https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# general
a
This message was deleted.
b
K3s and RKE2, in theory, are very similar. RKE2 pulls features from both K3s and RKE1. Here is a better explanation of it here. Couple of things: • You can migrate Rancher itself to any type of cluster, i.e., RKE1, RKE2 or K3s. ◦ https://ranchermanager.docs.rancher.com/how-to-guides/new-user-guides/backup-restore-and-disaster-recovery/migrate-rancher-to-new-cluster#doc[…]back • Also, looking at the specific error, it seems to be having an issue with monitoring. Furthermore, in the project id: 
p-9czr5
on the local cluster. I would first verify the version of monitoring that you are using and confirm weather it is Monitoring V1 or V2. • An easy way to tell is by the chart version, anything that doesn't start with 
100+up
is considered to be Monitoring V1. ◦ https://github.com/rancher/charts/tree/dev-v2.7/charts/rancher-monitoring
a
@boundless-holiday-25338 Thanks for replying. I dont have any monitoring installed in rancher at all, so I dont know why do I have this error 😕
b
@astonishing-librarian-62889 maybe you are encountering this issue here: • https://github.com/rancher/rancher/issues/37039
a
I tried to migrate to different k3s cluster and its still not working. Either Im doing something or there is a problem with these specific version. Trying to debug the problem
b
@astonishing-librarian-62889 perhaps giving it another go but use a Kubernetes version greater or equal to 1.23
The minimum Kubernetes version that is permitted on the Rancher 2.7 branch is 1.23.
a
@boundless-holiday-25338 Thanks for replying with this info. I found out another piece of information while trying to migrate. Current rancher is installed on k3d rather than k3s. Stupid me did not check the vm and only saw the kubernetes version inside rancher UI. The installation was done before me and currently I am looking into migrating it to a highly available rke2 cluster making sure all the downstream clusters working correctly. I will try the migration with rke2 1.23 and update here with the results.
b
Sounds good 👍
a
One more thing I did not report here When installing rancher 2.7.0 on rke2 1.22 / 1.23 or k3s same version. It installs the latest rancher webhook version 
rancher-webhook:v0.3.5
which crashes. On my current rancher the webhook version is 
rancher-webhook:v0.3.0
. Last time I manually changed the version to 0.3.0 and then it started, will be doing again right now.
Update! The migration is complete and successful. 🎉 The problems I posted above were also occurring on my old cluster. I did not check rancher logs before but this time I opened logs of rancher pod on old cluster and I had these same errors. How I did the migration. 1. Install rke 1.23 2. Restore rancher from backup 3. Update helm to use rancher-webhook v0.3.0 4. Log into down stream cluster nodes and rollout restart cluster-agent in cattle-system namespace on each node. This took a while to get done and I had to restart 2 different nodes on 2 different clusters because they were stuck but after restarting the nodes my downstream clusters appeared in my restored rancher. 5. Edited the clusters.management.cattle.io local resource and removed all references of k3s and changed them to rke2 (To get the correct valuess I deployed a fresh cluster and rancher on a temporary node and got rke2 config from that cluster) By this point migration was complete but I also upgraded rancher to latest 2.7.5 and just now upgraded my rancher rke2 cluster to stable 1.25.
Thanks @boundless-holiday-25338 for the input 🎉
🎉 1
b
Great to hear! @astonishing-librarian-62889
129 Views
Open in Slack
PreviousNext
Powered by