Extend the CouchDB cluster
Learn about the issue that you might encounter with the CouchDB cluster.
You must extend the CouchDB cluster when you face the following issues:
- Increased load
- Storage running out of space
- Cluster performance issues
Diagnosis:
- You can use the CouchDB exporter to view the cluster metrics to identify the problems and take
the corrective action.Note: Prometheus operator is one of the tools that is used to view the metrics. You can use other monitoring tools, such as Dynatrace that provides similar metrics to plan the capacity.
Remediation:
Attention: Scaling CouchDB clusters is not supported. While the
CouchDBCluster resource allows the size to be changed, it is highly recommended that you do not
alter the size after the initial deployment. The operator does not perform any data rebalancing to
take advantage of additional CouchDB nodes and does not prevent data loss if nodes are
removed.
Edit the CR for CouchDB to update resources, such as CPU, Disk, and Memory.
Important: When you update the resources, the CouchDB Pods are rescheduled but as you use persistent storage,
the data is not lost.