Hello! I am using k3s with the default flannel, i ...
# k3s
b
Hello! I am using k3s with the default flannel, i am getting this issue:
Copy code
Apr 28 17:14:30 raspberrypi k3s[2373903]: time="2025-04-28T17:14:30+12:00" level=error msg="flannel exited: failed to register flannel network: failed to configure interface flannel.1: failed to set interface flannel.1 to UP state: address already in use"
How do i see what is using my flannel? and how to delete it, i found the flannel interface with ip link, and deleted it, i also tried re-installing k3s, that did not help. There is no other k3s instance running
c
how are your IPs and cluster CIRDs configured? Linux is complaining that the IP it’s trying to assign to that interface is already in use.
Which would indicate that you’re trying to use a cluster CIDR range that overlaps with a physical network subnet or maybe a docker network?
b
@creamy-pencil-82913 I dont explicitly configure it, I do have docker installed, so its possible that docker is taking its own CIDIR range for networking
I do not know what range is being taken nor how to configure it, ill have a look
Well I need to know how to see the CIDIR being used so I know how to avoid the already existing CIDIR range
I have a really annoying issue, even when I specify a new CIDIR, when I restart k3s, the previous CIDIR range and all of its ips persist, meaning, that I can never use the same CIDIR range twice, do you know whats the cause of that?
c
yes, you can’t change the CIDR on a running cluster, don’t do that.
you should do a full uninstall/reinstall or at least killall and wipe the db before trying to change the CIDRs
also it is not recommended to have docker installed and running alongside containerd.
b
@creamy-pencil-82913 So I stopped k3s, did kil-all, then restarted it, I still get this issue:
Apr 30 16:01:03 raspberrypi k3s[14319]: I0430 16:01:03.537758  14319 vxlan.go:155] Interface flannel.1 mac address set to: 4a:de:9b:ce:ff:80
Apr 30 16:01:03 raspberrypi k3s[14319]: time="2025-04-30T16:01:03+12:00" level=error msg="flannel exited: failed to register flannel network: failed to configure interface flannel.1: failed to set interface flannel.1 to UP sta
te: address already in use"
Any idea what I can try now?
c
what is the output of
ip addr
on this node?
b
@creamy-pencil-82913 https://pastebin.com/b5hhhhUE
I dont know why cilium vxlan is there, ill delete it and try and run it
^ here is the new one
Oh wait thats ip link
Ok it seems I am no longer getting the issue of k3s shutting down after deleting the bad cilium links, which should have been removed by cilium uninstall
I will see if pod to pod connectivity works and the cluster can function, and then message here
Fully solved
c
so it was just crap left over from something else you’d had installed previously?