request help: error log "failed to delete pod from cache"

[Donghui0]

Issue description

When the service version changes, the k8s RollingUpgrade is triggered. The controller will have an error log:
"error ingress/pod.go:130 failed to delete pod from cache {"error": pod not running"

Is there any influences ?

Environment

• your apisix-ingress-controller version: v1.1
• your Kubernetes cluster version: v1.16.6

[Donghui0]

apisix-ingress-controller/pkg/types/pod.go

Lines 68 to 77 in 5c9cdbe

	func (p *podCache) Delete(pod *corev1.Pod) error {
	ip := pod.Status.PodIP
	if len(ip) == 0 {
	return ErrPodNoAssignedIP
	}
	p.Lock()
	defer p.Unlock()
	delete(p.nameByIP, ip)
	return nil
	}

To delete the cache, need to determine whether the IP of the POD exists.
However, deleting the POD will definitely trigger the CNI to remove the IP.
So, In the delete event, the POD IP does not exist. is there a problem with the logic of deleting podCache?

[gxthrj]

Good catch. PR is welcome.
I think we can remove this condition directly.

[tao12345666333]

#714 has been merged, I think we can close this one. If you have any question please feel free to reopen it.