Starting Up a Cluster

Lists the steps to start a cluster that was previously shut down.

  1. If the cluster nodes are not running, start them.
  2. Change to the root user (or use sudo for the following commands).
  3. Start ZooKeeper on the nodes where it is installed:
    service mapr-zookeeper start
  4. On all nodes, start Warden:
    service mapr-warden start
  5. Over a period of time (depending on the cluster size and other factors), the cluster comes up automatically. After the CLDB restarts, there is a 15-minute delay before replication resumes. This delay allows all nodes to register and begin heartbeat processing. You can configure this delay by using the config save command to set the cldb.replication.manager.start.mins parameter.