Hi everyone! We hit an interesting issue today with Airbyte 2.1.0 on GKE where a MySQL CDC sync job failed after 15 attempts (2.5h, 0 records synced) and now every sync trigger fails. After digging through GKE logs we though we traced the root cause to the cluster autoscaler evicting the

temporal

and

workload-api-server

pods mid-flight — not a connector issue at all. The failure surface in the UI as

Source process exited with non-zero exit code 3

+

TransientErrorException: streams without terminal status

, which made diagnosis quite tricky without GKE-level visibility. We have a full write-up with relevant log excerpts ready. Is there a dedicated channel or thread where it's best to share a detailed bug report?

I'm guessing it would probably be best to put this in as a GitHub issue. Are you using Autopilot Mode or Standard Mode in GKE?

Ciao @Justin Beasley, ok thanks, we're using autopilot. I would add this detail on github issue