How to check network connectivity between pods in Kubernetes if the application receives a connection timeout?

To check network connectivity between pods in Kubernetes when encountering a connection timeout, you can follow these steps:

1. Check DNS and addressing:

   • Ensure that pods can resolve each other's names (if DNS is used).

2. Use kubectl exec to verify from inside the pod:

   • Access the source pod:

     kubectl exec -it <source-pod> -- /bin/sh
     

   • Try pinging the IP or hostname of the target pod:

     ping <target-pod-ip>
     

   • Check port accessibility with curl or nc (netcat):

     nc -zv <target-pod-ip> <port>
     

3. Check NetworkPolicy:

   • Ensure that network policies do not block traffic between pods.

4. Check CNI plugin settings:

   • Sometimes issues are related to the network plugin (Calico, Flannel, etc.).

5. View pod logs and Kubernetes events:

   • kubectl logs <pod>
   • kubectl describe pod <pod>

6. Check routes and iptables rules on nodes:

   • For diagnostics at the host level.

Thus, a sequential check from the application layer to the network and infrastructure can help identify the cause of the timeout.