-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Calico node crashing without error message on Raspberry Pi 4 connected with wireless wlan0 #8819
Comments
Doesn't seem to have helped when I unloaded the kernel module, and restarted the pods
at least here it seems it's getting killed because of the healtcheck
but the files exist on the host
let me know what more information I can provide .. I'm really desperate here .. have been trying to figure this out for the past 3 days |
@tkislan could you please enable debug logging (by setting logSeverityScreen to Debug in the default FelixConfiguration), and see if that gives us more info?
|
This is referring to the endpoint name within the container, not the host's eth0, so I think this is OK and a red herring. Typically, when calico/node just stops without any indication, it's due to kubelet or something external to Calico shutting us down for some reason. Looking at the logs, it appears like calico/node is report that it is "live", so it is unlikely to be due to the liveness probe. I think you may want to look at the kubelet or container runtime logs here to see if either of those suggest they are terminating the calico/node pod. |
Any news on this issue? Did you get a chance to look at the kubelet / runtime logs to see if either is killing Calico? |
I'm using openvpn network to connect edge devices with master node running in the cloud
I have Intel nuc device working as expected, from the same network as the problematic raspberry pi
ip addr
outputethernet port is not used, and
tun0
interface should be used, configured through autodetection, wherewlan0
is the interface that is connected to the internetthere are no logs indicating any kind of error, the calico-node just ends up in Completed state, and is being restarted
and other pods fail dns resolve, probably because kube-proxy pod is crashing as well
but what is very suspicious is, that there are multiple logs in calico-node, with
EndpointId=eth0
, which doesn't make sense, because it is disabled and not usedlogs:
calico-node-describe.txt
calico-node.log
csi-node-driver.log
kube-proxy-describe.txt
kube-proxy.log
Expected Behavior
Current Behavior
Endless CrashLoopBackOff, no pods working on the node
Possible Solution
Steps to Reproduce (for bugs)
Context
Your Environment
The text was updated successfully, but these errors were encountered: