googleapis / googleapis/google-cloud-rust
Ideal exactly-once + `NackImmediately` shutdown behavior
Open
api: pubsub
type: cleanup
- Dominant language
- Rust
- Stars
- 955
- Forks
- 144
- Avg merge
- 1d 5h
- Merged PRs (30d)
- 279
Description
Ideally for exactly-once messages, if the client is configured to `NackImmediately`, it should:
- flush all pending acks
- await all pending ack RPCs
The client should not retry transient errors for an acknowledge RPC after shutdown.
---
This is a nice to have. The current behavior is well-defined. It could just be more helpful by returning fewer `Shutdown` errors, where the application does not know if acks were successful or not.
Contributor guide
Assessment
This issue has not been assessed yet.