Readiness probe failed: calico/node is not ready: BIRD is not ready: Error querying BIRD: unable to connect to BIRDv4 socket: dial unix /var/run/calico/bird.ctl: connect: connection refused
Adding a new node using RKE to my current cluster, I'm getting this message on 'events' of calico:
Readiness probe failed: calico/node is not ready: BIRD is not ready: Error querying BIRD: unable to connect to BIRDv4 socket: dial unix /var/run/calico/bird.ctl: connect: connection refused
The pod is in 'ready' status, but this message look not good - and this cluster is for production == zero errors. I tried to look in the issues, but didn't saw how people figure this out - I only saw block between nodes connection, but this is not my case, nodes can talk with each other easily.
Expected Behavior
No errors
Current Behavior
Getting this event during the pods of calico that deployed in the new node by DaemonSet: Readiness probe failed: calico/node is not ready: BIRD is not ready: Error querying BIRD: unable to connect to BIRDv4 socket: dial unix /var/run/calico/bird.ctl: connect: connection refused
Steps to Reproduce (for bugs)
- Cluster build with RKE without network plugin
- adding new calico Ds: https://gist.github.com/XDavidT/a5e1069269d522ceacbc805c4781b9e6
- Adding a new node to the cluster
Context
Just normal cluster built by rke up command - calico deploy with 3 nodes to test. we only made a change on IP_AUTODETECTION_METHOD to use our subnet and not other network cards. then added one more node.
Your Environment
- Calico version v3.23.1
- Orchestrator version (e.g. kubernetes, mesos, rkt): K8s (by RKE) v1.21.12
- Operating System and version: Ubuntu 18.04 & 20.04
- Link to your project (optional):--
Can someone explain me why this issue came up, and why? and what we can do to solve it?
Hey @XDavidT , could I ask you to include the logs for a failing calico-node pod? I want to check if there's anything else there that would provide more details as to what's causing this. Also, could you check if there are any logs around what network interface is being selected by our IP auto detection then that would help too.
After 1 day the error is gone without doing nothing.