Intelligent health checks

[KasonBraley]

Describe the solution you would like

I am looking to see if it is currently possible to implement this health check idea https://blog.cloudflare.com/intelligent-automatic-restarts-for-unhealthy-kafka-consumers/ with kakfa-go.

Some initial struggles I am having with implementing this:

• Detecting a rebalancing event. Based on this open issue, Need a way to detect rebalancing event #1023, it seems it may be possible, but requires creating a new GroupRebalancer just to be notified of the event.
• Listing the partitions assigned to a Reader that is within a Consumer Group. This closed issue How do I get a list of partitions assigned to a consumer within a Consumer Group ?  #717 mentions it is not available, but I think the newer ConsumerGroup type possibly makes this possible?
• Retrieving the offsets for the Consumer Group. I believe I can make use of the methods mentioned here Is there some way to get current lag of the current consumer group? #862 (comment) once I figure out how to retrieve the current partitions assigned to the Consumer Group.

If these are all currently possible, I would greatly appreciate some advice and tips on how to implement this. As these lower level API's are a little more involved than just using the Reader.