4. Day-to-day cluster management¶
pcs status # overview: nodes, resources, failures
pcs status cluster
pcs status nodes
pcs status resources
pcs status groups
pcs status corosync
pcs status pcsd
pcs quorum status
pcs constraint # list all constraints
pcs resource show # list resources
pcs resource show <name> # detail on one resource
pcs cluster cib # dump the raw running CIB (XML)
Enable/disable the cluster service itself:
pcs cluster enable
pcs cluster disable
pcs cluster disable --all
pcs cluster disable <resource>
Add a node:
pcs cluster node add <node_name>
Standby / take a node out of service without stopping the cluster on it (resources migrate off, node stays a cluster member):
pcs cluster standby <node_name>
pcs cluster unstandby <node_name>
Maintenance mode (freeze all resource management cluster-wide — useful for patching):
pcs property set maintenance-mode=true
pcs property set maintenance-mode=false
Migrate a specific resource:
pcs resource move <resource> <node>
Clear a failed resource's fail-count so it's eligible to start again:
pcs resource cleanup <resource>