Logs on Kafka connector failure

DevOps Kafka
#1

When creating a Connector (specifically AWS Lambda Sink) in Confluent Cloud and it fails how do I gain information as to why it failed?

For example.

     ID     |            Name            | Status | Type |             Trace
+-----------+----------------------------+--------+------+--------------------------------+
  lcc-8p33m | stocks-highlow-lambda-sink | FAILED | sink | Unexpected error occurred with
            |                            |        |      | connector. Support has been
            |                            |        |      | notified about the issue.```
Can I get logs somewhere?

I see I have a describe command but its also pretty uninformative.
#2

Hi, Unfortunately Confluent does not expose the logs directly today. This is an enhancement we hope to have available to users in early Q3. In the meantime could you please try running the describe with the -vvv flag? I’ll look on the backend for clues as to why this failed as well.

#3

The team did notice this message:

org.apache.kafka.common.errors.GroupAuthorizationException: Not authorized to access group: connect-lcc-8p33m```
The team wasn’t able to reproduce but the “theory” is that you deleted the group read ACL after validation. Its a weak theory but if you retry and provide the steps to reproduce I’ll update the team that is looking into this.
#4

Ok so there is still some maturation of the product required but, good to hear it’s on the roadmap. I killed off this connector a couple of days ago as I wasn’t sure if I was getting charged for it

#5

When I recreated it though it worked

#6

My guess is that maybe the API key/secret pair hadn’t fully registered with the cluster yet or maybe the ACL hadn’t fully replicated / taken effect?

#7

That seems possible to me, we had instances of that in the past but we addressed it. I’ll propose this to the team though as a potential scenario maybe we have some regression.