You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jun 14, 2018. It is now read-only.
If anyone is interested in extracting coverage from k8s pods, feel free to tackle this!
We're at a pretty low coverage right now, and we need to improve the situation.
AFAICT, golang test coverage is only gathered from go unit tests, i.e. "go test -cover". There doesn't seem to be a way to extract coverage stats from non-tests. If we wanted to improve test coverage we'd need to beef up existing tests (e.g. with k8s/envoy mocks) and/or write additional unit tests that run on pods that can verify the glue logic that isn't covered by mocks, etc.
Starting to work on this. First step is getting basic unit test coverage published to github.com/istio/manager via jenkins. We can then follow-up with incremental PR to increase coverage.
Thanks! Take a look at mixer set-up with code coverage website integration. The problem with bazel tests is that it's really hard to do e2e stuff like iptables and envoy call-out. So we might need to do extensive mocking to get unit tests working for config generation.
No description provided.
The text was updated successfully, but these errors were encountered: