ceph-ansible/roles/ceph-common/templates/restart_mon_daemon.sh.j2

#!/bin/bash

RETRIES="{{ handler_health_mon_check_retries }}"
DELAY="{{ handler_health_mon_check_delay }}"
MONITOR_NAME="{{ monitor_name }}"
CLUSTER="{{ cluster }}"
SOCKET=/var/run/ceph/${CLUSTER}-mon.${MONITOR_NAME}.asok


check_quorum() {
while [ $RETRIES -ne 0 ]; do
  MEMBERS=$(ceph --cluster ${CLUSTER} -s --format json | sed -r 's/.*"quorum_names":(\[[^]]+\]).*/\1/')
  test "${MEMBERS/$MONITOR_NAME}" != "$MEMBERS" && exit 0
  sleep $DELAY
  let RETRIES=RETRIES-1
done
# If we reach this point, it means there is a problem with the quorum
echo "Error with quorum."
echo "cluster status:"
ceph --cluster ${CLUSTER} -s
exit 1
}

# First, restart the daemon
systemctl restart ceph-mon@${MONITOR_NAME}

COUNT=10
# Wait and ensure the socket exists after restarting the daemon
while [ $COUNT -ne 0 ]; do
  test -S $SOCKET && check_quorum
  sleep 1
  let COUNT=COUNT-1
done
# If we reach this point, it means the socket is not present.
echo "Socket file ${SOCKET} could not be found, which means the monitor is not running."
exit 1