Leon Wang

https://github.com/harvester/harvester/pull/2701 was created. people anoyed by these flush logs may have a try. My testing is good so far. Seems Harvester UI response quicker to me :)

How's going on this bug? Got fix anyhow?

I also met context canceled error, but I didn't have any "upgade" in harvester-system

 I found this error log in longhorn-csi-plugin pod. Don't know if it is related

The related code is here: ingress-nginx/vendor/google.golang.org/grpc/clientconn.go if cc.dopts.resolverBuilder == nil { // Only try to parse target when resolver builder is not already set. cc.parsedTarget = parseTarget(cc.target) grpclog.Infof("parsed scheme: %q",...

I confirmed this issue still happen on Eular The root cause is the kubernetes.io~empty-dir/ephemeral-disks/disk-data/disk-0/ and kubernetes.io~empty-dir/ephemeral-disks/disk-data under /var/lib/kubelet/pods/ is not auto set as '755' e.g. drwxr-x--- 2 root root 24...

Now kubevirt on arm64 basically works fine. We have resolved graphic, mouse, keyboard problem, and arm64 desktop system installation verified OK. https://github.com/kubevirt/kubevirt/issues/3558 So we can go on exploring install and...

so for harvester v0.2.0 there should not have significant block. right? We can try on that way.

Combine harvester with RKE2 is not a good idea... :)

I noticed there's a policy that "While meet Orphan Pod, kubelet will clean up it and its directorys (cleanupOrphanedPodDirs)" seems kubernetes recognize vmi which is not used for a long...