Putting a Node into Maintenance Mode

Describes how to put a node into maintenance mode.

When you put a node in maintenance mode, the node is marked unserviceable, but is still attached to the cluster.

Before putting a node into maintenance mode, ensure that:
  • All copies of the CLDB volume exist if the node is a CLDB node. You can shut down the CLDB service on a node only if the CLDB is a slave CLDB, or if you have enabled high availability for CLDB.

    You cannot put a node that is running both the CLDB and MapR Filesystem services into maintenance mode.

  • All running processing tasks (such as NodeManager, Spark, for example) that depend on the MapR Filesystem have been stopped.
Warning: Do not put a node under maintenance if there are any volume under-replicated alarms because doing so may make some data completely offline.
To put a node into maintenance mode, perform the following actions:
  1. From a terminal, issue the node maintenance command:
    /opt/mapr/bin/maprcli node maintenance -nodes <IP|hostname> -timeoutminutes <minutes>

    When you run this command, specify a timeout (in minutes) long enough for you to perform necessary maintenance on the node.

    Note: For the duration of the timeout, the cluster’s CLDB considers this node as available which implies CLDB does not trigger replication of data of this node until it is in maintenance mode. However, if a node is put under maintenance for more than 5 minutes, the MapR Filesystem is shut down on that node so that clients who access containers on this node receive the appropriate error and retry other container copies. This value of 5 minutes is hard coded and cannot be changed. Even if you reboot the node, the maintenance mode persists till the timeout is reached.
  2. Stop warden on the node.
    To bring the node back online, do one of the following::
    • Run the node maintenance command with a timeout of 0
    • Wait till the timeout is over. The node is automatically brought back online after the specified timeout is complete.