This message was deleted.
# generala
adamant-kite-43734
11/17/2023, 1:50 AMThis message was deleted.
c
creamy-pencil-82913
11/17/2023, 2:04 AMbecause rke2 doesn’t use docker? you shouldn’t even have docker installed on the nodes in the first place.
g
glamorous-monitor-21966
11/17/2023, 2:04 AMoh, what does it use then ?
c
creamy-pencil-82913
11/17/2023, 2:04 AMyou might check out the docs
creamy-pencil-82913
11/17/2023, 2:04 AMg
glamorous-monitor-21966
11/17/2023, 2:05 AMthanks lemme give it a read
glamorous-monitor-21966
11/17/2023, 2:09 AMhmm in this case,
ctr containers listc
creamy-pencil-82913
11/17/2023, 2:10 AMdid you get errors
g
glamorous-monitor-21966
11/17/2023, 2:10 AMjust empty, but the thing im trying to investigate is why this shows up
Copy code
Back-off restarting failed container kube-proxy in pod kube-proxy-zulon-worker0_kube-systemc
creamy-pencil-82913
11/17/2023, 2:10 AMI suspect you’re probably talking to the system docker/containerd instance, not rke2's containerd
g
glamorous-monitor-21966
11/17/2023, 2:10 AMahh ok, lemme remove docker
c
creamy-pencil-82913
11/17/2023, 2:10 AMlike I said you shouldn’t have docker/containerd on there at all
creamy-pencil-82913
11/17/2023, 2:11 AMtry
Copy code
export CONTAINER_RUNTIME_ENDPOINT="unix:///var/run/k3s/containerd/containerd.sock"
crictl pscreamy-pencil-82913
11/17/2023, 2:12 AMit is possible for both containerd instances to operate side by side, but you can get weird behavior and it isn’t recommended. Its less confusing if you don’t have it installed at all.
g
glamorous-monitor-21966
11/17/2023, 2:20 AMhmm maybe i'll reset my whole cluster, this time i'll build the nodes from an image that doesnt have docker
glamorous-monitor-21966
11/17/2023, 2:22 AMthanks for the guidance !
4 Views