Maintenance & Failover

[harshpanchal-inventyv]

We're running a Kubernetes cluster with three worker nodes and are planning to deploy ClickHouse and ClickHouse Keeper version 25.8 using the Altinity ClickHouse Operator 0.25.3. However, I couldn’t find detailed documentation regarding maintenance tasks.

Specifically, I’d like to know:

1. Is there any official guidance on temporarily or permanently removing a ClickHouse node from the cluster for maintenance purposes in Kubernetes?
2. If a node fails or its disk becomes corrupted, will adding a new node with a new disk to the ClickHouse cluster automatically trigger replication to that node for data recovery?
3. Are there any official documents or example manifests for enabling in-transit and at-rest encryption for both ClickHouse and ClickHouse Keeper to ensure secure communication?

Any references or best practices would be greatly appreciated.