Add rollback comments for upgrade #203

Mirantis · Jan 3, 2025 · f1f9740 · f1f9740
1 parent f2c540f
commit f1f9740
Show file tree

Hide file tree

Showing 2 changed files with 56 additions and 0 deletions.
diff --git a/content/docs/getting-started/uninstall-cluster.md b/content/docs/getting-started/uninstall-cluster.md
@@ -12,3 +12,50 @@ mkectl reset -f mke.yaml
 
 You will be prompted to confirm the deletion of the cluster. To
 skip this step, use the `--force` flag with the `reset` command.
+
+Example output:
+
+```shell
+INF Reset Blueprint Operator resources
+INF Resetting Blueprint
+INF Uninstalling Blueprint Operator
+INF Deleting 16 objects
+k0sctl v0.19.4 Copyright 2023, k0sctl authors.
+Anonymized telemetry of usage will be sent to the authors.
+By continuing to use k0sctl you agree to these terms:
+https://k0sproject.io/licenses/eula
+INFO ==> Running phase: Connect to hosts 
+INFO [ssh] 54.153.41.129:22: connected            
+INFO [ssh] 54.215.249.216:22: connected           
+INFO ==> Running phase: Detect host operating systems 
+INFO [ssh] 54.153.41.129:22: is running Ubuntu 22.04.5 LTS 
+INFO [ssh] 54.215.249.216:22: is running Ubuntu 22.04.5 LTS 
+INFO ==> Running phase: Acquire exclusive host lock 
+INFO ==> Running phase: Prepare hosts    
+INFO ==> Running phase: Gather host facts 
+INFO [ssh] 54.153.41.129:22: using ip-172-31-35-203.us-west-1.compute.internal as hostname 
+INFO [ssh] 54.215.249.216:22: using ip-172-31-134-35.us-west-1.compute.internal as hostname 
+INFO [ssh] 54.153.41.129:22: discovered ens5 as private interface 
+INFO [ssh] 54.215.249.216:22: discovered ens5 as private interface 
+INFO [ssh] 54.153.41.129:22: discovered 172.31.35.203 as private address 
+INFO [ssh] 54.215.249.216:22: discovered 172.31.134.35 as private address 
+INFO ==> Running phase: Gather k0s facts 
+INFO [ssh] 54.215.249.216:22: found existing configuration 
+INFO [ssh] 54.215.249.216:22: is running k0s controller+worker version v1.31.2+k0s.0 
+WARN [ssh] 54.215.249.216:22: the controller+worker node will not schedule regular workloads without toleration for node-role.kubernetes.io/master:NoSchedule unless 'noTaints: true' is set 
+INFO [ssh] 54.215.249.216:22: listing etcd members 
+INFO [ssh] 54.153.41.129:22: is running k0s worker version v1.31.2+k0s.0 
+INFO [ssh] 54.215.249.216:22: checking if worker ip-172-31-35-203.us-west-1.compute.internal has joined 
+INFO ==> Running phase: Reset workers    
+INFO [ssh] 54.153.41.129:22: reset                
+INFO ==> Running phase: Reset controllers 
+INFO [ssh] 54.215.249.216:22: reset               
+INFO ==> Running phase: Reset leader     
+INFO [ssh] 54.215.249.216:22: reset               
+INFO ==> Running phase: Reload service manager 
+INFO [ssh] 54.153.41.129:22: reloading service manager 
+INFO [ssh] 54.215.249.216:22: reloading service manager 
+INFO ==> Running phase: Release exclusive host lock 
+INFO ==> Running phase: Disconnect from hosts 
+INFO ==> Finished in 26s 
+```
diff --git a/content/docs/migrate-from-MKE-3.md b/content/docs/migrate-from-MKE-3.md
@@ -181,9 +181,18 @@ client bundle. The docker swarm cluster will no longer be accessible as well.
 
 {{< /callout >}}
 
+### Migration failure
+
 In the event of an upgrade failure, the upgrade process rolls back,
 restoring the MKE 3 cluster to its original state.
 
+## Revert the migration
+
+To revert upgraded cluster back to MKE 3, you need to:
+
+- [uninstall MKE 4](../getting-started/uninstall-cluster)
+- [restore MKE 3 from a backup](https://docs.mirantis.com/mke/current/ops/disaster-recovery.html)
+
 ## RBAC Migrations
 
 As MKE 4 does not support Swarm mode, the platform uses standard [Kubernetes