[RHOCP 4] HAProxy container status 'unready' in openshift-vsphere-infra namespace, port 30000 already in use by 'ovnkube' process

Solution Verified - Updated -

Issue

  • After upgrading Openshift cluster from v4.8.45 to v4.9.43, the HAProxy container in openshift-vsphere-infra namespace is unable to startand shows status as unready

    LAST SEEN  TYPE     REASON      OBJECT                           MESSAGE
    5h56m      Normal   Pulled      pod/haproxy-ocp4-xxxx-xxxx  Container image "quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:xxxxxx" already present on machine
    5h30m      Warning  ProbeError  pod/haproxy-ocp4-xxxx-xxxx  Liveness probe error: Get "http://10.xx.xx.xx:30936/haproxy_ready": dial tcp 10.xx.xx.xx:30936: connect: connection refused
                                                                 body:
    5h36m      Warning  BackOff     pod/haproxy-ocp4-xxxx-xxxx  Back-off restarting failed container
    
    
  • HAProxy logs show Starting proxy stats: cannot bind socket [::1:30000]

  • netstat output shows ovnkube process is bound to port 30000

Environment

  • Red Hat OpenShift Container Platform (RHOCP)
    • 4.9

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content