They both have many failing testcases but one can at least check buildability and manually to check if some patch does not add new failing testcases.
Details
Details
Diff Detail
Diff Detail
Event Timeline
Comment Actions
Looks fine to me. I believe these bots will be a very useful addition. I have one random idea you can consider if you'd like: I believe there is a way to mark (I think it involves setting flunkOnFailure=False) a particular step so that it is still run, but it does not fail the overall build. That way, you might be able to mark these bots "stable" and get breakage notifications if the build breaks, but still leave open the possibility for manual inspection of test results.