prometheus-to-sd latest image includes breaking changes?

[lechen26]

Hi

on our rabbitmq deployment within GKE i was using marketplace.gcr.io/google/rabbitmq/prometheus-to-sd:3.7 like the following:

      - name: prometheus-to-sd
        image: marketplace.gcr.io/google/rabbitmq/prometheus-to-sd:3.7
        ports:
        - name: profiler
          containerPort: 6060
        command:
        - /monitor
        - --stackdriver-prefix=custom.googleapis.com
        - --source=rabbitmq:http://localhost:15672/api/metrics?whitelisted=queue_messages_ready_ram,queue_messages_ready,queue_messages_unacknowledged,queue_messages_unacknowledged_ram,queue_consumer_utilisation,queue_memory,queue_messages_persistent,queue_messages
        - --pod-id=$(POD_NAME)
        - --namespace-id=$(POD_NAMESPACE)
        - --monitored-resource-types=k8s
        env:
        - name: POD_NAME
          valueFrom:
            fieldRef:
              fieldPath: metadata.name
        - name: POD_NAMESPACE
          valueFrom:
            fieldRef:
              fieldPath: metadata.namespace 

and it worked.
now when deploying the rabbitmq deployment on a new cluster i've noticed that prometheus-to-sd:3.7 latest image has changed with changes on flag that seems to be broken.

the flag monitored-resource-types does not work anymore and i had to change it to monitored-resource-type-prefix.
is it documented anywhere?

thanks
Chen

[lechen26]

also, looks like when using the latest image and the new flag i'm not getting the RabbitMQ metrics to stackdriver. (such as queue_messages)

once updated the version to the SHA256 of the previous image (sha256:8e3679a6e059d1806daae335ab08b304fd1d8d35cdff457baded7306b5af9ba5) and used the old flag (monitored-resource-types=k8s) things are working again.

thanks
Chen

[lechen26]

anything? 10x

[x13n]

This is not an image maintained by us:

marketplace.gcr.io/google/rabbitmq/prometheus-to-sd

Please follow up with the marketplace team instead.