Inspired by bug 825654. In certain cases it's easy to miss real failures when using only-unstarred mode. eg: * Blue retry jobs that are continually retrying / running away * Perma-orange that is being inadvertently starred push after push We should perhaps consider showing these jobs if there are more than a certain number of them consecutively (on the same push in the case of retries, or in subsequent pushes in the case of perma-orange).
Would need to make sure we take SETA into account for the second case.
Could you explain more?
Just want to make sure pushes where the job doesn't get run at all are taken into account with this. If a job is permafail on pushes 1, 3, and 4 but it isn't even scheduled on push 2, Treeherder should still surface the failures the same as if they happened on all four pushes.
Yeah agreed, by "in subsequent pushes" in comment 0 I meant (but didn't write) "in subsequent runs, where the runs are on different pushes".
Status: NEW → RESOLVED
Last Resolved: a year ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.