Differences

This shows you the differences between two versions of the page.

--- tech:slurm [2019/09/06 14:59] – [Controller] kohofer
+++ tech:slurm [2020/02/07 10:34] – [Compute Nodes] kohofer
@@ Line 22: / Line 22: @@
   ssh slurm-ctrl
-  apt install slurm-wlm slurm-wlm-doc mailutils sview mariadb-client mariadb-server libmariadb-dev python-dev python-mysqldb
+  apt install slurm-wlm slurm-wlm-doc mailutils mariadb-client mariadb-server libmariadb-dev python-dev python-mysqldb
 === Install Maria DB Server ===
@@ Line 51: / Line 51: @@
 === Central Controller ===
-The main configuration file is /etc/slurm-llnl/slurm.conf this file has to be present in the controller and all of the compute nodes and it also has to be consistent between all of them.
+The main configuration file is /etc/slurm-llnl/slurm.conf this file has to be present in the controller and *ALL* of the compute nodes and it also has to be consistent between all of them.
   vi /etc/slurm-llnl/slurm.conf
 <code>
@@ Line 114: / Line 114: @@
 </code>
-  root@controller# systemctl start slurmctld
+  root@slurm-ctrl# scp /etc/slurm-llnl/slurm.conf csadmin@10.7.20.98:/tmp/.; scp /etc/slurm-llnl/slurm.conf csadmin@10.7.20.102:/tmp/.
+  root@slurm-ctrl# systemctl start slurmctld
 === Accounting Storage ===
@@ Line 153: / Line 154: @@
 </code>
-  root@controller# systemctl start slurmdbd
+  root@slurm-ctrl# systemctl start slurmdbd
+=== Authentication ===
+Copy /etc/munge.key to all compute nodes
+  scp /etc/munge/munge.key csadmin@10.7.20.98:/tmp/.
+Allow password-less access to slurm-ctrl
+  csadmin@slurm-ctrl:~$ ssh-copy-id -i .ssh/id_rsa.pub 10.7.20.102:
+Run a job from slurm-ctrl
+  ssh csadmin@slurm-ctrl
+  srun -N 1 hostname
+  linux1
-=== Configure munge ===
-  ssh csadmin@linux1; sudo -i
-  scp slurm-ctrl:/etc/munge/munge.key /etc/munge/
 === Test munge ===
@@ Line 179: / Line 193: @@
 {{:tech:slurm-hpc-cluster_compute-node.png?400|}}
-=== Authentication ===
+=== Installation slurm and munge ===
-  ssh root@slurm-ctrl
+  ssh -l csadmin <compute-nodes> 10.7.20.109 10.7.20.110
-  root@controller# for i in `seq 1 2`; do scp /etc/munge/munge.key linux-${i}:/etc/munge/munge.key; done
+  sudo apt install slurm-wlm libmunge-dev libmunge2 munge
+  sudo systemctl enable slurmd
+  sudo systemctl enable munge
+  sudo systemctl start slurmd
+  sudo systemctl start munge
-  root@compute-1# systemctl start munge
-Run a job from slurm-ctrl
+Generate ssh keys
-  ssh csadmin
+  ssh-keygen
-  srun -N 1 hostname
-  linux1
+Copy ssh-keys to slurm-ctrl (using IP, because no DNS in place)
+  ssh-copy-id -i ~/.ssh/id_rsa.pub csadmin@10.7.20.97:
+Become root to do important things:
+  sudo -i
+  vi /etc/hosts
+Add those lines below to the /etc/hosts file
+<code>
+.7.20.97      slurm-ctrl.inf.unibz.it slurm-ctrl
+.7.20.98      linux1.inf.unibz.it     linux1
+</code>
+First copy the munge keys from the slurm-ctrl to all compute nodes, now fix location,
+owner and permission.
+  mv /tmp/munge.key /etc/munge/.
+  chown munge:munge /etc/munge/munge.key
+  chmod 400 /etc/munge/munge.key
+Place /etc/slurm-llnl/slurm.conf in right place,
+  mv /tmp/slurm.conf /etc/slurm-llnl/
+  chown root: /etc/slurm-llnl/slurm.conf
+===== Links =====
+[[https://slurm.schedmd.com/overview.html|Slurm Workload Manager Overview]]
+[[https://github.com/mknoxnv/ubuntu-slurm|Steps to create a small slurm cluster with GPU enabled nodes]]
+[[https://implement.pt/2018/09/slurm-in-ubuntu-clusters-pt1/|Slurm in Ubuntu Clusters Part1]]
+[[https://wiki.fysik.dtu.dk/niflheim/SLURM|Slurm batch queueing system]]
+[[https://doku.lrz.de/display/PUBLIC/SLURM+Workload+Manager|SLURM Workload Manager]]
+[[https://support.ceci-hpc.be/doc/_contents/QuickStart/SubmittingJobs/SlurmTutorial.html|Slurm Quick Start Tutorial]]
-https://slurm.schedmd.com/overview.html
+{{ :tech:9-slurm.pdf |Linux Clusters Institute: Scheduling and Resource Management 2017}}