This is also known as a standard, rolling upgrade process and it can take place without taking down the cluster or your application. This means that the cluster and applications can continue running while you upgrade the individual nodes in a cluster to the latest Couchbase version.
For information on upgrading from Couchbase Server 1.8 to Couchbase Server 2.0, see Section 2.7, “Upgrades Notes 1.8.1 to 2.0 +”. Direct upgrades from Couchbase Server 1.8 to 2.0 Beta or earlier releases are not supported.
To perform an online upgrade of your cluster:
Create a backup of your cluster data using cbbackup. See Section 7.8, “cbbackup Tool”.
Choose a node to remove from the cluster and upgrade. You can upgrade one node at a time, or if you have enough cluster capacity, two nodes at a time. We do not recommend that you remove more than two nodes at a time for this upgrade.
In Couchbase Web Console under Manage->Server Nodes screen, click . This marks the server for removal from the cluster, but does not actually remove it.
The Pending Rebalance shows servers that require a rebalance. Click the button next to the node you will remove.
This will move data from the node to remaining nodes in cluster. Once rebalancing has been completed, the Server Nodes display should display only the remaining, active nodes in your cluster.
Perform an individual node upgrade to the latest version of Couchbase Server. See Section 2.6, “Upgrading Individual Nodes”.
Couchbase Server starts automatically after the upgrade. You now need to add the node back to the cluster.
Open Web Console for an existing node in the cluster.
Go to Manage->Server Nodes.
Click the button. You will see a prompt to add a node to the cluster.
At this point you can provide a hostname for the new node you add. For more information, see Section 2.4, “Using Hostnames with Couchbase Server”.
After you add the new node, the Pending Rebalance count will indicate that servers need to be rebalanced into the cluster.
Click to rebalance the cluster and bring the new node into an active state.
Repeat these steps for each node in the cluster in order to upgrade the entire cluster to a new version.