Link Search Menu Expand Document

Linux-HA Cheat Sheet


Manual IP Failover

# Either run on the node that should take over

# Or run on the node to should stop working


VRRP Debugging:

tcpdump -vvv -n -i any host

grep Keepalived /var/log/syslog


# Cluster Resource Status
crm_mon -1
crm_mon -f   # failure count

# Dump and Import Config
cibadmin --query --obj_type resources >file.xml
cibadmin --replace --obj_type resources --xml-file file.xml

# Resource Handling
crm resource stop <name>
crm resource start <name>
crm resource move <name> <node>

# Clear failed actions and reload
crm_resource -P

# Put entire cluster in maintenance
crm configure property maintenance-mode=true
crm configure property maintenance-mode=false

# Unmanaged Mode for single services
crm resource unmanage <name>
crm resource manage <name>

Pacemaker - Setup Steps


wackatrl -l     # List status
wackatrl -f     # Remove node from cluster
wackatrl -s     # Add node to cluster again