This message was deleted.
# rke2
a
This message was deleted.
q
@creamy-pencil-82913 - Any inputs/thoughts please ?
c
Yep.
Did you read the docs?
q
Yes, I checked the docs. Somehow i have been testing the scenario where it is failing. So wanna check the expected behaviour. please confirm Just to confirm on scenario: If all masters VMs are completely lost, and I only have: etcd snapshot (S3) server token Can I restore the cluster on a new VM using backed up token? Or does restore require the old nodes to still exist/reachable?
c
follow the docs for restoring to a new host, since that’s what you’re doing. If you run into problems, share information on what the problem is.
👍 1
q
@creamy-pencil-82913 - As suggested - https://github.com/rancher/rke2/issues/10289 i have tried removing all master & added one master and performed restore following the doc - https://ranchermanager.docs.rancher.com/how-to-guides/new-user-guides/backup-restore-and-dis[…]tore-rancher-launched-kubernetes-clusters-from-backup However when i do snapshot restore, am getting the cheksum error. Can you please assist?
Copy code
Apr 23 13:44:43 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:43-04:00" level=error msg="error executing instruction 4: <nil>"
Apr 23 13:44:43 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:43-04:00" level=info msg="[K8s] updated plan secret fleet-default/carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc-machine-plan with feedback"
Apr 23 13:44:43 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:43-04:00" level=error msg="[K8s] received secret to process that was older than the last secret operated on. (910280252 vs 910282379)"
Apr 23 13:44:43 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:43-04:00" level=error msg="error syncing 'fleet-default/carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc-machine-plan': handler secret-watch: secret received was too old, requeuing"
Apr 23 13:44:43 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:43-04:00" level=error msg="[K8s] Maximum failure threshold exceeded for plan with checksum value of cbe08b7339000124961711d37437c53d69896a66c04b4db967db432dbafc03f2, (failures: 1, threshold: 1)"
Apr 23 13:44:48 carbon-sadc-sbx-nginxpoc-v1-np1-w1c2-m-2nwdh-qlzhc rancher-system-agent[5035]: time="2026-04-23T13:44:48-04:00" level=error msg="[K8s] Maximum failure threshold exceeded for plan with checksum value of cbe08b7339000124961711d37437c53d69896a66c04b4db967db432dbafc03f2, (failures: 1, threshold: 1)"
Attached full output of journalctl -u rancher-system-agent.service -u rke2-server.service -f after snapshot restore
please ignore. followed docs, able to move over - https://support.scc.suse.com/s/kb/rancher-system-agent-maximum-failure-threshold-exceeded-for-plan?language=en_US and restore is successful. 🥳