SLURM - Revision history

D2acosta: added why

2026-10-23T23:52:11Z

added why

← Older revision		Revision as of 18:52, 23 October 2026
Line 1:		Line 1:
			== Why ==
			Slurm (Simple Linux Utility for Resource Management) is a job scheduler. Basically use this if too many people are trying to use a system at the same time and its crashing or you want to distribute resources more fairly.

	== Installation/Config ==		== Installation/Config ==
	First you need to install the "slurm" packages and dependencies so run<syntaxhighlight lang="bash" start="1">		First you need to install the "slurm" packages and dependencies so run<syntaxhighlight lang="bash" start="1">

D2acosta at 09:34, 23 October 2026

2026-10-23T09:34:54Z

← Older revision		Revision as of 04:34, 23 October 2026
Line 47:		Line 47:
	systemctl status slurmd		systemctl status slurmd
	sinfo		sinfo
	</syntaxhighlight>If sinfo shows the node as down or drain, you can set it to idle by<syntaxhighlight lang="bash">sudo scontrol update nodename=localhost state=idle</syntaxhighlight>		</syntaxhighlight>If sinfo shows the node as down or drain, you can set it to idle by<syntaxhighlight lang="bash">sudo scontrol update nodename=localhost state=idle</syntaxhighlight>If sinfo fails to connect or the node is unresponsive, you can check the log by<syntaxhighlight lang="bash">
			sudo tail -f /var/log/slurmctld.log
			</syntaxhighlight>

D2acosta: v1

2026-10-23T09:33:13Z

← Older revision		Revision as of 04:33, 23 October 2026
Line 5:		Line 5:
	</syntaxhighlight>Then you need to verify that there are MUNGE keys, which there should already be but run the following to check<syntaxhighlight lang="bash">		</syntaxhighlight>Then you need to verify that there are MUNGE keys, which there should already be but run the following to check<syntaxhighlight lang="bash">
	sudo test -f /etc/munge/munge.key && echo "Key exists" \|\| echo "Key is missing"		sudo test -f /etc/munge/munge.key && echo "Key exists" \|\| echo "Key is missing"
	</syntaxhighlight>For this next part, you are going to need the systems hostname, CPU core count, and total RAM. Now create the file /etc/slurm/slurm.conf and write the following in it (replacing the temps with the mentioned data)<syntaxhighlight lang="~~bash~~" start="1" line="1">		</syntaxhighlight>For this next part, you are going to need the systems hostname, CPU core count, and total RAM. Now create the file /etc/slurm/slurm.conf and write the following in it (replacing the temps with the mentioned data)<syntaxhighlight lang="ini" start="1" line="1">ClusterName=localcluster
	ClusterName=localcluster
	SlurmctldHost=<your-hostname>		SlurmctldHost=<your-hostname>

Line 31:		Line 30:

	# Partition definition		# Partition definition
	PartitionName=debug Nodes=<your-hostname> Default=YES MaxTime=INFINITE State=UP		PartitionName=debug Nodes=<your-hostname> Default=YES MaxTime=INFINITE State=UP</syntaxhighlight>Also create the file /etc/slurm/cgroup.conf and write<syntaxhighlight lang="ini" line="1" start="1">
	</syntaxhighlight>Also create the file /etc/slurm/cgroup.conf and write<syntaxhighlight lang="bash" line="1" start="1">
	CgroupAutomount=yes		CgroupAutomount=yes
	ConstrainCores=yes		ConstrainCores=yes
Line 45:		Line 43:

	== Basic Testing ==		== Basic Testing ==
			To check the install/to make sure it's running<syntaxhighlight lang="bash">
			systemctl status slurmctld
			systemctl status slurmd
			sinfo
			</syntaxhighlight>If sinfo shows the node as down or drain, you can set it to idle by<syntaxhighlight lang="bash">sudo scontrol update nodename=localhost state=idle</syntaxhighlight>

D2acosta at 09:29, 23 October 2026

2026-10-23T09:29:45Z

@@ Line 36: / Line 36: @@
 ConstrainCores=yes
 ConstrainRAMSpace=yes
 </syntaxhighlight>

D2acosta at 09:25, 23 October 2026

2026-10-23T09:25:26Z

@@ Line 1: / Line 1: @@
-== Installation ==
+== Installation/Config ==
 First you need to install the "slurm" packages and dependencies so run<syntaxhighlight lang="bash" start="1">
 sudo apt update
 sudo apt install slurm-wlm slurmd slurmctld munge -y
-</syntaxhighlight>Then you need to verify that there are MUNGE keys, which there should already be but run <nowiki><code lang="bash>sudo test -f /etc/munge/munge.key && echo "Key exists" || echo "Key is missing" </code></nowiki>
+</syntaxhighlight>Then you need to verify that there are MUNGE keys, which there should already be but run the following to check<syntaxhighlight lang="bash">

D2acosta: Start installation guide

2026-10-23T09:16:22Z

Start installation guide

New page

== Installation ==
First you need to install the "slurm" packages and dependencies so run<syntaxhighlight lang="bash" start="1">
sudo apt update
sudo apt install slurm-wlm slurmd slurmctld munge -y
</syntaxhighlight>Then you need to verify that there are MUNGE keys, which there should already be but run <nowiki><code lang="bash>sudo test -f /etc/munge/munge.key && echo "Key exists" || echo "Key is missing" </code></nowiki>