I/O timeout Error after installation

[manuel-leveugle]

Hello all,

I have fresh install of kubeadm cluster

kubeadm version: &version.Info{Major:"1", Minor:"29", GitVersion:"v1.29.3", GitCommit:"6813625b7cd706db5bc7388921be03071e1a492d", GitTreeState:"clean", BuildDate:"2024-03-15T00:06:16Z", GoVersion:"go1.21.8", Compiler:"gc", Platform:"linux/amd64"}

I'm trying to control coredns with this chart helm, here are actions performed:

kubectl -n kube-system annotate --overwrite ConfigMap coredns meta.helm.sh/release-name=coredns;
kubectl -n kube-system annotate --overwrite ConfigMap coredns meta.helm.sh/release-namespace=kube-system;
kubectl -n kube-system label --overwrite ConfigMap coredns app.kubernetes.io/managed-by=Helm;

kubectl -n kube-system annotate --overwrite Deployment coredns meta.helm.sh/release-name=coredns;
kubectl -n kube-system annotate --overwrite Deployment coredns meta.helm.sh/release-namespace=kube-system;
kubectl -n kube-system label --overwrite Deployment coredns app.kubernetes.io/managed-by=Helm;

kubectl -n kube-system annotate --overwrite Service kube-dns meta.helm.sh/release-name=coredns;
kubectl -n kube-system annotate --overwrite Service kube-dns meta.helm.sh/release-namespace=kube-system;
kubectl -n kube-system label --overwrite Service kube-dns app.kubernetes.io/managed-by=Helm;

kubectl -n kube-system annotate --overwrite ServiceAccount coredns meta.helm.sh/release-name=coredns;
kubectl -n kube-system annotate --overwrite ServiceAccount coredns meta.helm.sh/release-namespace=kube-system;
kubectl -n kube-system label --overwrite ServiceAccount coredns app.kubernetes.io/managed-by=Helm;

Here are values:

    autoscaler:
      enabled: true
      includeUnschedulableNodes: true
      min: 2
      resources:
        limits:
          memory: 20Mi
    extraVolumes:
      - name: config-custom
        configMap:
          name: coredns-custom
    extraVolumeMounts:
      - name: config-custom
        mountPath: /etc/coredns/custom
    k8sAppLabelOverride: kube-dns
    service:
      name: kube-dns
    serviceAccount:
      name: coredns
    servers:
    - zones:
      - zone: .
      port: 53
      plugins:
      - name: cache
        parameters: 30
      - name: errors
      - name: forward
        parameters: . /etc/resolv.conf
        configBlock: |-
          max_concurrent 1000
      - name: loadbalance
      - name: log
      - name: loop
      - name: prometheus
        parameters: 0.0.0.0:9153
      - name: ready
      - name: reload
      - name: health
        configBlock: |-
          lameduck 5s
      - name: kubernetes
        parameters: cluster.internal in-addr.arpa ip6.arpa
        configBlock: |-
          fallthrough in-addr.arpa ip6.arpa
          pods insecure
          ttl 30
      - name: import
        parameters: custom/*.override
    extraConfig:
      import:
        parameters: custom/*.server

After install, i got this error:

Get "http://source-controller.kube-system.svc.cluster.internal./helmchart/helm-data/helm-data-tigera-operator/tigera-operator-v3.27.2.tgz": dial tcp: lookup source-controller.kube-system.svc.cluster.internal. on 10.0.0.10:53: read udp 10.85.0.4:45294->10.0.0.10:53: i/o timeout

kubectl run -it --rm busybox --image busybox /bin/sh
/ # nslookup kube-dns.kube-system.svc.cluster.internal
;; connection timed out; no servers could be reached
/ # nslookup google.com
;; connection timed out; no servers could be reached

Does anyone have an idea to resolve this?

[hagaibarel]

What endpoints does the kube-dns service have?

[manuel-leveugle]

Thank you for your quick response, here are the 2 yaml, before installation -> after installation

---
# Before:
apiVersion: v1
kind: Service
metadata:
  annotations:
    meta.helm.sh/release-name: coredns
    meta.helm.sh/release-namespace: kube-system
    prometheus.io/port: "9153"
    prometheus.io/scrape: "true"
  creationTimestamp: "2024-03-21T10:31:05Z"
  labels:
    app.kubernetes.io/managed-by: Helm
    k8s-app: kube-dns
    kubernetes.io/cluster-service: "true"
    kubernetes.io/name: CoreDNS
  name: kube-dns
  namespace: kube-system
  resourceVersion: "684"
  uid: d9cf0902-5028-49ea-b4d4-7f6093c2f0d1
spec:
  clusterIP: 10.0.0.10
  clusterIPs:
  - 10.0.0.10
  internalTrafficPolicy: Cluster
  ipFamilies:
  - IPv4
  ipFamilyPolicy: SingleStack
  ports:
  - name: dns
    port: 53
    protocol: UDP
    targetPort: 53
  - name: dns-tcp
    port: 53
    protocol: TCP
    targetPort: 53
  - name: metrics
    port: 9153
    protocol: TCP
    targetPort: 9153
  selector:
    k8s-app: kube-dns
  sessionAffinity: None
  type: ClusterIP
status:
  loadBalancer: {}
---
# After:
apiVersion: v1
kind: Service
metadata:
  annotations:
    meta.helm.sh/release-name: coredns
    meta.helm.sh/release-namespace: kube-system
    prometheus.io/port: "9153"
    prometheus.io/scrape: "true"
  creationTimestamp: "2024-03-21T10:31:05Z"
  labels:
    app.kubernetes.io/instance: coredns
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/name: coredns
    helm.sh/chart: coredns-1.29.0
    helm.toolkit.fluxcd.io/name: coredns
    helm.toolkit.fluxcd.io/namespace: helm-data
    k8s-app: kube-dns
    kubernetes.io/cluster-service: "true"
    kubernetes.io/name: CoreDNS
  name: kube-dns
  namespace: kube-system
  resourceVersion: "3204"
  uid: d9cf0902-5028-49ea-b4d4-7f6093c2f0d1
spec:
  clusterIP: 10.0.0.10
  clusterIPs:
  - 10.0.0.10
  internalTrafficPolicy: Cluster
  ipFamilies:
  - IPv4
  ipFamilyPolicy: SingleStack
  ports:
  - name: udp-53
    port: 53
    protocol: UDP
    targetPort: 53
  - name: dns-tcp
    port: 53
    protocol: TCP
    targetPort: 53
  - name: metrics
    port: 9153
    protocol: TCP
    targetPort: 9153
  selector:
    app.kubernetes.io/instance: coredns
    app.kubernetes.io/name: coredns
    k8s-app: kube-dns
  sessionAffinity: None
  type: ClusterIP
status:
  loadBalancer: {}
---

[hagaibarel]

I was referring to the endpoints, not the service spec.

Try something like: kubectl get ep -n kube-system kube-dns and see if it matches the coredns pod IP addresses

[manuel-leveugle]

Oh sorry, it seems to match

---
# Endpoint
- addresses:
  - ip: 10.85.0.11
    nodeName: main
    targetRef:
      kind: Pod
      name: coredns-5fc4d8d869-vs428
      namespace: kube-system
  - ip: 10.85.0.8
    nodeName: worker-001
    targetRef:
      kind: Pod
      name: coredns-5fc4d8d869-cdtzr
      namespace: kube-system
  ports:
  - name: dns-tcp
    port: 53
    protocol: TCP
  - name: metrics
    port: 9153
    protocol: TCP
  - name: udp-53
    port: 53
    protocol: UDP
---
metadata:
  name: coredns-5fc4d8d869-vs428
podIP: 10.85.0.11
podIPs:
  - ip: 10.85.0.11
---
metadata:
  name: coredns-5fc4d8d869-cdtzr
podIP: 10.85.0.8
podIPs:
  - ip: 10.85.0.8
---

Edit1: I have the impression that the error does not come from coredns, but from the tigera-calico installation. I'm looking, can we pause this issue, I'll close it if so.

[davirezendegb]

I have something similar to AWS, I carried out the installation, and sometimes I start to experience timeouts, did you manage to find out anything? [ERROR] plugin/errors: 2 metadata.google.internal. AAAA: read udp 10.72.35.115:45212->10.72.32.2:53: i/o timeout