Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-8265

Master node br-ex interface timeout after 18-24 hours uptime, lost IPv4 and IPv6 address

XMLWordPrintable

    • Important
    • No
    • SDN Sprint 234, SDN Sprint 235, SDN Sprint 236
    • 3
    • Rejected
    • False
    • Hide

      None

      Show
      None
    • 6/28: Chad is following up on this defect.

      Description of problem:

      Deployed dual stack hub cluster (IPv4 and IPv6) , Cluster gets deployed and ready . After 18 to 24 hours  master nodes goes not ready state . The infra machine can't reach br-ex interface either through IPv4 or IPv6 address.
      Logging into console of master-0  the br-ex interface lost both address. performing nmcli connection down/up br-ex  the interface timeout  get ipv4/ipv6 address.
      The NIC 100G -card used (Intel Corporation Ethernet Controller E810-C for QSFP (rev 02)
      workaround: reboot node , The ipv4/ipv6 address get assigned to br-ex interface.
      Note : The DHCP/DHCPv6 lease refresh is working properly.   

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

      4.12.0-0.nightly-2023-03-01-040753 

      How reproducible:

      Always, every 18 to 24 hours run time 

      Steps to Reproduce:

      1.Bring dual stack hub cluster 4.12.0.0 nightly build 
      2.Monitor nodes are in ready  state after 24 hours 
      3. The nodes go to not ready state
      4.[kni@infra6 ~]$ oc get nodes
      NAME                           STATUS     ROLES                         AGE   VERSION
      master-0.slcm.nokia.bos2.lab   NotReady      control-plane,master,worker   26h   v1.25.4+18eadca
      master-1.slcm.nokia.bos2.lab   Ready   control-plane,master,worker   26h   v1.25.4+18eadca
      master-2.slcm.nokia.bos2.lab   Ready      control-plane,master,worker   26h   v1.25.4+18eadca

      Actual results:

      Nodes are 'NotReady' state and lost ipv4 and ipv6 connectivity 

      Expected results:

      Nodes are up running always and reachable through ipv4 or ipv6 network

      Additional info:

      [core@master-0 ~]$ ethtool -i ens5
      driver: ice
      version: 4.18.0-372.46.1.el8_6.x86_64
      firmware-version: 2.40 0x80007063 1.2898.0
      expansion-rom-version:
      bus-info: 0000:b1:00.0
      supports-statistics: yes
      supports-test: yes
      supports-eeprom-access: yes
      supports-register-dump: yes
      supports-priv-flags: yes
      
      br-ex: connected to ovs-if-br-ex
              "br-ex"
              ovs-interface (openvswitch), 40:A6:B7:7B:CA:58, sw, mtu 1500
              master br-ex, ip4 default, ip6 default
              inet4 192.168.44.48/32
              inet4 192.168.44.49/32
              inet4 192.168.44.50/25
              route4 192.168.44.0/25 metric 48
              route4 default via 192.168.44.1 metric 48
              inet6 2001:db8:44::50/128
              inet6 2001:db8:44:0:42a6:b7ff:fe7b:ca58/64
              inet6 fe80::42a6:b7ff:fe7b:ca58/64
              route6 2001:db8:44::/64 metric 48
              route6 fe80::/64 metric 1024
              route6 default via fe80::77c:a757:2808:9cb5 metric 48
              route6 2001:db8:44::50/128 metric 48
      
      ens5: connected to ovs-port-phys0
              "Intel E810-C for QSFP"
              ovs-port (openvswitch), sw
              master br-ex

            nmoraiti Nikolaos Moraitis
            shreepat@redhat.com Shreemanth Patil
            Anurag Saxena Anurag Saxena
            Shreemanth Patil
            Votes:
            0 Vote for this issue
            Watchers:
            8 Start watching this issue

              Created:
              Updated:
              Resolved: