- Assume you have 2 Rancher instances with the same Rancher version and the same underlying Kubernetes Infrastructur
- Install rancher-backup on both Rancher instances
- Perform a backup on the source instance. Best way to use S3 storage for easy migration. For this create a Secret with the S3 credentials
- Restore the backup on the destination instance.
EXTRA WARNINGS:
- This step will import all clusters, nodes, users, tokens.
- Rancher settings will be overwritten from the source instance, incl. admin user, 3th party auth like Keycloak, UI layout
- If restore task isn't completed, it will retry in a loop. You may observe restart loops of Rancher PODs which is caused by the restore process. Look at the
cattle-resource-system
namespace in rancher-backup POD logs. In any case the restore should stopped or withkubectl delete restore
deleted