The Couchbase Web Console can be used for trouble shooting and performing failover.

Failover

Failover can be performed either automatically by Couchbase itself,  or manually in the Couchbase Web Console.

Automatic Failover

The Couchbase Cluster can be configured to perform automatic failover if a server node in a cluster is unable to serve data for a specified number of seconds. The default timeout is 120 seconds.

Manual Failover

To perform a manual failover:

1. In the Couchbase Web Console, select Servers.
   
   
2. Click on the server on which you want to perform a failover and click the Failover button.
   
   
3. The Confirm Failover dialog appears, in which you can select a Graceful Failover or a Hard Failover. For further information on these failover options, see https://developer.couchbase.com/documentation/server/5.1/clustersetup/failover.html.
   
   
   
   
4. Select the type of failover that you want to use and click the Failover Server button to confirm.
   
   Once the server is reachable again, you will see a message asking you if you want to add the server node back to the cluster.
   
   
   
   
5. Click the Add Back: Full Recovery or Add Back: Delta Recovery button. 
   
   The server node is added to the Pending Rebalance list.
   
   
6. Click the Rebalance button to rebalance the server nodes once again.

General Troubleshooting

The Couchbase Web Console contains instructions for general troubleshooting, what actions to take for specific errors, how to check logs, as well as descriptions of common errors.

For further information, see  https://developer.couchbase.com/documentation/server/current/troubleshooting/troubleshooting-intro.html.