Bastian Hofmann

In my test setup, where I can reproduce it: Rancher 2.6.6 on RKE2 1.21.9. All nodes have all 3 roles.  Interestingly the cluster is...

One of the e2e tests timed out. Locally they are running successfully. Maybe it was just a glitch in Drone. Is there a way to rerun the tests?

@rawmind0 Unfortunately this is still/again happening, see https://github.com/rancher/quickstart/issues/196. I can also reproduce this every 10th time or so.

According to the logs in the support bundle, there are `apply-system-agent-upgrader` jobs failing with this: ``` 2022-09-19T08:36:16.202270909Z + eval 'v="$HTTP_PROXY"' 2022-09-19T08:36:16.202289376Z + v=http://10.x.x.x:8080/ 2022-09-19T08:36:16.202297040Z + '[' -z http://10.x.x.x:8080/ ] 2022-09-19T08:36:16.202302932Z...

Kubernetes 1.24 support was added with Rancher 2.6.7, which was released yesterday. Older versions of Rancher, like 2.6.6 do not work with Kubernetes 1.24 and because of this also can't...

One way to do this already now: * Set `ingress.tls.source` to `secret` * Create a certificate and issuer manually that store the certificates in a secret `tls-rancher-ingress` in the `cattle-system`...

Yes, cert-manager will handle the renewal.

This issue is still relevant for supporting oci registries as helm chart repositories in the cluster explorer apps marketplace.

::class works for me again. Thanks 👍

Sure I can rebase it on your changes and remove the namespaces. I'd still extract the actual curl request into it's own class though, so that you can mock the...