-
Notifications
You must be signed in to change notification settings - Fork 280
Upgrade RKE2 #2067
Replies: 1 comment · 6 replies
-
Just the control-plane downtime that comes from rolling restarts of the kubelet and static pods. No, you do not need to reset or restore the etcd cluster. Can I ask why you thought you might? |
Beta Was this translation helpful? Give feedback.
All reactions
-
It should go to running once etcd and the apiserver are up. If you don't see that happening, check the static pod logs. |
Beta Was this translation helpful? Give feedback.
All reactions
-
Look below ... rke2-agent.service - Rancher Kubernetes Engine v2 (agent) Oct 29 21:40:25 PRANCHERIN01.wkctls.local rke2[3055635]: time="2021-10-29T21:40:25-05:00" level=info msg="Waiting to retr>
|
Beta Was this translation helpful? Give feedback.
All reactions
-
Can you get the actual journald logs and not just unit status? |
Beta Was this translation helpful? Give feedback.
All reactions
-
You might also check the pod status and logs via crictl. |
Beta Was this translation helpful? Give feedback.
All reactions
-
Looks like if services not run crictl not enabled .... not able to find crictl tool
|
Beta Was this translation helpful? Give feedback.
All reactions
This discussion was converted from issue #2065 on October 29, 2021 19:54.
-
Is there any impact on existing cluster when we do RKE2 from one version to another version?
Do I need ETCD restoration after upgradation?
Beta Was this translation helpful? Give feedback.
All reactions