Readiness Probe (chkmqready) intermittently times out (when MQ Listener IS running)

[lsabados]

We experience an intermittent issue with our readiness probe (chkmqready) on MQ pod start up. The probe fails and the pod never goes to a "ready" state. We ended up scraping its call in the sts due to production outages. Has anyone else experienced this?

[arthurbarr]

The only issues I've ever seen are due to Kubernetes clusters which are really slow. There really isn't much code in chkmqready to go wrong (famous last words).

[arthurbarr]

What do you mean by "scraping its call in the sts"?

[lsabados]

In our environment, we found the chkmqready to be both intermittently unreliable and ineffective. We replaced (in the MQ pod sts) the execution of the chkmqready with this

      readinessProbe:
        exec:
          command:
            - chkmqready

Replaced with

    readinessProbe:
      tcpSocket:
        port: listener

Oh..and the listener value is set earlier in the sts

ports:
- containerPort: 1414
name: listener

[lsabados]

Oh..and the listener value is set earlier in the sts

[lsabados]

Accidentally closed

[lsabados]

Can someone please share the code executed by the chkmqready?

[arthurbarr]

https://github.com/ibm-messaging/mq-container/blob/master/cmd/chkmqready/main.go