES/gpu=1.0 Set TRESBillingWeights to something small on the dedicated hardware partition, such as: CPU=0.25 Set PriorityWeightFairshare and PriorityWeightAge to values such that Fairshare dominates when jobs are young, and Age takes over if they've been pending a long time The observed behaviour: 1. production association jobs have a high priority; this is working well 2. research jobs are still getting heavily penalised in fairshare, and we don't understand why; they seem to have enormous RawUsage, largely coming from memory: Here's what I see from sshare (sensitive details removed, obviously): sshare -l -A prod, research -a -o Account,RawUsage,EffectvUsage,FairShare,LevelFS,TRESRunMins%80 | grep -v cpu=0 > ' Account RawUsage EffectvUsage FairShare LevelFS TRESRunMins -------------------- ----------- ------------- ---------- ---------- -------------------------------------------------------------------------------- prod 1587283 0.884373 0.226149 cpu=81371,mem=669457237,energy=0,node=20610,billing=100833,fs/disk=0,vmem=0,pag+ prod 1082008 0.681681 0.963786 0.366740 cpu=81281,mem=669273429,energy=0,node=20520,billing=100833,fs/disk=0,vmem=0,pag+ prod 505090 0.318202 0.964027 0.785664 cpu=90,mem=184320,energy=0,node=90,billing=0,fs/disk=0,vmem=0,pages=0,gres/gpu=+ research 1043560787 0.380577 0.121648 cpu=17181098808,mem=35196566339054,energy=0,node=4295361360,billing=25773481938+ research 146841 0.000141 0.005311 124.679238 cpu=824,mem=3375923,energy=0,node=824,billing=824,fs/disk=0,vmem=0,pages=0,gres+ research 17530141 0.016798 0.001449 1.044377 cpu=254484,mem=3379938816,energy=0,node=161907,billing=893592,fs/disk=0,vmem=0,+ research 167597 0.000161 0.005070 109.238498 cpu=7275,mem=223516160,energy=0,node=7275,billing=50931,fs/disk=0,vmem=0,pages=+ research 12712481 0.012182 0.001931 1.440166 cpu=186327,mem=95399526,energy=0,node=23290,billing=232909,fs/disk=0,vmem=0,pag+ research 11521011 0.011040 0.002173 1.589104 cpu=8167,mem=267626086,energy=0,node=8167,billing=65338,fs/disk=0,vmem=0,pages=+ research 9719735 0.009314 0.002414 1.883599 cpu=15020,mem=69214617,energy=0,node=1877,billing=3755,fs/disk=0,vmem=0,pages=0+ research 25004766 0.023961 0.001207 0.732184 cpu=590778,mem=6464600473,energy=0,node=98910,billing=2266887,fs/disk=0,vmem=0,+ research 68938740 0.066061 0.000724 0.265570 cpu=159332,mem=963064985,energy=0,node=89957,billing=192706,fs/disk=0,vmem=0,pa+ research 7359413 0.007052 0.002656 2.487710 cpu=81401,mem=583487624,energy=0,node=20350,billing=20350,fs/disk=0,vmem=0,page+ research 718714430 0.688714 0.000241 0.025473 cpu=20616,mem=337774728,energy=0,node=5154,billing=92772,fs/disk=0,vmem=0,pages+ research 1016606 0.000974 0.003863 18.009010 cpu=17179774580,mem=35184178340113,energy=0,node=4294943645,billing=25769661870+ Firstly, why are the mem TRES numbers so enormous? Secondly, what's going on with the last user, where the rawusage is tiny, but the TRESRunMins is ridiculously big? That could be messing up the whole thing. Thanks in advance for any advice (either that can help explain what I've misunderstood, or to suggestions of "there's a better way to achieve what you want") Tim -- Tim Cutts Scientific Computing Platform Lead AstraZeneca ________________________________ AstraZeneca UK Limited is a company incorporated in England and Wales with registered number:03674842 and its registered office at 1 Francis Crick Avenue, Cambridge Biomedical Campus, Cambridge, CB2 0AA. This e-mail and its attachments are intended for the above named recipient only and may contain confidential and privileged information. If they have come to you in error, you must not copy or show them to anyone; instead, please reply to this e-mail, highlighting the error to the sender and then immediately delete the message. For information about how AstraZeneca UK Limited and its affiliates may process information, personal data and monitor communications, please see our privacy notice at www.astrazeneca.com<https://www.astrazeneca.com>

1 0

making a maint reservation on a specific GPU
by Paul Raines 22 Nov '24

22 Nov '24

We have a 8 GPU server in which one GPU has gone into an error state that will require a reboot to clear. I have jobs on the server running on good GPUs that will take another 3 days to complete. In the meantime, I would like short jobs to run on the good free GPUs till I reboot. I set a reservation for the time window I plan to reboot on the whole node with scontrol create reservation reservationName=rtx-01_reboot users=root starttime=2024-11-25T06:00:00 duration=720 Nodes=rtx-01 flags=maint,ignore_jobs But I would like to set a reservation on just the bad GPU (gpu_id=7) from now till 2024-11-25T06:00:00 so no job runs that will use it. Is that possible? --------------------------------------------------------------- Paul Raines http://help.nmr.mgh.harvard.edu MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging 149 (2301) 13th Street Charlestown, MA 02129 USA The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Mass General Brigham Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline <https://www.massgeneralbrigham.org/complianceline> . Please note that this e-mail is not secure (encrypted). If you do not wish to continue communication over unencrypted e-mail, please notify the sender of this message immediately. Continuing to send or respond to e-mail after receiving this message means you understand and accept this risk and wish to continue to communicate over unencrypted e-mail.

1 0

Suspending jobs and resuming
by Ratnasamy, Fritz 22 Nov '24

22 Nov '24

Hi, I am using an old slurm version 20.11.8 and we had to reboot our cluster today for maintenance. I suspended all the jobs on it with the command scontrol suspend list_job_ids and all the jobs paused and were suspended. However, when I tried to resume them after the reboot, scontrol resume did not work (it was showing in the reason column " (JobHeldAdmin)". I was able to release them with scontrol release and the jobs started to run back. However, the SLURM recorded time on it resetted (Time columns, showing 0:00 for all the jobs) though the jobs seem to have re-started from the last point before he got suspended. 1- Did I follow the right procedure to suspend, reboot and resume/release? 2- In this case, does the wall time for all the jobs goes into reset and therefore anyone with slurm admin rights will be able to have their jobs last longer than the wall time limit by suspending and resuming a job? Best, *Fritz Ratnasamy* Data Scientist Information Technology

2 1

Slurm PID Files
by Matthias Leopold 20 Nov '24

20 Nov '24

Hi, I compiled and installed Slurm 24.05 on Ubuntu 22.04 following this tutorial: https://www.schedmd.com/slurm/installation-tutorial/ Systemd service files are from deb packages that result from this. Do I have to worry that slurmctld and slurmd don't write PID files although SlurmctldPidFile and SlurmdPidFile are defined in slurm.conf? Paths for PID files exist and are writeable, logs don't show any error. slurmdbd does write a PID file as defined in slurmdbd.conf. thx Matthias

1 0

Does Slurm support DSP
by shaobo liu 20 Nov '24

20 Nov '24

Dear all Does slurm support DSP (Digital Signal Processing)? slurm website does not see DSP related content。

3 4

Does Slurm support DSP
by shaobo liu 20 Nov '24

20 Nov '24

Dear all Does slurm support DSP (Digital Signal Processing)? slurm website does not see DSP related content。

1 0

error: Unable to contact slurm controller (connect failure)
by Daniel Rodriguez Lopez (ext) 19 Nov '24

19 Nov '24

Dear all, We recently tried to fix our version of slurm in every node of our cluster. After the instalation (slurm 20.11.9) in one of the compute nodes, most of the commads (squeue, sinfo, scontrol show config etc) returns this error: error: Unable to contact slurm controller (connect failure) The .log files don't show any errors, we have both debugs values equal to debug5. Also, the rest of the cluster works as usual. I appreciate any insight on what could be the cause. Thank you and regards, Daniel

4 3

Slurm services status=1/FAILURE for CentOs cluster
by Mojtaba Farrokhbin 19 Nov '24

19 Nov '24

<https://superuser.com/posts/1862116/timeline> I have reinstalled slurm resource management on a HPC cluster. But it seems there is a problem on starting slurmd services. Here is the system status: "systemctl status slurmd" shows: ● slurmd.service - Slurm node daemon Loaded: loaded (/usr/lib/systemd/system/slurmd.service; enabled; vendor preset: disabled) Active: failed (Result: exit-code) since Sun 2024-11-10 11:31:24 +0130; 1 weeks 1 days ago ------------------------------------------- The slurm.conf output is: # # Example slurm.conf file. Please run configurator.html # (in doc/html) to build a configuration file customized # for your environment. # # # slurm.conf file generated by configurator.html. # # See the slurm.conf man page for more information. # ClusterName=linux ControlMachine=master.cluster.... #ControlAddr= #BackupController= #BackupAddr= # SlurmUser=slurm #SlurmdUser=root SlurmctldPort=6817 SlurmdPort=6818 AuthType=auth/munge #JobCredentialPrivateKey= #JobCredentialPublicCertificate= StateSaveLocation=/var/spool/slurm/ctld SlurmdSpoolDir=/var/spool/slurm/d SwitchType=switch/none MpiDefault=none SlurmctldPidFile=/var/run/slurmctld.pid SlurmdPidFile=/var/run/slurmd.pid #ProctrackType=proctrack/pgid ProctrackType=proctrack/cgroup TaskPlugin=task/cgroup TaskPluginParam=cpusets #PluginDir= #FirstJobId= #MaxJobCount= #PlugStackConfig= #PropagatePrioProcess= #PropagateResourceLimits= #PropagateResourceLimitsExcept= #Prolog= #Epilog= #SrunProlog= #SrunEpilog= #TaskProlog= #TaskEpilog= #TaskPlugin= #TrackWCKey=no #TreeWidth=50 #TmpFS= #UsePAM= # # GPU definition (Added by S 2022/11) GresTypes=gpu # TIMERS SlurmctldTimeout=300 SlurmdTimeout=300 InactiveLimit=0 MinJobAge=300 KillWait=30 Waittime=0 # # SCHEDULING SchedulerType=sched/backfill #SchedulerAuth= #SelectType=select/linear SelectType=select/cons_res # select partial node #SelectTypeParameters=CR_CPU_Memory SelectTypeParameters=CR_Core_Memory #FastSchedule=1 FastSchedule=0 PriorityType=priority/multifactor PriorityDecayHalfLife=0 PriorityUsageResetPeriod=NONE PriorityWeightFairshare=100000 #PriorityWeightAge=1000 PriorityWeightPartition=10000 #PriorityType=priority/multifactor #PriorityDecayHalfLife=14-0 #PriorityUsageResetPeriod=14-0 #PriorityWeightFairshare=100000 #PriorityWeightAge=1000 #PriorityWeightPartition=10000 #PriorityWeightJobSize=1000 #PriorityMaxAge=1-0 DefMemPerCPU=2000 # LOGGING SlurmctldDebug=3 SlurmctldLogFile=/var/log/slurmctld.log SlurmdDebug=3 SlurmdLogFile=/var/log/slurmd.log JobCompType=jobcomp/none #JobCompLoc= # # ACCOUNTING #JobAcctGatherType=jobacct_gather/linux #JobAcctGatherFrequency=30 # AccountingStorageType=accounting_storage/slurmdbd AccountingStorageHost=master #AccountingStorageLoc= #AccountingStoragePass= #AccountingStorageUser= # # COMPUTE NODES # OpenHPC default configuration # Edited by Surin 1399/11 #AccountingStorageEnforce=limits AccountingStorageEnforce=QOS,Limits,Associations #TaskPlugin=task/affinity #PropagateResourceLimitsExcept=MEMLOCK #AccountingStorageType=accounting_storage/filetxt Epilog=/etc/slurm/slurm.epilog.clean NodeName=cn0[1-5] NodeHostName=cn0[1-5] RealMemory=128307 Sockets=2 CoresPerSocket=16 ThreadsPerCore=2 Feature=HyperThread State=UNKNOWN NodeName=gp01 NodeHostName=gp01 RealMemory=128307 Sockets=2 CoresPerSocket=8 ThreadsPerCore=2 Feature=HyperThread Gres=gpu:4 State=UNKNOWN #PartitionName=all Nodes=cn0[1-5],gp01 MaxTime=INFINITE State=UP Oversubscribe=EXCLUSIVE PartitionName=all Nodes=cn0[1-5],gp01 MaxTime=10-00:00:00 State=UP MaxNodes=1 PartitionName=normal Nodes=cn0[1-5] Default=YES MaxTime=10-00:00:00 State=UP MaxNodes=1 #PartitionName=normal Nodes=cn0[1-5] Default=YES MaxTime=INFINITE State=UP Oversubscribe=EXCLUSIVE PartitionName=gpu Nodes=gp01 MaxTime=10-00:00:00 State=UP SlurmctldParameters=enable_configless ReturnToService=1 HealthCheckProgram=/usr/sbin/nhc HealthCheckInterval=300 ------------------------------------------- scontrol show node cn01 shows: NodeName=cn01 CoresPerSocket=16 CPUAlloc=0 CPUTot=64 CPULoad=N/A AvailableFeatures=HyperThread ActiveFeatures=HyperThread Gres=(null) NodeAddr=cn01 NodeHostName=cn01 RealMemory=128557 AllocMem=0 FreeMem=N/A Sockets=2 Boards=1 State=DOWN* ThreadsPerCore=2 TmpDisk=64278 Weight=1 Owner=N/A MCS_label=N/A Partitions=all,normal BootTime=None SlurmdStartTime=None CfgTRES=cpu=64,mem=128557M,billing=64 AllocTRES= CapWatts=n/a CurrentWatts=0 LowestJoules=0 ConsumedJoules=0 ExtSensorsJoules=n/s ExtSensorsWatts=0 ExtSensorsTemp=n/s Reason=undraining ------------------------------------------- "scontrol ping" also works as: Slurmctld(primary) at master.cluster... is UP ------------------------------------------- "systemctl start slurmd" shows: Job for slurmd.service failed because the control process exited with error code. See "systemctl status slurmd.service" and "journalctl -xe" for details. ------------------------------------------- "systemctl status slurmd.service" shows: ● slurmd.service - Slurm node daemon Loaded: loaded (/usr/lib/systemd/system/slurmd.service; enabled; vendor preset: disabled) Active: failed (Result: exit-code) since Mon 2024-11-18 17:02:50 +0130; 41s ago Process: 219025 ExecStart=/usr/sbin/slurmd $SLURMD_OPTIONS (code=exited, status=1/FAILURE) Nov 18 17:02:50 master.cluster.... systemd[1]: Starting Slurm node daemon... Nov 18 17:02:50 master.cluster.... slurmd[219025]: fatal: Unable to determine this slurmd's NodeName Nov 18 17:02:50 master.cluster.... systemd[1]: slurmd.service: control process exited, code=exited status=1 Nov 18 17:02:50 master.cluster.... systemd[1]: Failed to start Slurm node daemon. Nov 18 17:02:50 master.cluster.... systemd[1]: Unit slurmd.service entered failed state. Nov 18 17:02:50 master.cluster.... systemd[1]: slurmd.service failed. ------------------------------------------- "journalctl -xe" output is: Nov 18 17:04:54 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:05:04 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:05:08 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:05:12 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:05:22 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:05:26 master.cluster.... dhcpd[2250]: DHCPDISCOVER from 70:35:09:f8:13:40 via eno2: network eno2: no free leases Nov 18 17:06:34 master.cluster.... munged[2514]: Purged 2 credentials from replay hash ------------------------------------------- slurmd.services contains the following information: [Unit] Description=Slurm node daemon After=munge.service network.target remote-fs.target ConditionPathExists=/etc/slurm/slurm.conf [Service] Type=forking EnvironmentFile=-/etc/sysconfig/slurmd ExecStart=/usr/sbin/slurmd $SLURMD_OPTIONS ExecReload=/bin/kill -HUP $MAINPID PIDFile=/var/run/slurmd.pid KillMode=process LimitNOFILE=131072 LimitMEMLOCK=infinity LimitSTACK=infinity Delegate=yes [Install] WantedBy=multi-user.target [root@master system]# which slurmd /usr/sbin/slurmd [root@master system]# cat slurmd.service [Unit] Description=Slurm node daemon After=munge.service network.target remote-fs.target ConditionPathExists=/etc/slurm/slurm.conf [Service] Type=forking EnvironmentFile=-/etc/sysconfig/slurmd ExecStart=/usr/sbin/slurmd $SLURMD_OPTIONS ExecReload=/bin/kill -HUP $MAINPID PIDFile=/var/run/slurmd.pid KillMode=process LimitNOFILE=131072 LimitMEMLOCK=infinity LimitSTACK=infinity Delegate=yes [Install] WantedBy=multi-user.target ------------------------------------------- For "which slurmd" command I get this address: /usr/sbin/slurm ------------------------------------------- ls -l /usr/sbin/slurmd shows: lrwxrwxrwx 1 root root 69 Nov 10 10:54 /usr/sbin/slurmd -> /install/centos7.9/compute_gpu/rootimg/usr/sbin/slurmd

1 0

InvalidAccount
by hmeij＠wesleyan.edu 18 Nov '24

18 Nov '24

Manual compilation of 24.05.4. slurmctld and slurmd run on same server. All works ok but all test jobs end up pending with InvalidAccount message. I do not use slurm database and have not enabled accounting. Can not find an answer for this behavior or a misconfiguration. slurm.conf file was generated using easy config tool. Any ideas how to fix this? Thx, -Henk ## looks like all users have access to test queue [hmeij@sharptail2 slurm]$ sinfo -o "%g %.10R %.20l" GROUPS PARTITION TIMELIMIT all test infinite [hmeij@sharptail2 slurm]$ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST test* up infinite 1 idle sharptail2 ## simple sleep job [hmeij@sharptail2 slurm]$ sbatch sleep Submitted batch job 8 [hmeij@sharptail2 slurm]$ squeue JOBID PARTITION NAME USER ST TIME NODES CPUS MIN_MEMORY NODELIST(REASON) 8 test sleep hmeij PD 0:00 1 1 1G (InvalidAccount) [hmeij@sharptail2 slurm]$ scontrol show job 8 JobId=8 JobName=sleep UserId=hmeij(8216) GroupId=its(623) MCS_label=N/A Priority=1 Nice=0 Account=(null) QOS=(null) JobState=PENDING Reason=InvalidAccount Dependency=(null) Requeue=1 Restarts=0 BatchFlag=1 Reboot=0 ExitCode=0:0 RunTime=00:00:00 TimeLimit=UNLIMITED TimeMin=N/A SubmitTime=2024-11-11T13:27:14 EligibleTime=2024-11-11T13:27:14 AccrueTime=2024-11-11T13:27:14 StartTime=Unknown EndTime=Unknown Deadline=N/A SuspendTime=None SecsPreSuspend=0 LastSchedEval=2024-11-11T13:27:14 Scheduler=Main Partition=test AllocNode:Sid=sharptail2:644662 ReqNodeList=(null) ExcNodeList=(null) NodeList= NumNodes=1-1 NumCPUs=1 NumTasks=1 CPUs/Task=1 ReqB:S:C:T=0:0:1:1 ReqTRES=cpu=1,mem=1G,node=1,billing=1 AllocTRES=(null) Socks/Node=1 NtasksPerN:B:S:C=1:0:*:* CoreSpec=* MinCPUsNode=1 MinMemoryNode=1G MinTmpDiskNode=0 Features=(null) DelayBoot=00:00:00 OverSubscribe=OK Contiguous=0 Licenses=(null) Network=(null) Command=/zfshomes/hmeij/slurm/sleep WorkDir=/zfshomes/hmeij/slurm StdErr=/zfshomes/hmeij/slurm/err StdIn=/dev/null StdOut=/zfshomes/hmeij/slurm/out TresPerTask=cpu=1 ## within a minute or so that InvalidAccount changes to None ( ## but job remains pending; 1-7 stuck over the weekend) [hmeij@sharptail2 slurm]$ squeue JOBID PARTITION NAME USER ST TIME NODES CPUS MIN_MEMORY NODELIST(REASON) 8 test sleep hmeij PD 0:00 1 1 1G (None) ## in the slurmctld.log slurmctld: sched: JobId=8 has invalid account slurmctld: debug: set_job_failed_assoc_qos_ptr: Filling in assoc for JobId=8 Assoc=0 slurmctld: debug: sched: Running job scheduler for full queue. slurmctld: error: _refresh_assoc_mgr_qos_list: no new list given back keeping cached one. ##and the slurm.conf accounting section (both AccountingStorageType lines yield same behavior) #AccountingStorageType= AccountingStorageType=accounting_storage/none #JobAcctGatherFrequency=30 #JobAcctGatherType= ## using SchedulerType = sched/builtin

3 5

Listen to job state changes
by egonle＠posteo.me 18 Nov '24

18 Nov '24

Hello, is there any way to listen to job state changes of slurm 23.x or newer? I’d like to kind of subscribe to job state changes instead of polling for job states. Adding this feature to slurm accounting DB seems to be last option right now, although I’d like to avoid it. Thanks&Best Egonle

4 3

How to power up all ~idle nodes and verify that they have started up without issue programmatically
by Xaver Stiensmeier 15 Nov '24

15 Nov '24

Dear Slurm User list, I would like to startup all ~idle (idle and powered down) nodes and check programmatically if all came up as expected. For context: this is for a program that sets up slurm clusters with on demand cloud scheduling. In the most easiest fashion this could be executing a command like *srun FORALL hostname* which would return the names of the nodes if it succeeds and an error message otherwise. However, there's no such input value like FORALL as far as I am aware. One could use -N{total node number} as all nodes are ~idle when this executes, but I don't know an easy way to get the total number of nodes. Best regards, Xaver

3 5

First setup of slurm with a GPU node
by Patrick Begou 13 Nov '24

13 Nov '24

Hi, I'm using slurm on a small 8 nodes cluster. I've recently added one GPU node with two Nvidia A100, one with 40Gb of RAM and one with 80Gb. As using this GPU resource increase I would like to manage this resource with Gres to avoid usage conflict. But at this time my setup do not works as I can reach a GPU without reserving it: srun -n 1 -p tenibre-gpu ./a.out can use a GPU even if the reservation do not specify this resource (checked with running nvidia-smi on the node). "tenibre-gpu" is a slurm partition with only this gpu node. From the documentation I've created a gres.conf file and it is propagated on all the nodes (9 compute nodes, 1 login node and the management node) and slurmd has been restarted. gres.conf is:* ## GPU setup on tenibre-gpu-0 NodeName=tenibre-gpu-0 Name=gpu Type=A100-40 File=/dev/nvidia0 Flags=nvidia_gpu_env NodeName=tenibre-gpu-0 Name=gpu Type=A100-80 File=/dev/nvidia1 Flags=nvidia_gpu_env * * In slurm.conf I have checked these flags: ## Basic scheduling SelectTypeParameters=CR_Core_Memory SchedulerType=sched/backfill SelectType=select/cons_tres ## Generic resources GresTypes=gpu ## Nodes list .... Nodename=tenibre-gpu-0 RealMemory=257270 Sockets=2 CoresPerSocket=16 ThreadsPerCore=1 State=UNKNOWN .... #partitions PartitionName=tenibre-gpu MaxTime=48:00:00 DefaultTime=12:00:00 DefMemPerCPU=4096 MaxMemPerCPU=8192 Shared=YES State=UP Nodes=tenibre-gpu-0 ... May be I've missed something ? I'm running Slurm 20.11.7-1. Thanks for your advices. Patrick

5 6

Slurm and NVIDIA NVML
by Matthias Leopold 13 Nov '24

13 Nov '24

Hi, I'm trying to compile Slurm with NVIDIA NVML support, but the result is unexpected. I get /usr/lib/x86_64-linux-gnu/slurm/gpu_nvml.so, but when I do "ldd /usr/lib/x86_64-linux-gnu/slurm/gpu_nvml.so" there is no reference to /lib/x86_64-linux-gnu/libnvidia-ml.so.1 (which I would expect). ~$ ldd /usr/lib/x86_64-linux-gnu/slurm/gpu_nvml.so linux-vdso.so.1 (0x00007ffd9a3f4000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f0bc2c06000) /lib64/ld-linux-x86-64.so.2 (0x00007f0bc2e47000) /lib/x86_64-linux-gnu/libnvidia-ml.so.1 is present during compilation. Also I can see that NVML headers where found in config.status (else I wouldn't get gpu_nvml.so at all to my understanding). Our old cluster was deployed with NVIDIA deepops (which compiles Slurm on every node) and also has NVML support. There ldd brings the expected result ~$ ldd /usr/local/lib/slurm/gpu_nvml.so ... libnvidia-ml.so.1 => /lib/x86_64-linux-gnu/libnvidia-ml.so.1 (0x00007f3b10120000) ... I can't test actual functionality with my new binaries because I don't have a node with GPUs yet. Am I missing something? thank you Matthias

2 2

Slurm release candidate version 24.11.0rc1 available for testing
by Marshall Garey 12 Nov '24

12 Nov '24

We are pleased to announce the availability of Slurm release candidate 24.11.0rc1. To highlight some new features coming in 24.11: - New gpu/nvidia plugin. This does not rely on any NVIDIA libraries, and will build by default on all systems. It supports basic GPU detection and management, but cannot currently identify GPU-to-GPU links, or provide usage data as these are not exposed by the kernel driver. - Add autodetected GPUs to the output from "slurmd -C". - Added new QOS-based reports to "sreport". - Revamped network I/O with the "conmgr" thread-pool model. - Added new "hostlist function" syntax for management commands and configuration files. - switch/hpe_slingshot - Added support for hardware collectives setup through the fabric manager. (Requires SlurmctldParameters=enable_stepmgr) - Added SchedulerParameters=bf_allow_magnetic_slot configuration option to allow backfill planning for magnetic reservations. - Added new "scontrol listjobs" and "liststeps" commands to complement "listpids", and provide --json/--yaml output for all three subcommands. - Allow jobs to be submitted against multiple QOSes. - Added new experimental "oracle" backfill scheduling support, which permits jobs to be delayed if the oracle function determines the reduced fragmentation of the network topology is sufficiently advantageous. - Improved responsiveness of the controller when jobs are requeued by replacing the "db_index" identifier with a slurmctld-generated unique identifier. ("SLUID") - New options to job_container/tmpfs to permit site-specific scripts to modify the namespace before user steps are launched, and to ensure all steps are completely captured within that namespace. This is the first release candidate of the upcoming 24.11 release series, and represents the end of development for this release, and a finalization of the RPC and state file formats. If any issues are identified with this release candidate, please report them through https://bugs.schedmd.com against the 24.11.x version and we will address them before the first production 24.11.0 release is made. Please note that the release candidates are not intended for production use. A preview of the updated documentation can be found at https://slurm.schedmd.com/archive/slurm-master/ . Slurm can be downloaded from https://www.schedmd.com/downloads.php . -- Marshall Garey Release Management, Support, and Development SchedMD LLC - Commercial Slurm Development and Support

1 0

Container Problems with various oci.conf settings
by Jason Pascucci 12 Nov '24

12 Nov '24

Hi, I'm running on Ubuntu 20.04. I've got a clean configuration of slurmctld and slurmd on one node. 1) I've configured oci.conf to the defaults defined by "OCI.CONF EXAMPLE FOR RUNC USING RUN (RECOMMENDED OVER USING CREATE/START):". I have a container that I can run by hand: runc --rootless true run -b /opt/pilot_results/results.20241108-184831/step1 test sh-4.2# exit exit and it returns. However, when I srun --container=/opt/pilot_results/results.20241108-184831/step1 ls it hangs after completing the ls, and I have to double ctrl-c out of it. 2) I tried using the configuration for RUNC with Create/Start and it hangs on start. 3) I tried using the configuration for CRUN using RUN, I can run the container by hand with crun, but srun fails with: srun --container=/opt/pilot_results/results.20241108-194914/step1 bash bind socket to `/run/user/1008//pd-builds-bench-1.jrp.34.0.0/notify`: Address already in use sync socket closed srun: error: pd-builds-bench-1: task 0: Exited with exit code 1 4) I tried using the configuration for CRUN with Create/Start and it errors repeatedly with: slurmstepd: error: _get_container_state: RunTimeQuery failed rc:256 output:error opening file `/run/user/1008//pd-builds-bench-1.jrp.51.0.0/status`: No such file or directory I went through the (open and closed) support tickets and couldn't find anything that reflects any of these errors, and I'm pretty stuck at this point. Any help would be welcome. Thanks, JRP

1 0

slurmctld keeps segfaulting, possibly during or just after backfill
by Marcus Lauer 12 Nov '24

12 Nov '24

We are running into a problem where slurmctld is segfaulting a few times a day. We had this problem with SLURM 23.11.8 and now with 23.11.10 as well, though the problem only appears on one of the several SLURM clusters we have, and all of them use one of those versions of SLURM. I was wondering if anyone has encountered a similar issue and has any thoughts on how to prevent this. Obviously we use "SchedulerType=sched/backfill" but strangely when I switched to sched/builtin for a while there were still slurmctld segfaults. We also set "SchedulerParameters=enable_user_top,bf_max_job_test=2000". I have tried turning those off but it did not help. I have also tried tweaking several other settings to no avail. Most of the cluster runs Rocky Linux 8.10 (including the slurmctld system) though we still have some Scientific Linux 7.9 compute nodes (we compile SLURM separately for those). Here is the crash-time error from journalctl: Oct 02 06:31:20 our.host.name kernel: sched_agent[2048355]: segfault at 8 ip 00007fec755d7ea8 sp 00007fec6bffe7e8 error 4 in libslurmfull.so[7fec7555a000+1f4000] Oct 02 06:31:20 our.host.name kernel: Code: 48 39 c1 7e 19 48 c1 f8 06 ba 01 00 00 00 48 d3 e2 48 f7 da 48 0b 54 c6 10 48 21 54 c7 10 c3 b8 00 00 00 00 eb da 48 8b 4f 08 <48> 39 4e 08 48 0f 4e 4e 08 49 89 c9 48 83 f9 3f 76 4e ba 40 00 00 Oct 02 06:31:20 our.host.name systemd[1]: Started Process Core Dump (PID 2169426/UID 0). Oct 02 06:31:20 our.host.name systemd-coredump[2169427]: Process 2048344 ( slurmctld) of user 991 dumped core. This is followed by a list of each of the dozen or so related threads. The one which is dumping core is first and looks like this: Stack trace of thread 2048355: #0 0x00007fec755d7ea8 bit_and_not (libslurmfull.so) #1 0x000000000044531f _job_alloc (slurmctld) #2 0x000000000044576b _job_alloc_whole_node_internal (slurmctld) #3 0x0000000000446e6d gres_ctld_job_alloc_whole_node (slurmctld) #4 0x00007fec722e29b8 job_res_add_job (select_cons_tres.so) #5 0x00007fec722f7c32 select_p_select_nodeinfo_set (select_cons_tres.so) #6 0x00007fec756e7dc7 select_g_select_nodeinfo_set (libslurmfull.so) #7 0x0000000000496eb3 select_nodes (slurmctld) #8 0x0000000000480826 _schedule (slurmctld) #9 0x00007fec753421ca start_thread (libpthread.so.0) #10 0x00007fec745f78d3 __clone (libc.so.6) I have run slurmctld with "debug5" level logging and it appears that the error occurs right after backfill considers a large number of jobs. Slurmctld could be failing at the end of backfill or when doing something which happens just after backfill runs. Usually this is the last message before the crash: [2024-09-25T18:39:42.076] slurmscriptd: debug: _slurmscriptd_mainloop: finished If anyone has any thoughts or advice on this that would be appreciated. Thank you. -- Marcus Lauer Systems Administrator CETS Group, Research Support

2 2

The hostname resolution case sensitive
by Bill 11 Nov '24

11 Nov '24

Hi, I want to confirm that the hostname resolution is case sensitive in SLURM ? Many thanks, Bill

3 5

Discrepancy in GPU Hour Reporting Between sacct and sreport in SLURM
by Manisha Yadav 11 Nov '24

11 Nov '24

I am trying to find the GPU hour utilization for a user during a specific time period using the sacct and sreport commands. However, I am noticing a significant difference between the outputs of these two commands. Could you explain the reasons for this discrepancy? Are there specific factors or configurations in SLURM that could lead to variations in the reported GPU hours? There is a significant discrepancy in the results produced for GPU hour utilization by the sacct and sreport commands in SLURM Thanks, Manisha ------------------------------------------------------------------------------------------------------------ [ C-DAC is on Social-Media too. Kindly follow us at: Facebook: https://www.facebook.com/CDACINDIA & Twitter: @cdacindia ] This e-mail is for the sole use of the intended recipient(s) and may contain confidential and privileged information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies and the original message. Any unauthorized review, use, disclosure, dissemination, forwarding, printing or copying of this email is strictly prohibited and appropriate legal action will be taken. ------------------------------------------------------------------------------------------------------------

1 0

Access denied by pam_slurm_adopt
by John Hearns 10 Nov '24

10 Nov '24

I have cluster which uses Slurm 23.11.6 When I submit a multi-node job and run something like clush -b -w $SLURM_JOB_NODELIST "date" very often the ssh command fails with: Access denied by pam_slurm_adopt: you have no active jobs on this node This will happen maybe on 50% of the nodes There is the same behaviour of I salloc a number of nodes then try to ssh to a node. I have traced this to slurmstepd spawning a long sleep, which I believe allows proctrackd to 'see' if a job is active. On nodes that I can ssh into: root 3211 1 0 Nov08 ? 00:00:00 /usr/sbin/slurmd --systemd root 3227 1 0 Nov08 ? 00:00:00 /usr/sbin/slurmstepd infinity root 24322 1 0 15:40 ? 00:00:00 slurmstepd: [15709.extern] root 24326 24322 0 15:40 ? 00:00:00 \_ sleep 100000000 On nodes where I cannot ssh: root 3226 1 0 Nov08 ? 00:00:00 /usr/sbin/slurmd --systemd root 3258 1 0 Nov08 ? 00:00:00 /usr/sbin/slurmstepd infinity Maybe I am not understanding something here? ps. I ahve tried to run the pam_slurm_adopt module with options to debug, and have not found anything useful John H

2 2

Change primary alloc node
by Bhaskar Chakraborty 04 Nov '24

04 Nov '24

Hi, Is there a way to change/control the primary node (i.e. where the initial task starts) as part of a job's allocation. For eg, if a job requires 6 CPUs & its allocation is distributed over 3 hosts h1, h2 & h3 I find that it always starts the task in 1 particularnode (say h1) irrespective of how many slots were available in the hosts. Can we somehow let slurm have the primary node as h2? Is there any C-API inside select plugin which can do this trick if we were to control it through the configured select plugin? Thanks.-Bhaskar.

2 4

Slurm Job Sched Priority
by Bhaskar Chakraborty 03 Nov '24

03 Nov '24

Hello, Is there any DS in slurmctld which portrays the dynamic relative priority of pending jobs? We are trying to use slurm for developing a scheduling solution and 1 of the problems we face at the outset is how to determinethe order of scheduling for pending jobs. One option is to find scheduling iteration window begin & close pointers & cache the job ids as seen in order & then make them the priority order at that point of time. ( This means for 500 pending jobs, say, if we can find which are the slurmctld calls which mark the beginning & end of a sched iteration then we can use the scheduling orderof jobs as the relative priority order for that period of time, of course it may change depending on fairshare, user initiated priority modification etc. ) A concrete existing data structure showing the dynamic priority itself from slurmctld would be handy. Help appreciated. Thanks! Bhaskar.

3 4

Re: Change primary alloc node
by Bhaskar Chakraborty 31 Oct '24

31 Oct '24

Hello, Just to add some context here. We plan to use slurm for developing a sched solution which interacts with a backend system. Now, the backend system has pieces of h/w which require specific host in the allocation to be the primary/master host wherein the initial task would be launched, this in turn is driven by the job's placement orientation on the h/w itself. So, our primary task should launch in the asked primary host while secondary / remote tasks would subsequently get started on other hosts. Hope this brings some context to the problem as to why a specific host is necessary to be the starting host. Regards,Bhaskar. On Thursday 31 October, 2024 at 12:04:37 am IST, Laura Hild <lsh(a)jlab.org> wrote: I think if you tell the list why you care which of the Nodes is BatchHost, they may be able to provide you with a better solution. ________________________________________ Od: Bhaskar Chakraborty via slurm-users <slurm-users(a)lists.schedmd.com> Poslano: sreda, 30. oktober 2024 12:35 Za: slurm-users(a)schedmd.com Zadeva: [slurm-users] Change primary alloc node Hi, Is there a way to change/control the primary node (i.e. where the initial task starts) as part of a job's allocation. For eg, if a job requires 6 CPUs & its allocation is distributed over 3 hosts h1, h2 & h3 I find that it always starts the task in 1 particular node (say h1) irrespective of how many slots were available in the hosts. Can we somehow let slurm have the primary node as h2? Is there any C-API inside select plugin which can do this trick if we were to control it through the configured select plugin? Thanks. -Bhaskar.

3 2

Why AllowAccounts not work in slurm-23.11.6
by daijiangkuicgo＠gmail.com 30 Oct '24

30 Oct '24

I have set AllowAccounts=sunlabc5hpc,root, but it doesn’t seem to work. User c010637 is not part of the sunlabc5hpc account but is still able to use the sunlabc5hpc partition. I have tried setting EnforcePartLimits to ALL, ANY, and NO, but none of these options resolved the issue. [c010637@sl-login ~]$ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST cpu* up infinite 3 mix sl-c[0035,0042-0043] cpu* up infinite 1 idle sl-c0036 gpu up infinite 3 idle sl-c[0045-0047] sunlabc5hpc up infinite 1 idle sl-c0048 [c010637@sl-login ~]$ scontrol show partition sunlabc5hpc PartitionName=sunlabc5hpc AllowGroups=ALL AllowAccounts=sunlabc5hpc,root AllowQos=ALL AllocNodes=ALL Default=NO QoS=N/A DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO GraceTime=0 Hidden=NO MaxNodes=UNLIMITED MaxTime=UNLIMITED MinNodes=0 LLN=NO MaxCPUsPerNode=UNLIMITED MaxCPUsPerSocket=UNLIMITED Nodes=sl-c0048 PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO OverTimeLimit=NONE PreemptMode=OFF State=UP TotalCPUs=256 TotalNodes=1 SelectTypeParameters=NONE JobDefaults=(null) DefMemPerNode=UNLIMITED MaxMemPerNode=UNLIMITED TRES=cpu=256,mem=515000M,node=1,billing=256,gres/gpu=8 [c010637@sl-login ~]$ sacctmgr list assoc format=cluster,user,account%20,qos user=$USER Cluster User Account QOS ---------- ---------- -------------------- -------------------- snowhpc c010637 c010637_bank normal [c010637@sl-login ~]$ sacctmgr list account sunlabc5hpc Account Descr Org ---------- -------------------- -------------------- sunlabc5h+ sunlabc5hpc sunlabc5hpc [c010637@sl-login ~]$ sacctmgr show assoc where Account=sunlabc5hpc format=User,Account User Account ---------- ---------- sunlabc5h+ c010751 sunlabc5h+ snowdai sunlabc5h+

5 9

Re: 转发: What is the safe upgrade path when upgrade from slurm21.08 and mariadb5.5?
by taleintervenor＠sjtu.edu.cn 30 Oct '24

30 Oct '24

Thanks for all your help. So it seems we can skip the trouble of compiling SLURM over different mariadb versions. Tianyang Zhang SJTU Network Information Center 发件人: Sid Young <sid.young(a)gmail.com> 发送时间: 2024年10月30日 7:19 收件人: Andrej Sec <andrej.sec(a)savba.sk> 抄送: taleintervenor(a)sjtu.edu.cn; slurm-users(a)lists.schedmd.com 主题: Re: [slurm-users] Re: 转发: What is the safe upgrade path when upgrade from slurm21.08 and mariadb5.5? I recently upgraded from 20.11 to 24.05.2, before moving the cluster from CentOS 7.9 to Oracle Linux 8.10 The DB upgrade should be pretty simple, do a mysqldump first, then uninstall the old DB, change the repo's and install the new DB version. It should recognise the DB files on disk and access them. Do another DB backup on the new DB version. then roll through the Slurm upgrades. I picked the first and last version of each release, and systematically went through each node till it was done. First the slurm controller node, then the compute nodes. To avoid Job loss, drain the nodes or you end up with a situation where the slurmd can't talk to the running slurmstepd and the job(s) gets lost. (Shows as a "Protocol Error"). Ole sent me a link to this guide which mostly worked. https://wiki.fysik.dtu.dk/Niflheim_system/Slurm_installation/#upgrade-slurm… Sid Young W: https://off-grid-engineering.com On Tue, Oct 29, 2024 at 6:33 PM Andrej Sec via slurm-users <slurm-users(a)lists.schedmd.com <mailto:slurm-users@lists.schedmd.com> > wrote: Hi, we are facing a similar task. We have a Slurm 22.05 / MariaDB 5.5.68 environment and want to upgrade to a newer version. According to the documentation, it’s recommended to upgrade from 22.05 to a maximum of 23.11 in one step. With the MariaDB upgrade, there’s a challenge between 10.1 and 10.2+ due to incompatible changes (https://mariadb.com/kb/en/changes-improvements-in-mariadb-10-2). This upgrade, as I understand from the documentation, requires at least slurm 22.05, where it is automatically handled by the slurmdbd service. In the test lab, we performed the following tests: a. Incremental upgrade - according to MariaDB recommendations: 1. Upgrade MariaDB 5.5.68 -> 10.1.48 -> 10.2.44 2. Start the Slurm suite 22.05, checking content after each MariaDB upgrade step. During the 10.1 -> 10.2 upgrade, the slurmdbd service automatically converted the database to the required format. We had enabled general.log in MariaDB, allowing detailed inspection of database changes during conversion. 3. Upgrade slurmdbd to version 23.11 4. Upgrade slurmctld to version 23.11 5. Upgrade slurmd to version 23.11 6. Check the database content and compare tests before and after the upgrade (we used various reports with scontrol, sreport, sacct, sacctmgr for verification). b. Direct MariaDB upgrade from 5.5.68 to 10.2.44 using the same approach. According to the tests, this resulted in the same state as the incremental approach. PS: If you proceed with the upgrade, I would appreciate it if you could let us know about any potential challenges you encountered. Andrej Sec nscc, Bratislava, Slovakia _____ Od: "hermes via slurm-users" <slurm-users(a)lists.schedmd.com <mailto:slurm-users@lists.schedmd.com> > Komu: slurm-users(a)lists.schedmd.com <mailto:slurm-users@lists.schedmd.com> Odoslané: pondelok, 28. október 2024 8:48:19 Predmet: [slurm-users] =?eucgb2312_cn?q?=D7=AA=B7=A2=3A_What_is_the_safe_upgrade_path_when_upgrade_from_slurm21=2E08_and_mariadb5=2E5=3F?= Hi everyone: We are currently running business on SLURM21.08 and mariadb5.5. When talking about the upgrade, we need to keep all the users and jobs history data. And we see the official document wrote: “When upgrading an existing accounting database to MariaDB 10.2.1 or later from an older version of MariaDB or any version of MySQL, ensure you are running slurmdbd 22.05.7 or later. These versions will gracefully handle changes to MariaDB default values that can cause problems for slurmdbd.” So is this mean we have to firstly build SLURM>22.05 over mariadb5.5, and do the SLURM upgrade. Then upgrade the mariadb to newer version, and rebuild the same version of SLURM over new mariadb-devel? And is it safe to jump directly from mariadb5.5 to latest version? How can we check whether the slurm have correctly inherited the historical data? Thanks, Tianyang Zhang SJTU Network Information Center -- slurm-users mailing list -- slurm-users(a)lists.schedmd.com <mailto:slurm-users@lists.schedmd.com> To unsubscribe send an email to slurm-users-leave(a)lists.schedmd.com <mailto:slurm-users-leave@lists.schedmd.com> -- slurm-users mailing list -- slurm-users(a)lists.schedmd.com <mailto:slurm-users@lists.schedmd.com> To unsubscribe send an email to slurm-users-leave(a)lists.schedmd.com <mailto:slurm-users-leave@lists.schedmd.com>

1 0

转发: What is the safe upgrade path when upgrade from slurm21.08 and mariadb5.5?
by taleintervenor＠sjtu.edu.cn 29 Oct '24

29 Oct '24

Hi everyone: We are currently running business on SLURM21.08 and mariadb5.5. When talking about the upgrade, we need to keep all the users and jobs history data. And we see the official document wrote: “When upgrading an existing accounting database to MariaDB 10.2.1 or later from an older version of MariaDB or any version of MySQL, ensure you are running slurmdbd 22.05.7 or later. These versions will gracefully handle changes to MariaDB default values that can cause problems for slurmdbd.” So is this mean we have to firstly build SLURM>22.05 over mariadb5.5, and do the SLURM upgrade. Then upgrade the mariadb to newer version, and rebuild the same version of SLURM over new mariadb-devel? And is it safe to jump directly from mariadb5.5 to latest version? How can we check whether the slurm have correctly inherited the historical data? Thanks, Tianyang Zhang SJTU Network Information Center

3 2

Job pre / post submit scripts
by Bhaskar Chakraborty 29 Oct '24

29 Oct '24

Hi, Is there an option in slurm to launch a custom script at the time of job submission through sbatchor salloc? The script should run with submit user permission in submit area. The idea is that we need to enquire something which characterises our job’s requirement like CPUslots, memory etc from a central server and we do need read access to user area prior to that. In our use case the user doesn’t necessarily know beforehand what kind of resource his job needs.(Hence, the need for such a script which will contact the server with user area info.) Based on it we can modify the job a little later. A post submit script, if available, would inform us the slurm job id as well, it would get called just after the job has entered the system and prior to its scheduling. Thanks,Bhaskar. Sent from Yahoo Mail for iPad

4 4

Scheduling oddity with multiple GPU types in same partition
by Kevin M. Hildebrand 29 Oct '24

29 Oct '24

We have a 'gpu' partition with 30 or so nodes, some with A100s, some with H100s, and a few others. It appears that when (for example) all of the A100 GPUs are in use, if there are additional jobs requesting A100 GPUs pending, and those jobs have the highest priority in the partition, then jobs submitted for H100s won't run even if there are idle H100s. This is a small subset of our present pending queue- the four bottom jobs should be running, but aren't. The top pending job shows reason 'Resources' while the rest all show 'Priority'. Any thoughts on why this might be happening? JOBID PRIORITY TRES_ALLOC 8317749 501490 cpu=4,mem=80000M,node=1,billing=48,gres/gpu=1,gres/gpu:a100=1 8317750 501490 cpu=4,mem=80000M,node=1,billing=48,gres/gpu=1,gres/gpu:a100=1 8317745 501490 cpu=4,mem=80000M,node=1,billing=48,gres/gpu=1,gres/gpu:a100=1 8317746 501490 cpu=4,mem=80000M,node=1,billing=48,gres/gpu=1,gres/gpu:a100=1 8338679 500060 cpu=4,mem=64G,node=1,billing=144,gres/gpu=1,gres/gpu:h100=1 8338678 500060 cpu=4,mem=64G,node=1,billing=144,gres/gpu=1,gres/gpu:h100=1 8338677 500060 cpu=4,mem=64G,node=1,billing=144,gres/gpu=1,gres/gpu:h100=1 8338676 500060 cpu=4,mem=64G,node=1,billing=144,gres/gpu=1,gres/gpu:h100=1 Thanks, Kevin -- Kevin Hildebrand University of Maryland Division of IT

2 1

I need to limit the number of jobs per user per partition
by Stephen Connolly 28 Oct '24

28 Oct '24

I am unable to limit the number of jobs per user per partition. I have searched the internet, the forums and the slurm documentation. I created a partition with a QOS having MaxJobsPU=1 and MaxJobsPA=1 Created a user stephen with account=stephen and MaxJobs=1 However if I sbatch a test job (sleep 180) multiple times they all run concurrently. I am at a loss of what else to do. Help would be very appreciated . Thank you -- Stephen Connolly JSI Data Systems Ltd 613-727-9353 stephen(a)jsidata.ca

2 1

jobs dropping
by Mihai Ciubancan 24 Oct '24

24 Oct '24

Hello, We are trying to run some PiconGPU codes on a machine with 8x100H, susing slurm. But the jobs don't run, and are not in the queue. In slurmd logs I have: [2024-10-24T09:50:40.934] CPU_BIND: _set_batch_job_limits: Memory extracted from credential for StepId=1079.batch job_mem_limit= 648000 [2024-10-24T09:50:40.934] Launching batch job 1079 for UID 1009 [2024-10-24T09:50:40.938] debug: acct_gather_energy/none: init: AcctGatherEnergy NONE plugin loaded [2024-10-24T09:50:40.938] debug: acct_gather_profile/none: init: AcctGatherProfile NONE plugin loaded [2024-10-24T09:50:40.938] debug: acct_gather_interconnect/none: init: AcctGatherInterconnect NONE plugin loaded [2024-10-24T09:50:40.938] debug: acct_gather_filesystem/none: init: AcctGatherFilesystem NONE plugin loaded [2024-10-24T09:50:40.939] debug: gres/gpu: init: loaded [2024-10-24T09:50:41.022] [1079.batch] debug: cgroup/v2: init: Cgroup v2 plugin loaded [2024-10-24T09:50:41.026] [1079.batch] debug: CPUs:192 Boards:1 Sockets:2 CoresPerSocket:48 ThreadsPerCore:2 [2024-10-24T09:50:41.026] [1079.batch] debug: jobacct_gather/cgroup: init: Job accounting gather cgroup plugin loaded [2024-10-24T09:50:41.026] [1079.batch] CPU_BIND: Memory extracted from credential for StepId=1079.batch job_mem_limit=648000 step_mem_limit=648000 [2024-10-24T09:50:41.027] [1079.batch] debug: laying out the 8 tasks on 1 hosts mihaigpu2 dist 2 [2024-10-24T09:50:41.027] [1079.batch] gres_job_state gres:gpu(7696487) type:(null)(0) job:1079 flags: [2024-10-24T09:50:41.027] [1079.batch] total_gres:8 [2024-10-24T09:50:41.027] [1079.batch] node_cnt:1 [2024-10-24T09:50:41.027] [1079.batch] gres_cnt_node_alloc[0]:8 [2024-10-24T09:50:41.027] [1079.batch] gres_bit_alloc[0]:0-7 of 8 [2024-10-24T09:50:41.027] [1079.batch] debug: Message thread started pid = 459054 [2024-10-24T09:50:41.027] [1079.batch] debug: Setting slurmstepd(459054) oom_score_adj to -1000 [2024-10-24T09:50:41.027] [1079.batch] debug: switch/none: init: switch NONE plugin loaded [2024-10-24T09:50:41.027] [1079.batch] debug: task/cgroup: init: core enforcement enabled [2024-10-24T09:50:41.027] [1079.batch] debug: task/cgroup: task_cgroup_memory_init: task/cgroup/memory: total:2063720M allowed:100%(enforced), swap:0%(permissive), max:100%(2063720M) max+swap:100%(4127440M) min:30M kmem:100%(2063720M permissive) min:30M [2024-10-24T09:50:41.027] [1079.batch] debug: task/cgroup: init: memory enforcement enabled [2024-10-24T09:50:41.027] [1079.batch] debug: task/cgroup: init: Tasks containment cgroup plugin loaded [2024-10-24T09:50:41.027] [1079.batch] cred/munge: init: Munge credential signature plugin loaded [2024-10-24T09:50:41.027] [1079.batch] debug: job_container/none: init: job_container none plugin loaded [2024-10-24T09:50:41.030] [1079.batch] debug: spank: opening plugin stack /etc/slurm/plugstack.conf [2024-10-24T09:50:41.030] [1079.batch] debug: task/cgroup: task_cgroup_cpuset_create: job abstract cores are '0-63' [2024-10-24T09:50:41.030] [1079.batch] debug: task/cgroup: task_cgroup_cpuset_create: step abstract cores are '0-63' [2024-10-24T09:50:41.030] [1079.batch] debug: task/cgroup: task_cgroup_cpuset_create: job physical CPUs are '0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,96,98,100,102,104,106,108,110,112,114,116,118,120,122,124,126,128,130,132,134,136,138,140,142,144,146,148,150,152,154,156,158' [2024-10-24T09:50:41.030] [1079.batch] debug: task/cgroup: task_cgroup_cpuset_create: step physical CPUs are '0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,96,98,100,102,104,106,108,110,112,114,116,118,120,122,124,126,128,130,132,134,136,138,140,142,144,146,148,150,152,154,156,158' [2024-10-24T09:50:41.031] [1079.batch] task/cgroup: _memcg_initialize: job: alloc=648000MB mem.limit=648000MB memsw.limit=unlimited [2024-10-24T09:50:41.031] [1079.batch] task/cgroup: _memcg_initialize: step: alloc=648000MB mem.limit=648000MB memsw.limit=unlimited [2024-10-24T09:50:41.064] [1079.batch] debug levels are stderr='error', logfile='debug', syslog='quiet' [2024-10-24T09:50:41.064] [1079.batch] starting 1 tasks [2024-10-24T09:50:41.064] [1079.batch] task 0 (459058) started 2024-10-24T09:50:41 [2024-10-24T09:50:41.069] [1079.batch] _set_limit: RLIMIT_NOFILE : reducing req:1048576 to max:131072 [2024-10-24T09:51:23.066] debug: _rpc_terminate_job: uid = 64030 JobId=1079 [2024-10-24T09:51:23.067] debug: credential for job 1079 revoked [2024-10-24T09:51:23.067] [1079.batch] debug: Handling REQUEST_SIGNAL_CONTAINER [2024-10-24T09:51:23.067] [1079.batch] debug: _handle_signal_container for StepId=1079.batch uid=64030 signal=18 [2024-10-24T09:51:23.068] [1079.batch] Sent signal 18 to StepId=1079.batch [2024-10-24T09:51:23.068] [1079.batch] debug: Handling REQUEST_SIGNAL_CONTAINER [2024-10-24T09:51:23.068] [1079.batch] debug: _handle_signal_container for StepId=1079.batch uid=64030 signal=15 [2024-10-24T09:51:23.068] [1079.batch] error: *** JOB 1079 ON mihaigpu2 CANCELLED AT 2024-10-24T09:51:23 *** [2024-10-24T09:51:23.069] [1079.batch] Sent signal 15 to StepId=1079.batch [2024-10-24T09:51:23.069] [1079.batch] debug: Handling REQUEST_STATE [2024-10-24T09:51:23.071] [1079.batch] task 0 (459058) exited. Killed by signal 15. [2024-10-24T09:51:23.090] [1079.batch] debug: Handling REQUEST_STATE [2024-10-24T09:51:23.141] [1079.batch] debug: Handling REQUEST_STATE [2024-10-24T09:51:23.241] [1079.batch] debug: Handling REQUEST_STATE [2024-10-24T09:51:23.741] [1079.batch] debug: Handling REQUEST_STATE [2024-10-24T09:51:24.073] [1079.batch] debug: signaling condition [2024-10-24T09:51:24.073] [1079.batch] debug: jobacct_gather/cgroup: fini: Job accounting gather cgroup plugin unloaded [2024-10-24T09:51:24.073] [1079.batch] debug: task/cgroup: fini: Tasks containment cgroup plugin unloaded [2024-10-24T09:51:24.073] [1079.batch] debug: get_exit_code task 0 killed by cmd [2024-10-24T09:51:24.073] [1079.batch] job 1079 completed with slurm_rc = 0, job_rc = 15 [2024-10-24T09:51:24.075] [1079.batch] debug: Message thread exited [2024-10-24T09:51:24.154] [1079.batch] done with job Anyone have any idea what could be the problem? Thank you, Mihai

2 1

Randomly draining nodes
by Nacereddine Laddaoui 24 Oct '24

24 Oct '24

Hello everyone, I’ve recently encountered an issue where some nodes in our cluster enter a drain state randomly, typically after completing long-running jobs. Below is the output from the |sinfo| command showing the reason *“Prolog error”* : |root@controller-node:~# sinfo -R REASON USER TIMESTAMP NODELIST Prolog error slurm 2024-09-24T21:18:05 node[24,31] | When checking the |slurmd.log| files on the nodes, I noticed the following errors: |[2024-09-24T17:18:22.386] [217703.extern] error: _handle_add_extern_pid_internal: Job 217703 can't add pid 3311892 to jobacct_gather plugin in the extern_step. **(repeated 90 times)** [2024-09-24T17:18:22.917] [217703.extern] error: _handle_add_extern_pid_internal: Job 217703 can't add pid 3313158 to jobacct_gather plugin in the extern_step. ... [2024-09-24T21:17:45.162] launch task StepId=217703.0 request from UID:54059 GID:1600 HOST:<SLURMCTLD_IP> PORT:53514 [2024-09-24T21:18:05.166] error: Waiting for JobId=217703 REQUEST_LAUNCH_PROLOG notification failed, giving up after 20 sec [2024-09-24T21:18:05.166] error: slurm_send_node_msg: [(null)] slurm_bufs_sendto(msg_type=RESPONSE_SLURM_RC_MSG) failed: Unexpected missing socket error [2024-09-24T21:18:05.166] error: _rpc_launch_tasks: unable to send return code to address:port=<SLURMCTLD_IP>:53514 msg_type=6001: No such file or directory | If you know how to solve these errors, please let me know. I would greatly appreciate any guidance or suggestions for further troubleshooting. Thank you in advance for your assistance. Best regards, -- Télécom Paris <https://www.telecom-paris.fr> *Nacereddine LADDAOUI* Ingénieur de Recherche et de Développement 19 place Marguerite Perey CS 20031 91123 Palaiseau Cedex Site web Télécom Paris <https://www.telecom-paris.fr>X Télécom Paris <https://twitter.com/TelecomParis_>Facebook Télécom Paris <https://www.facebook.com/TelecomParis>LinkedIn Télécom Paris <https://www.linkedin.com/school/telecom-paris/>Instagram Télécom Paris <https://www.instagram.com/telecom_paris/>Blog Télécom Paris <https://imtech.wp.imt.fr> Une école de l'IMT <https://www.imt.fr>

6 9

loss of "unchangeable" node features
by Laura Hild 23 Oct '24

23 Oct '24

Has anyone else noticed, somewhere between versions 22.05.11 and 23.11.9, losing fixed Features defined for a node in slurm.conf, and instead now just having those controlled by a NodeFeaturesPlugin like node_features/knl_generic?

1 0

Slurm version 24.05.4 is now available (CVE-2024-48936)
by Tim Wickberg 23 Oct '24

23 Oct '24

Slurm version 24.05.4 is now available and includes a fix for a recently discovered security issue with the new stepmgr subsystem. SchedMD customers were informed on October 9th and provided a patch on request; this process is documented in our security policy. [1] A mistake in authentication handling in stepmgr could permit an attacker to execute processes under other users' jobs. This is limited to jobs explicitly running with --stepmgr, or on systems that have globally enabled stepmgr through "SlurmctldParameters=enable_stepmgr" in their configuration. CVE-2024-48936. Downloads are available at https://www.schedmd.com/downloads.php . Release notes follow below. - Tim [1] https://www.schedmd.com/security-policy/ -- Tim Wickberg Chief Technology Officer, SchedMD LLC Commercial Slurm Development and Support > * Changes in Slurm 24.05.4 > ========================== > -- Fix generic int sort functions. > -- Fix user look up using possible unrealized uid in the dbd. > -- Fix FreeBSD compile issue with tls/none plugin. > -- slurmrestd - Fix regressions that allowed slurmrestd to be run as SlurmUser > when SlurmUser was not root. > -- mpi/pmix fix race conditions with het jobs at step start/end which could > make srun to hang. > -- Fix not showing some SelectTypeParameters in scontrol show config. > -- Avoid assert when dumping removed certain fields in JSON/YAML. > -- Improve how shards are scheduled with affinity in mind. > -- Fix MaxJobsAccruePU not being respected when MaxJobsAccruePA is set > in the same QOS. > -- Prevent backfill from planning jobs that use overlapping resources for the > same time slot if the job's time limit is less than bf_resolution. > -- Fix memory leak when requesting typed gres and --[cpus|mem]-per-gpu. > -- Prevent backfill from breaking out due to "system state changed" every 30 > seconds if reservations use REPLACE or REPLACE_DOWN flags. > -- slurmrestd - Make sure that scheduler_unset parameter defaults to true even > when the following flags are also set: show_duplicates, skip_steps, > disable_truncate_usage_time, run_away_jobs, whole_hetjob, > disable_whole_hetjob, disable_wait_for_result, usage_time_as_submit_time, > show_batch_script, and or show_job_environment. Additionaly, always make > sure show_duplicates and disable_truncate_usage_time default to true when > the following flags are also set: scheduler_unset, scheduled_on_submit, > scheduled_by_main, scheduled_by_backfill, and or job_started. This effects > the following endpoints: > 'GET /slurmdb/v0.0.40/jobs' > 'GET /slurmdb/v0.0.41/jobs' > -- Ignore --json and --yaml options for scontrol show config to prevent mixing > output types. > -- Fix not considering nodes in reservations with Maintenance or Overlap flags > when creating new reservations with nodecnt or when they replace down nodes. > -- Fix suspending/resuming steps running under a 23.02 slurmstepd process. > -- Fix options like sprio --me and squeue --me for users with a uid greater > than 2147483647. > -- fatal() if BlockSizes=0. This value is invalid and would otherwise cause the > slurmctld to crash. > -- sacctmgr - Fix issue where clearing out a preemption list using > preempt='' would cause the given qos to no longer be preempt-able until set > again. > -- Fix stepmgr creating job steps concurrently. > -- data_parser/v0.0.40 - Avoid dumping "Infinity" for NO_VAL tagged "number" > fields. > -- data_parser/v0.0.41 - Avoid dumping "Infinity" for NO_VAL tagged "number" > fields. > -- slurmctld - Fix a potential leak while updating a reservation. > -- slurmctld - Fix state save with reservation flags when a update fails. > -- Fix reservation update issues with parameters Accounts and Users, when > using +/- signs. > -- slurmrestd - Don't dump warning on empty wckeys in: > 'GET /slurmdb/v0.0.40/config' > 'GET /slurmdb/v0.0.41/config' > -- Fix slurmd possibly leaving zombie processes on start up in configless when > the initial attempt to fetch the config fails. > -- Fix crash when trying to drain a non-existing node (possibly deleted > before). > -- slurmctld - fix segfault when calculating limit decay for jobs with an > invalid association. > -- Fix IPMI energy gathering with multiple sensors. > -- data_parser/v0.0.39 - Remove xassert requiring errors and warnings to have a > source string. > -- slurmrestd - Prevent potential segfault when there is an error parsing an > array field which could lead to a double xfree. This applies to several > endpoints in data_parser v0.0.39, v0.0.40 and v0.0.41. > -- scancel - Fix a regression from 23.11.6 where using both the --ctld and > --sibling options would cancel the federated job on all clusters instead of > only the cluster(s) specified by --sibling. > -- accounting_storage/mysql - Fix bug when removing an association > specified with an empty partition. > -- Fix setting multiple partition state restore on a job correctly. > -- Fix difference in behavior when swapping partition order in job submission. > -- Fix security issue in stepmgr that could permit an attacker to execute > processes under other users' jobs. CVE-2024-48936.

1 0

HOST CPUs not used by jobs
by Bhaskar Chakraborty 21 Oct '24

21 Oct '24

I have a SLURM configuration of 2 hosts with 6 + 4 CPUs. I am submitting jobs with sbatch -n <CPU slots> <job script>. However, I see that even when I have exhausted all 10 CPU slots for the running jobs it's still allowing subsequent jobs to run ! The CPU slots availability is also show as full for the 2 hosts. No job is found pending. What could be problem? My Slurm.conf looks like (host names are changed to generic): ClusterName=MyClusterControlMachine=host1ControlAddr=<some address>SlurmUser=slurmsa#AuthType=auth/mungeStateSaveLocation=/var/spool/slurmdSlurmdSpoolDir=/var/spool/slurmdSlurmctldLogFile=/var/log/slurm/slurmctld.logSlurmdDebug=3SlurmctldDebug=6SlurmdLogFile=/var/log/slurm/slurmd.logAccountingStorageType=accounting_storage/slurmdbdAccountingStorageHost=host1#AccountingStoragePass=medslurmpass#AccountingStoragePass=/var/run/munge/munge.socket.2AccountingStorageUser=slurmsa#TaskPlugin=task/cgroupNodeName=host1 CPUs=6 SocketsPerBoard=3 CoresPerSocket=2 ThreadsPerCore=1 State=UNKNOWNNodeName=host2 CPUs=4 ThreadsPerCore=1 State=UNKNOWNPartitionName=debug Nodes=host1,host2 Default=YES MaxTime=INFINITE State=UPJobAcctGatherType=jobacct_gather/linuxJobAcctGatherFrequency=30 SelectType=select/cons_tresSelectTypeParameters=CR_CPUTaskPlugin=task/affinity Thanks in advance for any help! Regards,Bhaskar.

2 1

Customization of (error) messages after the job submission
by Sebastian Sitkiewicz 21 Oct '24

21 Oct '24

Dear SLUR Users and Administrators, I am interested in a way to customize the job submission exit statuses (mainly error codes) after the job has already been queued by the SLURM controller. We aim to provide more user-friendly messages and reminders in case of any errors or obstacles (also adjusted to our QoS/account system). For example, in the case of exceeding CPU minutes of given QoS (or account) and after the (successful) job submission, we would like to notify the user that his job has been queued (as it should) but won’t start until the CPU minutes limits are increased (and that he should contact the administrators to apply for more resources). Similarly, if the user queued a job that cannot be launched immediately because of exceeding the MaxJobs limit (per user), we would like to also give him some additional message after the srun/sbatch submission. We want to provide such information immediately after the job submission, without the need to check the status using `squeue` by the user. In the Job Launch Guide (https://slurm.schedmd.com/job_launch.html) there are distinguished following steps: 1. Call job_submit plugins to modify the request as appropriate 2. Validate that the options are valid for this user (e.g. valid partition name, valid limits, etc.) 3. Determine if this job is the highest priority runnable job, if so then really try to allocate resources for it now, otherwise only validate that it could run if no other jobs existed 4. Determine which nodes could be used for the job. If the feature specification uses an exclusive OR option, then multiple iterations of the selection process below will be required with disjoint sets of nodes 5. Call the select plugin to select the best resources for the request 6. The select plugin will consider network topology and the topology within a node (e.g. sockets, cores, and threads) to select the best resources for the job 7. If the job can not be initiated using available resources and preemption support is configured, the select plugin will also determine if the job can be initiated after preempting lower priority jobs. If so then initiate preemption as needed to start the job. From my understanding, to achieve our goal one would need to have access to source code or plugin related to point 2 (and some part of point 3). Unfortunately, the job_submit (lua) plugin from point 1 (and the cli_filter plugin as well) cannot be used because it only has access to the information on the parameters of the submitted job and the SLURM partitions (but not the QoS/account usage and their limits). Is there any way to extend the customization of job submission to include such features? Best regards, Sebastian -- dr inż. Sebastian Sitkiewicz Politechnika Wrocławska Wrocławskie Centrum Sieciowo-Superkomputerowe Dział Usług Obliczeniowych Wyb. Wyspiańskiego 27 50-370 Wrocław www.wcss.pl

1 0

Tracking costs - one single pool of credits, variable costs per partition
by John Snowdon 18 Oct '24

18 Oct '24

We are trying to design the charging and accounting system for our new institutional HPC facility and I'm having difficulty understanding exactly how we can use sacctmgr to achieve what we need. Until now, our previous HPC facilities have all operated as free delivery and we have not needed to track costs by user/group/project. Account codes have been purely optional. However, our new facility will be split into various resource types, with free partitions and paid/priority/reserved partitions across those resource types. All jobs will need to be submitted with an account code. For users submitting to 'free' partitions we don't need to track resource units against a balance, but the submitted account code would still be used for reporting purposes (i.e. "free resources accounted for % of all use by this project in August-September"). When submitting to a 'paid' partition, the account code needs to be checked to ensure it has a positive balance (or a balance that will not go past some negative threshold). Each of the 'paid' partitions may (will) have a different resource unit cost. A simple example: - Submit to a generic CPU paid partition -- 1 resource unit/token/credit/£/$ per allocated cpu, per hour of compute - Submit to a high-speed, non-blocking CPU paid partition -- 2 resource unit/token/credit/£/$ per allocated cpu, per hour of compute - Submit to a GPU paid partition -- 4 resource unit/token/credit/£/$ per allocated GPU card, per hour of compute We need to have *one* pool of resource units/tokens/credits per account - let's say 1000 credits, and a group of users may well decide to spend all of their credits on the generic CPU partition, all on the GPU partition, or some mixture of the two. So in the above examples, assuming one user (or group of users sharing the same account code) submit a 2 hour job to all three partitions, their one, single account code should be charged: - 2 units for the generic CPU partition - 4 units for the job on the low latency partition - 8 units for the gpu partition. - A total of 14 credits removed from their single account code Is this feasible to achieve without having to allocate credits to each of the partitions for an account, or creating a QOS variant for each and every combination of account and partition? John Snowdon Senior Research Infrastructure Engineer (HPC) Research Software Engineering Catalyst Building, Room 2.01 Newcastle University 3 Science Square Newcastle Helix Newcastle upon Tyne NE4 5TG https://rse.ncldata.dev/

2 1

How do you guys track which GPU is used by which job ?
by Sylvain MARET 18 Oct '24

18 Oct '24

Hey guys ! I'm looking to improve GPU monitoring on our cluster. I want to install this https://github.com/NVIDIA/dcgm-exporter and saw in the README that it can support tracking of job id : https://github.com/NVIDIA/dcgm-exporter?tab=readme-ov-file#enabling-hpc-job… However I haven't been able to see any examples on how to do it nor does slurm seem to expose this information by default. Does anyone do this here ? And if so do you have any examples I could try to follow ? If you have advise on best practices to monitor GPU I'd be happy to hear it out ! Regards, Sylvain Maret

5 6

Job information is not being added to accounting database on new setup
by Adrian Brady 17 Oct '24

17 Oct '24

Hi Everyone, I'm a new to slurm administration and looking for a bit of help! Just added Accounting to an existing cluster but job information is not being added to the Accounting Mariadb. When I submit a test job it gets scheduled fine and its visible with squeue, I get nothing returned from sacct! I have turned up the logging to debug5 on both slurmctld and slurmdbd logs and can't see any errors. I believe all the comms are ok between slurmctld and slurmdbd as when I enter the sacct command I can see the database is being queried but returning nothing, because nothing has been added to the tables. The cluster tables were created fine when I ran #sacctmgr add cluster ny5ktt $ sacct JobID JobName Partition Account AllocCPUS State ExitCode ------------ ---------- ---------- ---------- ---------- ---------- -------- # tail -f slurmdbd.log [2024-10-17T12:34:45.232] debug: REQUEST_PERSIST_INIT: CLUSTER:ny5ktt VERSION:9216 UID:10001 IP:10.202.233.117 CONN:10 [2024-10-17T12:34:45.232] debug2: accounting_storage/as_mysql: acct_storage_p_get_connection: acct_storage_p_get_connection: request new connection 1 [2024-10-17T12:34:45.233] debug2: Attempting to connect to localhost:3306 [2024-10-17T12:34:45.274] debug2: DBD_GET_JOBS_COND: called [2024-10-17T12:34:45.317] debug2: DBD_FINI: CLOSE:1 COMMIT:0 [2024-10-17T12:34:45.317] debug4: accounting_storage/as_mysql: acct_storage_p_commit: got 0 commits The Mariadb is running on it own node with slurmdbd and munged for authentication. I haven't setup any accounts, users, asssociations or enforcements yet. On my lab cluster, jobs were visible in the database without these being setup. I guess I must be missing something simple in the config that is stopping jobs being reported to slurmdbd. Master Node packages # rpm -qa |grep slurm slurm-slurmdbd-20.11.9-1.el8.x86_64 slurm-libs-20.11.9-1.el8.x86_64 slurm-20.11.9-1.el8.x86_64 slurm-slurmd-20.11.9-1.el8.x86_64 slurm-perlapi-20.11.9-1.el8.x86_64 slurm-doc-20.11.9-1.el8.x86_64 slurm-contribs-20.11.9-1.el8.x86_64 slurm-slurmctld-20.11.9-1.el8.x86_64 Database Node packages # rpm -qa |grep slurm slurm-slurmdbd-20.11.9-1.el8.x86_64 slurm-20.11.9-1.el8.x86_64 slurm-libs-20.11.9-1.el8.x86_64 slurm-devel-20.11.9-1.el8.x86_64 slurm.conf # # See the slurm.conf man page for more information. # ClusterName=ny5ktt ControlMachine=ny5-pr-kttslurm-01 ControlAddr=10.202.233.71 #BackupController= #BackupAddr= # AuthType=auth/munge #CheckpointType=checkpoint/none CryptoType=crypto/munge #DisableRootJobs=NO #EnforcePartLimits=NO #Epilog= #EpilogSlurmctld= #FirstJobId=1 #MaxJobId=999999 #GresTypes= #GroupUpdateForce=0 #GroupUpdateTime=600 #JobCheckpointDir=/var/slurm/checkpoint #JobCredentialPrivateKey= #JobCredentialPublicCertificate= #JobFileAppend=0 #JobRequeue=1 #JobSubmitPlugins= #KillOnBadExit=0 #LaunchType=launch/slurm #Licenses=foo*4,bar MailProg=/bin/true MaxJobCount=200000 #MaxStepCount=40000 #MaxTasksPerNode=128 MpiDefault=none #MpiParams=ports=#-# #PluginDir= #PlugStackConfig= #PrivateData=jobs ProctrackType=proctrack/cgroup #Prolog= #PrologFlags= #PrologSlurmctld= #PropagatePrioProcess=0 #PropagateResourceLimits= #PropagateResourceLimitsExcept= #RebootProgram= ReturnToService=1 #SallocDefaultCommand= SlurmctldPidFile=/var/run/slurm/slurmctld.pid SlurmctldPort=6817 SlurmdPidFile=/var/run/slurm/slurmd.pid SlurmdPort=6818 SlurmdSpoolDir=/var/spool/slurm/d SlurmUser=slurm #SlurmdUser=root #SrunEpilog= #SrunProlog= StateSaveLocation=/var/spool/slurm/ctld SwitchType=switch/none #TaskEpilog= TaskPlugin=task/none #TaskPluginParam= #TaskProlog= #TopologyPlugin=topology/tree #TmpFS=/tmp #TrackWCKey=no #TreeWidth= #UnkillableStepProgram= #UsePAM=0 # # # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=30 #MessageTimeout=10 #ResvOverRun=0 #MinJobAge=300 #MinJobAge=43200 # CHG0057915 MinJobAge=14400 # CHG0057915 #MaxJobCount=50000 #MaxJobCount=100000 #OverTimeLimit=0 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING DefMemPerCPU=3000 #FastSchedule=1 #MaxMemPerCPU=0 #SchedulerTimeSlice=30 SchedulerType=sched/backfill SelectType=select/cons_tres #SelectTypeParameters=CR_Core #SelectTypeParameters=CR_CPU SelectTypeParameters=CR_CPU_Memory # ECR CHG0056915 10/14/2023 MaxArraySize=5001 # # # JOB PRIORITY #PriorityFlags= #PriorityType=priority/basic #PriorityDecayHalfLife= #PriorityCalcPeriod= #PriorityFavorSmall= #PriorityMaxAge= #PriorityUsageResetPeriod= #PriorityWeightAge= #PriorityWeightFairshare= #PriorityWeightJobSize= #PriorityWeightPartition= #PriorityWeightQOS= # # # LOGGING AND ACCOUNTING #AccountingStorageEnforce=0 #AccountingStorageEnforce=limits AccountingStorageHost=ny5-pr-kttslurmdb-01.ktt.schonfeld.com #AccountingStorageLoc= #AccountingStoragePass= #AccountingStoragePort= #AccountingStorageType=accounting_storage/none AccountingStorageType=accounting_storage/slurmdbd #AccountingStorageUser= AccountingStoreJobComment=YES #DebugFlags= #JobCompHost= #JobCompLoc= #JobCompPass= #JobCompPort= JobCompType=jobcomp/none #JobCompUser= #JobContainerType=job_container/none JobAcctGatherFrequency=60 JobAcctGatherType=jobacct_gather/none SlurmctldDebug=info SlurmctldLogFile=/var/log/slurm/slurmctld.log SlurmdDebug=info SlurmdLogFile=/var/log/slurm/slurmd.log #SlurmdLogFile= #SlurmSchedLogFile= #SlurmSchedLogLevel= # # # POWER SAVE SUPPORT FOR IDLE NODES (optional) #SuspendProgram= #ResumeProgram= #SuspendTimeout= #ResumeTimeout= #ResumeRate= #SuspendExcNodes= #SuspendExcParts= #SuspendRate= #SuspendTime= # # # COMPUTE NODES ##using fqdn since the ctld domain is different. Can't use regex since it's not at the end ##save 17 and 18 as headnodes #NodeName=ny5-dv-kttres-17 Sockets=1 CoresPerSocket=18 ThreadsPerCore=2 Feature=HyperThread RealMemory=102400 #NodeName=ny5-dv-kttres-18 Sockets=1 CoresPerSocket=14 ThreadsPerCore=2 Feature=HyperThread RealMemory=102400 NodeName=ny5-dv-kttres-19 Sockets=1 CoresPerSocket=12 ThreadsPerCore=2 Feature=HyperThread RealMemory=102400 NodeName=ny5-dv-kttres-[20-21] Sockets=1 CoresPerSocket=18 ThreadsPerCore=2 Feature=HyperThread RealMemory=102400 NodeName=ny5-dv-kttres-[01-16] Sockets=2 CoresPerSocket=16 ThreadsPerCore=2 Feature=HyperThread RealMemory=233472 NodeName=ny5-dv-kttres-[22-35] Sockets=2 CoresPerSocket=32 ThreadsPerCore=2 Feature=HyperThread RealMemory=346884 PartitionName=ktt_slurm_light_1 Nodes=ny5-dv-kttres-[19-21] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_medium_1 Nodes=ny5-dv-kttres-[01-08] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_medium_2 Nodes=ny5-dv-kttres-[09-16] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_medium_3 Nodes=ny5-dv-kttres-[22-28] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_medium_4 Nodes=ny5-dv-kttres-[29-35] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_large_1 Nodes=ny5-dv-kttres-[01-16] Default=YES MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 PartitionName=ktt_slurm_large_2 Nodes=ny5-dv-kttres-[22-35] Default=NO MaxTime=INFINITE State=UP OverSubscribe=FORCE:2 Slurmdbd.conf AuthType=auth/munge DbdAddr=10.202.233.72 DbdHost=ny5-pr-kttslurmdb-01 DebugLevel=debug5 LogFile=/var/log/slurm/slurmdbd.log PidFile=/tmp/slurmdbd.pid StorageType=accounting_storage/mysql StorageHost=localhost #StorageHost=10.234.132.57 StorageUser=slurm SlurmUser=slurm StoragePass=xxxxxxx #StorageUser=slurm #StorageLoc=slurm_acct_db Database tables MariaDB [slurm_acct_db]> show tables; +--------------------------------+ | Tables_in_slurm_acct_db | +--------------------------------+ | acct_coord_table | | acct_table | | clus_res_table | | cluster_table | | convert_version_table | | federation_table | | ny5ktt_assoc_table | | ny5ktt_assoc_usage_day_table | | ny5ktt_assoc_usage_hour_table | | ny5ktt_assoc_usage_month_table | | ny5ktt_event_table | | ny5ktt_job_table | | ny5ktt_last_ran_table | | ny5ktt_resv_table | | ny5ktt_step_table | | ny5ktt_suspend_table | | ny5ktt_usage_day_table | | ny5ktt_usage_hour_table | | ny5ktt_usage_month_table | | ny5ktt_wckey_table | | ny5ktt_wckey_usage_day_table | | ny5ktt_wckey_usage_hour_table | | ny5ktt_wckey_usage_month_table | | qos_table | | res_table | | table_defs_table | | tres_table | | txn_table | | user_table | +--------------------------------+ Many Thanks Adrian Disclaimer Schonfeld Strategic Advisors (UK) LLP (“SSA UK”) is authorised and regulated by The Financial Conduct Authority. SSA UK is a limited liability partnership in England and Wales (No: OC420598) and its registered office is at 54 Jermyn Street, London, SW1Y 6LX. The contents of this message, including any attachments, are meant solely for the intended recipient and may be confidential, privileged, or otherwise protected from disclosure. If you receive this message in error, immediately alert the sender by reply e-mail, delete it and any attachments or copies from your systems, and do not read, disclose, distribute, or otherwise use the information contained herein. We do not waive any confidentiality or privilege if this message was misdirected. This e-mail does not constitute an offer to sell or a solicitation to buy any securities or an offer of any investment advisory services. If you reply to this email please note that we invest in securities and do not want to receive material, non-public information and you are instructed not to communicate any such information to us. We do not agree to keep confidential any information you provide nor restrict our trading activity, except as agreed pursuant to a written confidentiality agreement duly executed by us. We reserve the right to monitor and review the content of all messages sent to or from this e-mail address.

1 0

Issue with interactive jobs
by Nerjes, Onno 17 Oct '24

17 Oct '24

Dear all, we've set up SLURM 24.05.3 on our cluster and are experiencing an issue with interactive jobs. Before, we used 21.08 and pretty much the same settings, but without these issues. We've started with a fresh DB etc. The behavior of interactive jobs is very erratic. Sometimes they start absolutely fine, at other times they die silently in the background, while the user has to wait indefinitely. We have been unable to isolate certain users or nodes affected by this. On a given node, one user might be able to start an interactive job, while another user at the same time isn't able to. The day after, the situation might be the other way around. The exception are jobs that use a reservation. These start fine every time as far as we can tell. At the same time, the number of idle nodes does not seem to influence the behavior I described above. Failed allocation on the front end: [user1@login1 ~]$ salloc salloc: Pending job allocation 5052052 salloc: job 5052052 queued and waiting for resources The same job on the backend: 2024-10-14 11:41:57.680 slurmctld: _job_complete: JobId=5052052 done 2024-10-14 11:41:57.678 slurmctld: _job_complete: JobId=5052052 WEXITSTATUS 1 2024-10-14 11:41:57.678 slurmctld: Killing interactive JobId=5052052: Communication connection failure 2024-10-14 11:41:46.666 slurmctld: sched/backfill: _start_job: Started JobId=5052052 in devel on m02n01 2024-10-14 11:41:30.096 slurmctld: sched: _slurm_rpc_allocate_resources JobId=5052052 NodeList=(null) usec=6258 Raising the debug level has not brought additional information. We were hoping, that one of you might be able to provide some insight into what the next steps in troubleshooting might be. Best regards, Onno

1 0

Dependency jobs
by adam＠bramblecfd.com 17 Oct '24

17 Oct '24

Hi Slurm Users, I am trying to figure out if there is a way you can check if a running job has any jobs queued up after it that depend on the current running job, I know you can show job info and find what dependency a job is waiting for, But more after checking if there are jobs waiting on the current job to complete using the job ID, Many Thanks, Adam

3 2

AllowAccounts partition setting
by Marko Markoc 17 Oct '24

17 Oct '24

Hi All, At one point, when we were running slurm 22, we set up a few new partitions with `AllowAccounts` setting enabled. We limit access to these partitions only to a few accounts. We manually create these accounts and users using the `sacctmgr` command. At that time, everything was working. Only users in these accounts were able to run jobs in that partition. Few months ago we upgraded slurm to 23.02.7 but just a couple of days ago we noticed that anyone could submit jobs to these partitions. In addition to that, you can provide any value to the `--account` directive and slurm will accept and record it in the job accounting database. In our initial testing, you could only provide `--account` value that has been previously created. I'm still looking into this but I wanted to check if there was any change in version 23 that could have impacted this? I don't remember seeing anything in release notes. Thanks, Marko

6 8

Problem with nodes with 1 gpu
by Jörg Striewski 16 Oct '24

16 Oct '24

i cannot send jobs to nodes with one gpu, i don't find the bug in my configuration. can someone help me ? in slurm.conf GresTypes=gpu is set this are some nodes in slurm.conf NodeName=gpu-[001-003] CPUs=8 SocketsPerBoard=1 CoresPerSocket=4 RealMemory=31000 Gres=gpu:1080:1 NodeName=gpu-[010-019] CPUs=16 SocketsPerBoard=1 CoresPerSocket=8 RealMemory=64000 Gres=gpu:1080:2 the partition for this gpu nodes is # General GPU partitions PartitionName=GPU Nodes=gpu-[001-003,010-019] AllowAccounts=staff PreemptMode=REQUEUE PriorityTier=0 DefMemPerGPU=32000 DefCpuPerGPU=8 CpuBind=none TRESBillingWeights="GRES/gpu=1000" GraceTime=300 this are the entries for some nodes in gres.conf NodeName=gpu-[001-003] Name=gpu Type=1080 File=/dev/nvidia0 NodeName=gpu-[010-019] Name=gpu Type=1080 File=/dev/nvidia[0-1] when i send a job with sbatch to gpu-001 #SBATCH --job-name=hello #SBATCH --ntasks-per-node=1 #SBATCH --output=hello_%A.out #SBATCH --time=00:10:00 #SBATCH --mail-type=ALL #SBATCH --mail-user=striewski(a)ismll.de #SBATCH --partition=GPU #SBATCH --nodelist=gpu-001 #SBATCH --gres=gpu:1 [...] i get the error sbatch: error: Batch job submission failed: Requested node configuration is not available when i send the job to a node with 2 gpu's it runs with no error, just setting --nodelist=gpu-12 has someone a hint what i made wrong ? Mit freundlichen Grüßen / kind regards -- Jörg Striewski Information Systems and Machine Learning Lab (ISMLL) Institute of Computer Science University of Hildesheim Germany post address: Universitätsplatz 1, D-31141Hildesheim, Germany visitor address: Samelsonplatz 1, D-31141 Hildesheim,Germany Tel.(+49) 05121 / 883-40392 http://www.ismll.uni-hildesheim.de

2 2

2025

2024