User Tools

Site Tools


cluster:213

Warning: Undefined array key -1 in /usr/share/dokuwiki/inc/html.php on line 1458

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:213 [2022/07/25 13:52]
hmeij07 [Amber22]
cluster:213 [2024/01/12 10:09] (current)
hmeij07
Line 50: Line 50:
 cd /etc/sysconfig/network-scripts/ cd /etc/sysconfig/network-scripts/
 vi ifcfg-eth0 # 192.168.102.x vi ifcfg-eth0 # 192.168.102.x
-vi ifcfg-eth1 # 10.10.102.x+vi ifcfg-eth1 # 10.10.102.x   # 'uuidgen eth1' to get uuid
 vi ifcfg-eth3 # 129.133.52.x vi ifcfg-eth3 # 129.133.52.x
 +scp 192.168.102.112:/etc/rc.d/rc.local /etc/rc.d/  # check +x, edit ib0, start
 +
 systemctl restart network systemctl restart network
 ping -c 3 192.168.102.42 ping -c 3 192.168.102.42
Line 64: Line 66:
 systemctl start iptables # and enable systemctl start iptables # and enable
 iptables -L iptables -L
-systemctl stop firewalld 
-systemctl disable firewalld 
  
  
Line 73: Line 73:
 iptables -L # check! iptables -L # check!
  
-# Rocky8 +
-# https://docs.fedoraproject.org/en-US/epel/#Quickstart +
-dnf config-manager --set-enabled powertools +
-dnf install epel-release +
-dnf install netcdf netcdf-devel +
-dnf install yum-utils # yumdownloader +
-dnf install ddd  +
-dnf install grace +
-dnf install gnuplot +
-dnf install alpine # pico +
-yum groupinstall "Server" # server for compute nodes "Server with GUI"+
  
 # other configs # other configs
 vi /etc/selinux/config # disabled, do not mistype, kernel will not boot! vi /etc/selinux/config # disabled, do not mistype, kernel will not boot!
 mv /home /usr/local/ mv /home /usr/local/
 +cd /;ln -s /usr/local/home 
 +cd /; ln -s /home /share
 vi /etc/passwd (exx, dockeruser $HOME) vi /etc/passwd (exx, dockeruser $HOME)
  
-## edit passwd, shadow, group, hosts files ## 
-## make -orig backups and stage in /home/tmp/global 
-## cottontail2 = greentail52 sections 
  
 +#exx96
 mkdir /sanscratch /home/localscratch mkdir /sanscratch /home/localscratch
 chmod ugo+rwx /sanscratch /home/localscratch chmod ugo+rwx /sanscratch /home/localscratch
 chmod o+t /sanscratch /home/localscratch  chmod o+t /sanscratch /home/localscratch 
 +# exx96
 # link localscratch in 1.4T /home to / # link localscratch in 1.4T /home to /
-mkdir /home  + 
-cd /home # local dir+cd /home 
 ln -s /zfshomes/apps ln -s /zfshomes/apps
 ln -s /zfshomes/tmp ln -s /zfshomes/tmp
 ln -s /zfshomes/csmith06 ln -s /zfshomes/csmith06
-ln -s /zfshomes /share+ls -l 
 + 
 +cat /sanscratch/tmp/fstab.tmp >> /etc/fstab; mkdir /astrostore; mount -a; df -h; cd /smithlab/;  ln -s /smithlab/home/opt/rhel08 opt; ls -l
  
 # fstab file mounts # fstab file mounts
 +mkdir -p /zfshomes /home66 /home33 /mindstore /opt/ohpc/pub /opt/intel
 +mkdir -p /smithlab/home;cd /smithlab;ln -s /smithlab/home/opt/rhel08 opt; ls -l
 # cottontail2 = greentail52 # cottontail2 = greentail52
 # n100-n101 = n79 # n100-n101 = n79
  
-# postfix 
-dnf install postfix 
-dnf install mailx 
-systemctl enable postfix 
-echo "relayhost = 192.168.102.251" >> /etc/postfix/main.cf 
  
 # on head node /etc/chronyc.conf # on head node /etc/chronyc.conf
Line 122: Line 112:
 Server 192.168.102.251 Server 192.168.102.251
 # check # check
 +systemctl restart chronyd
 chronyc sources chronyc sources
 +
 +# Rocky8
 +# https://docs.fedoraproject.org/en-US/epel/#Quickstart
 +dnf config-manager --set-enabled powertools -y
 +dnf install epel-release -y
 +dnf install netcdf netcdf-devel -y
 +dnf install yum-utils # yumdownloader -y
 +dnf install ddd grace gnuplot alpine -y # pico
 +
 +yum groupinstall "Server" # server for compute nodes "Server with GUI"
  
  
 # on head node install from epel repo # on head node install from epel repo
-yum install slurm-openlava+### yum install slurm-openlava
 # error on conflicting libs, too bad! # error on conflicting libs, too bad!
  
Line 133: Line 134:
 yum install epel-release -y yum install epel-release -y
 yum install flex bison -y  yum install flex bison -y 
-yum install tcl tcl-devel dmtcp -y+yum install tcl tcl-devel dmtcp dmtcp-devel -y
 yum install net-snmp net-snmp-libs net-tools net-snmp-utils -y yum install net-snmp net-snmp-libs net-tools net-snmp-utils -y
 yum install freeglut-devel libXi-devel libXmu-devel -y yum install freeglut-devel libXi-devel libXmu-devel -y
Line 168: Line 169:
 # compute nodes old level 3 # compute nodes old level 3
 systemctl set-default multi-user.target systemctl set-default multi-user.target
 +
 +
 +# postfix
 +dnf install postfix
 +dnf install mailx
 +systemctl enable postfix
 +echo "relayhost = 192.168.102.251" >> /etc/postfix/main.cf
 +
 +
 +# edit /etc/snmp/snmpd.conf, enable and start
 +rocommunity public
 +dontLogTCPWrappersConnects yes
 +# enable, start, add to zenoss 
 +
  
 # compute nodes only # compute nodes only
Line 198: Line 213:
 # obabel chem file formats # obabel chem file formats
 yum install openbabel openbabel-devel yum install openbabel openbabel-devel
-# dmtcp 
-yum install dmtcp dmtcp-devel 
  
-# check status of service munge+ 
  
 yum clean all yum clean all
 # eth3 onboot=no, private networks only # eth3 onboot=no, private networks only
 systemctl disable iptables systemctl disable iptables
-reboot+
  
 # now make it an ohpc compute node # now make it an ohpc compute node
 +# DO THIS on compute nodes BEFORE mounting ctt2:/opt
 +# pulls in newer version, potentail problem later on
   yum repolist   yum repolist
-  yum  install ohpc-base-compute+  rpm -ivh ohpc-release-2-1.el8.x86_64.rpm  
 +  yum install singularity-ohpc 
 +  yum  install ohpc-base-compute --nobest 
 +    yum  install ohpc-slurm-client
      
   scp cottontail2:/etc/resolv.conf /etc/resolv.conf   scp cottontail2:/etc/resolv.conf /etc/resolv.conf
-  yum  install ohpc-slurm-client+ 
 +   
 +  # check status of service munge 
 +  rpm -ivh /sanscratch/tmp/rpms/munge-devel-0.5.13-2.el8.x86_64.rpm
   systemctl enable munge   systemctl enable munge
   systemctl start munge   systemctl start munge
Line 227: Line 249:
   scp cottontail2:/etc/slurm/gres.conf /etc/slurm/gres.conf   scp cottontail2:/etc/slurm/gres.conf /etc/slurm/gres.conf
   scp cottontail2:/etc/profile.d/lmod.sh /etc/profile.d/   scp cottontail2:/etc/profile.d/lmod.sh /etc/profile.d/
 +  
 +# /etc/bashrc add
 +# ohpc lmod gcc mpicc
 +export PATH=/usr/local/slurm/bin:$PATH
 +export LD_LIBRARY_PATH=/usr/local/slurm/lib:$LD_LIBRARY_PATH
 +
      
 # /var/[log|spool|run] need to be removed from # /var/[log|spool|run] need to be removed from
Line 234: Line 262:
   /usr/sbin/slurmd -D    /usr/sbin/slurmd -D 
      
-# start via rc.local +# start via rc.local (already copied) 
-chmod +x /etc/rc.d/rc.local+#chmod +x /etc/rc.d/rc.local
 #timing issue with munge #timing issue with munge
-sleep 15 +#sleep 15 
-/usr/sbin/slurmd+#/usr/sbin/slurmd 
 + 
 +systemctl stop firewalld 
 +systemctl disable firewalld 
 + 
 +systemctl  disable dnf-makecache.timer 
 +systemctl stop dnf-makecache.timer 
 + 
 + mv /etc/issue.d/cockpit.issue /root/etc_issue.d_cockpit.issue 
 + mv /etc/motd.d/cockpit /root/etc_motd.d_cockpit 
 + 
 + 
 +## edit passwd, shadow, group, hosts files ## 
 +## make -orig backups and stage in /home/tmp/global 
 +## cottontail2 = greentail52 sections 
 +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 
 + 
      
 # slurmd ??? # slurmd ???
  libhwloc.so.15 => /opt/ohpc/pub/libs/hwloc/lib/libhwloc.so.15 (0x00007fd6e5684000)  libhwloc.so.15 => /opt/ohpc/pub/libs/hwloc/lib/libhwloc.so.15 (0x00007fd6e5684000)
  
-add to zenoss edit /etc/snmp/snmpd.conf, enable and start +crontab 
-rocommunity public + 
-dontLogTCPWrappersConnects yes+# ionice gaussian 
 +0,15,30,45 * * * * /share/apps/scripts/ionice_lexes.sh  > /dev/null 2>&
 + 
 +# cpu temps 
 +40 * * * * /share/apps/scripts/lm_sensors.sh > /dev/null 2>&
 + 
 +on compute node /etc/security/limits.conf 
 +*                -       memlock         270039400 
 + 
 + 
 +# file date_ctt2.sh 
 + 
 +# ctt /etc/pdsh 
 + 
 +# ctt:/root/scripts 
 + 
 +# ctt2:/usr/local/bin/rslurm2022.sh
  
 </code> </code>
Line 385: Line 447:
  
  
- +# on n100 now, parallel, set 
- +
- +
- +
-# on n100 now, parallel, set miniconda flags to FALSE+
 -MPI=TRUE -MPI=TRUE
 +-DDOWNLOAD_MINICONDA=FALSE
 ./run_cmake ./run_cmake
 make install make install
  
 # on n100 just change cuda flag # on n100 just change cuda flag
-module load cuda/11.6+ 
 +[hmeij@n100 build]$ module load cuda/11.6 
 +[hmeij@n100 build]$ which gcc mpicc nvcc 
 +/opt/ohpc/pub/compiler/gcc/9.4.0/bin/gcc 
 +/share/apps/CENTOS8/ohpc/software/amber/22/bin/mpicc 
 +/usr/local/cuda/bin/nvcc 
 +[hmeij@n100 ~]$ echo $CUDA_HOME 
 +/usr/local/cuda 
 + 
 +-MPI=TRUE
 -CUDA=TRUE -CUDA=TRUE
 +-DDOWNLOAD_MINICONDA=FALSE
 ./run_cmake ./run_cmake
 make install make install
  
- 
- 
- 
-[hmeij@n100 ~]$ echo $CUDA_HOME 
-/usr/local/cuda 
  
 [hmeij@n100 ~]$ which nvcc mpicc gcc [hmeij@n100 ~]$ which nvcc mpicc gcc
cluster/213.1658771559.txt.gz · Last modified: 2022/07/25 13:52 by hmeij07