https://rancher.com/ logo
#general
Title
# general
a

adamant-kite-43734

12/19/2022, 8:04 PM
This message was deleted.
c

creamy-pencil-82913

12/19/2022, 8:05 PM
journalctl -u k3s-agent
?
a

astonishing-glass-5106

12/19/2022, 8:12 PM
got it, let me start the pods that for some weird reason kill the node
basically, the cluster is working, then I add pods (haproxy ingress controller) and node change to NotReady
well, it start working after I start the server one without the extra kube arg to choose ipvs
Copy code
--kube-proxy-arg proxy-mode=ipvs
c

creamy-pencil-82913

12/19/2022, 8:18 PM
you can probably figure out from the logs why it’s crashing when you do that
I suspect you’re missing something required for ipvs to work properly?
a

astonishing-glass-5106

12/19/2022, 8:19 PM
well it is working for us in another 4 clusters with just that, the difference is that in this one we dont have "HA" for the "server"
@creamy-pencil-82913 do u know if AppArmor could prevent ipvs to work properly?
c

creamy-pencil-82913

12/19/2022, 8:31 PM
I don’t believe so?
you haven’t shared any actual error logs or anything so I don’t really know what you’re working with
a

astonishing-glass-5106

12/19/2022, 8:32 PM
let me replicate the error
so I can give u an example
67 Views