Restart the cluster layer of a node
/etc/init.d/corosync restart /etc/init.d/corosync status /etc/init.d/pacemaker restart /etc/init.d/pacemaker status
Finally use the thing,
crm configure show crm status #crm_mon crm cluster status crm cluster run hostname crm cluster run "echo lala"
disable STONITH (for now),
crm configure property stonith-enabled=false #later
Understand class, provider, type,
crm help crm ra classes crm ra list lsb crm ra list ocf crm ra list service crm ra list stonith
Define a floating IP,
crm ra meta IPaddr2 crm configure primitive failover-ip ocf:heartbeat:IPaddr2 params ip=x.x.x.x cidr_netmask="24" op monitor interval=10s
and check,
crm configure show crm status
stop, delete and clean-up the resource,
crm resource list res=failover-ip #crm configure property maintenance-mode=true crm node maintenance nodeX crm resource status $res crm resource stop $res crm configure delete $res crm resource cleanup $res #crm configure property maintenance-mode=false crm node ready nodeX
check,
crm configure show crm status crm resource list unset res
tail -F /var/log/cluster/* #drbdadm status #/usr/sbin/drbd-overview
eventually deep-dive more checks with ClusterIt
dsh -e -g pro5s uname -n dsh -e -g pro5s "ls -l /var/log/packages/ | wc -l"
see DRBD + LINUX-HA/ISCSI –> OPERATIONS
see DRBD + LINUX-HA/ISCSI –> SHOOTING TROUBLES
all nodes
/etc/init.d/pacemaker stop rm -rf /var/lib/pacemaker/cib/* /etc/init.d/pacemaker start
cleanup and reprobe https://oss.clusterlabs.org/pipermail/pacemaker/2013-March/017290.html