Per-user message queue backpressure feedback
- Dominant language
- Rust
- Stars
- 0
- Forks
- 6
- PR merge metrics
- No merged PRs in 30d
Description
When a recipient's channel is full, the message is silently dropped:
```rust
Err(mpsc::error::TrySendError::Full(_)) => {
debug!(
"Recipient {} message queue is full, dropping message",
to_clone
);
}
```
Instead of dropping, the sender could receive a MessageAck with a new MessageStatus::RecipientBusy or similar, so the client knows to retry. Alternatively, bump a Prometheus counter so you can alert on it -- right now message loss is invisible.
Contributor guide
Research direction
Start at the mpsc::error::TrySendError::Full handler shown in the issue, where full recipient queues are logged and messages are dropped. Clarify whether the intended result is a MessageAck with a busy status or a Prometheus counter; done means queue-full message loss is visible to the sender or monitoring system.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- prometheus, rust
- Domain
- backend, observability
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100