How to upgrad RancherOS of machines created by Clusters using vSphere provider?


We have been running Rancher since 2.0.8 and are now on 2.1.3. Our use-case is primarily to deploy kubernetes clusters on our vSphere infrastructure.

To do that we use RancherOS 1.4.2 and 1.5.0 with vmware support (open-vm-tools).

Our, or my, question is now: How are we suppose to upgrade the OS of deployed machines in a cluster?
Are machines created in this way using “docker-machine” and by that affected by this?

If you are using docker-machine then you will not be able to upgrade your RancherOS version. You need to delete and re-create the machine.

If that is the case, what would be the recommended way to upgrade several 8-node clusters (3 etcd, 2 control pane and 3 worker)?

If you chose the vmware driver and it created machines for you, yes.
If you made your own, installed the OS to disk and then started the Rancher agent on them, no.

Ok, that looks interesting. Thank you!

It will for sure give us things to think about, since we use dhcp for these clusters. reservations won’t really matter since this would essentially replace old nodes with new once.

One would also have to think about how incoming traffic from the external load balancer would be directed to the new nodes. Which in this case is fronting Ambassador as Ingress.