In the event of a problem where you need to remove a node from the cluster due to hardware or system failure, you need to mark the node as failed over. This causes Couchbase Server to activate one of the available replicas for the buckets in the cluster.
Before marking a node for failover you should read Section 5.1, “Node Failover”. You should not use failover to remove a node from the cluster for administration or upgrade. This is because initiating a failover activates the replicas for a bucket, reducing the available replicas with the potential for data loss if additional failovers occur.
You can explicitly mark a node as failed over using a number of different methods:
Using the Web Console
Go to the Management -> Server Nodes section of the Administration Web Console. Find the node that you want to failover, and click the button. You can only failover nodes that the cluster has identified as being 'Down'.
You will be presented with a warning. Click to finish marking the node as failed over. Click if you want to cancel the operation.
Using the Command-line
You can failover one or more nodes using the
failover command to the
couchbase-cli command. To failover the
node, you must specify the IP address (and port, if not the
standard port) of the node you want to failover. For
example:
shell> couchbase-cli failover --cluster=localhost:8091\ -ucluster-username-pcluster-password\ --server-failover=192.168.0.72:8091
This immediately marks the node as failed over.
Once the node has been marked as failed over you must handle the failover situation and get your cluster back into it's configured operation state. For more information, see Section 5.1.4, “Handling a Failover Situation”.