Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-5923

[4.12] nodeport not reachable port connection timeout

XMLWordPrintable

    • SDN Sprint 230
    • 1
    • Rejected
    • False
    • Hide

      None

      Show
      None

      Description of problem:

      NodePort port not accessible

      Version-Release number of selected component (if applicable):

      OCP 4.8.20

      How reproducible:

      $oc -n ui-nprd get services -o wide
      NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR
      docker-registry ClusterIP 10.201.219.240 <none> 5000/TCP 24d app=registry
      docker-registry-lb LoadBalancer 10.201.252.253 internal-xxxxxx.xx-xxxx-1.elb.amazonaws.com 5000:30779/TCP 3d22h app=registry
      docker-registry-np NodePort 10.201.216.26 <none> 5000:32428/TCP 3d16h app=registry

      $oc debug node/ip-xxx.ca-central-1.compute.internal
      Starting pod/ip-xxx.ca-central-1computeinternal-debug ...
      To use host binaries, run `chroot /host`
      Pod IP: 10.81.23.96
      If you don't see a command prompt, try pressing enter.
      sh-4.2# chroot /host
      sh-4.4# nc -vz 10.81.23.96 32428
      Ncat: Version 7.70 ( https://nmap.org/ncat )
      Ncat: Connection timed out.

      In a new-created namespaces the same deployment works:

      [RHEL7:> oc project
      Using project "test-c1" on server "https://api.xx.xx.xxxx.xx.xx:6443".
      [RHEL7:- ~/tmp]> oc port-forward service/docker-registry-np 5000:5000
      Forwarding from 127.0.0.1:5000 -> 5000

      [1]+ Stopped oc4 port-forward service/docker-registry-np 5000:5000
      [RHEL7: ~/tmp]> bg %1
      [1]+ oc4 port-forward service/docker-registry-np 5000:5000 &
      [RHEL7: ~/tmp]> nc -v localhost 5000
      Ncat: Version 7.50 ( https://nmap.org/ncat )
      Ncat: Connected to 127.0.0.1:5000.
      Handling connection for 5000

      [RHEL7: ~/tmp]> kill %1
      [RHEL7: ~/tmp]>
      [1]+ Terminated oc4 port-forward service/docker-registry-np 5000:5000
      [RHEL7: ~/tmp]> oc get services
      NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
      docker-registry-np NodePort 10.201.224.174 <none> 5000:31793/TCP 68s

      [RHEL7: ~/tmp]> oc get pods -o wide
      NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
      registry-75b7c7fd94-rx29j 1/1 Running 0 7m5s 10.201.1.29 ip-xxx.ca-central-1.compute.internal <none> <none>
      [RHEL7: ~/tmp]> oc debug node/ip-xxx.ca-central-1.compute.internal
      Starting pod/ip-xxxca-central-1computeinternal-debug ...
      To use host binaries, run `chroot /host`
      Pod IP: 10.81.23.87
      If you don't see a command prompt, try pressing enter.
      sh-4.2# chroot /host
      sh-4.4# nc -v 10.81.23.87 31793
      Ncat: Version 7.70 ( https://nmap.org/ncat )
      Ncat: Connected to 10.81.23.87:31793.

      Actual results:

      • Working on new created namespace
      • Not working on already created namespace

      Expected results:

      • Suppose to work on all namespaces.

      Additional info:

      • This cluster get upgrade from 4.7.x to 4.8 and then they manually enable OVN.
      • The issue was happening on all namespaces but after restarting the ovnkube-master-xxxx pods only the newly created namespaces work.

            npinaeva@redhat.com Nadia Pinaeva
            npinaeva@redhat.com Nadia Pinaeva
            Anurag Saxena Anurag Saxena
            Votes:
            0 Vote for this issue
            Watchers:
            6 Start watching this issue

              Created:
              Updated:
              Resolved: