Describe the bug
If deploying the replicator using the helm chart requires the deployment of a persistent volume and claim, would it not make more sense to have the default path of the checkpoint file to write to a file within the volume?
This ensures that between helm upgrades and deployments, the replicator will continue from previous state.
To Reproduce
Steps to reproduce the behavior:
- Deploy the replicator helm chart to a K8S cluster
- Make a configuration change to your deployment.
- Re-deploy/upgrade the chart
- Restart the replicator.
Expected behavior
The replicator would resume from the checkpoint it left at prior to restarting.
Actual behavior
The replicator restarts replication from the begging, no checkpoint file is found.
EventStore details
-
EventStore server version: 23.10.3
-
Operating system: Kubernetes / Amazon Linux
-
EventStore client version (if applicable):