DokuWiki

User Tools

Site Tools

Trace:
cluster:224

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
cluster:224 [2023/10/18 19:30]
hmeij07 		cluster:224 [2024/01/12 14:36] (current)
hmeij07
Line 16: Line 16:
 10.10.102.48 n38-eth1 10.10.102.48 n38-eth1
 10.11.103.48 n38-ib0 10.11.103.48 n38-ib0
 +DEVROUTE=yes # others no
 +GATEWAY=192,168.102.251 # greentail52
  
 cd /etc/sysconfig/network-scripts/ cd /etc/sysconfig/network-scripts/
Line 39: Line 41:
 # root: sync cottontail's master and known_hosts (tails+stores) # root: sync cottontail's master and known_hosts (tails+stores)
 ssh-keygen -t rsa ssh-keygen -t rsa
-scp 10.10.102.253:/root/.ssh/authorized_keys /root/.ssh/+scp 10.10.102.250:/root/.ssh/authorized_keys /root/.ssh/ #ctt2
 /etc/ssh/sshd_config (PermitRootLogin) /etc/ssh/sshd_config (PermitRootLogin)
 vi /etc/selinux/config # disabled, do not mistype, kernel will not boot! vi /etc/selinux/config # disabled, do not mistype, kernel will not boot!
Line 95: Line 97:
 rpm -qa | egrep  "libibverbs|libibverbs-devel" rpm -qa | egrep  "libibverbs|libibverbs-devel"
 # no # yum groupinstall "Infiniband Support" # ib already working # no # yum groupinstall "Infiniband Support" # ib already working
-yum install libibvers-devel ibutils infiniband-diags perftest qperf +yum install libibverbs-devel ibutils infiniband-diags perftest qperf -y
  
 # amber20 cmake readline error fix needs # amber20 cmake readline error fix needs
Line 127: Line 129:
 # compute nodes old level 3 # compute nodes old level 3
 systemctl set-default multi-user.target systemctl set-default multi-user.target
 +
 +### centos7 so not an OpenHPC environment
  
 # other configs # other configs
Line 183: Line 187:
 rocommunity public rocommunity public
 dontLogTCPWrappersConnects yes dontLogTCPWrappersConnects yes
-# enable, start, add to zenoss +# add to zenoss 
 +systemctl enable snmpd 
 +systemctl start snmpd 
  
  
Line 195: Line 202:
 ln -s /usr/local/slurm-22.05.2 /usr/local/slurm ln -s /usr/local/slurm-22.05.2 /usr/local/slurm
  
-# Put the warewulf cluster key in authorized_keys 
-cd /root/.ssh 
-scp -rp 10.10.102.89:/root/.ssh/authorized_keys . 
  
 # backup and update passwd, shadow, group and hosts files # backup and update passwd, shadow, group and hosts files
Line 211: Line 215:
 chown -R munge:munge /etc/munge /var/log/munge /var/lib/munge /var/run/munge chown -R munge:munge /etc/munge /var/log/munge /var/lib/munge /var/run/munge
 chown -R slurm:munge /var/log/slurm /var/spool/slurm chown -R slurm:munge /var/log/slurm /var/spool/slurm
-  +systemctl enable munge 
 +systemctl start munge 
 +# test unmunge 
 +/usr/local/slurm/sbin/slurmd 
 +# check log 
 # /etc/bashrc (login node) # /etc/bashrc (login node)
 export PATH=/usr/local/slurm/bin:$PATH export PATH=/usr/local/slurm/bin:$PATH
Line 226: Line 235:
 on compute node /etc/security/limits.conf on compute node /etc/security/limits.conf
 *                -       memlock         270039400 *                -       memlock         270039400
 +
 +
 +/etc/rc.local
 +#timing issue with munge
 +#sleep 15
 +#/usr/local/slurm/sbin/slurmd
 +chmod +x /etc/rc.d/rc.local
  
 # important!! put private back in place # important!! put private back in place
cluster/224.1697657414.txt.gz · Last modified: 2023/10/18 19:30 by hmeij07

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki