gRPC Status in queue proxy metrics
- Dominant language
- Go
- Stars
- 6.1k
- Forks
- 1.2k
- Avg merge
- 2d 7h
- Merged PRs (30d)
- 2
Description
/area monitoring
## Ask your question here:
Hi, I am exploring metrics exposed by queue-proxy for gRPC service. As of now, I noticed that in my current knative-serving (1.0.1) the gRPC status is not part of any metrics emitted by queue-proxy. `response_code` and `response_code_class` are there, however the value is always successful even for failing gRPC request.
For example, following query
```
round(sum(rate(revision_request_count[1m])) by (response_code), 0.001)
```
Produces below graph even though 100% of the request are actually failing.

Is gRPC metrics supported in more recent version? Or is there any plan to expose it?
Contributor guide
Research direction
No file or test is named. Start by tracing queue-proxy's emitted request metrics and how gRPC responses are classified, then verify the behavior against a failing gRPC request; done means the metrics expose an accurate gRPC status or the supported limitation is clearly established.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go, grpc
- Domain
- backend, observability
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 28/100