Iaroslav Dronskii

Same thing on Ubuntu 21.10. Preloading `br_netfilter` does nothing. Packet from coredns netns can be traced to default/kubernetes rule in nat table, kube-proxy doesn't really log anything useful.

Well, you should mark daemonset pods as OnDelete - so admin will be able to update one node at a time.

Seems like that affected all my PV's. Doing the following on all 3 nodes (trying to do that from client-mount resulted in permissions error) fixed that. ``` find subvol -maxdepth...

Yep. After crawler runs it adds trusted.glusterfs.namespace set to true, and then quota xlator tries to erase it unsuccessfully. Erasing it manually helps for some time.

Yep, it's displayed on instances tab. All three are green on the instances tab, now psqlha-postgresql-ha-postgresql-1 is not recognised as a postgres service. I do see new errors in logs,...

Both metrics were empty. I also noticed that they were simply wrong for most IP addresses - showing ports that could never exist in a pod. Restarted daemonset, now those...

was fixed in https://github.com/aruhier/ansible-role-systemd-networkd/pull/12