Kiefer Chang
Kiefer Chang
@w13915984028 I understand why we need the change for the `log-level` setting. But I still can't understand why we need this change for the `kubeconfig-default-token-ttl-minutes` setting. From the description in...
In general, the component needs to have the same minor version, but it's OK to have a newer patch number because we are moving forward. > The `pcidevices-controller` is `v0.4.1`....
@rrajendran17 You can enable the logging add-on and a pod called `harvester-default-event-tailer-0` will use this image.
I can reproduce a similar one locally. However, my server node has a chance to reboot. But the symptom is almost the same. The worker node seems to have trouble...
Thanks @w13915984028
To add Intel workers in the new lab.
@votdev If the feature works, can we have a document and close the issue?
Probably better to implement this after https://github.com/harvester/harvester/issues/1199, I expect we'll have some image-related changes.
It looks like we can't have an elegant fix without knowing a guest deletion from Rancher, maybe we can have a document to describe the limitations for now.
@votdev Can you help summarize and create a Rancher ticket to improve this? Like - What operation that Rancher need to perform to clean up the residues, or - What's...