Closed Bug 1403037 Opened 2 years ago Closed 2 years ago
TEST-UNEXPECTED-FAIL | toolkit/components/telemetry/tests/unit/test
_Telemetry Environment .js "undefined" == "google .com"
59 bytes, text/x-review-board-request
TEST-UNEXPECTED-FAIL | toolkit/components/telemetry/tests/unit/test_TelemetryEnvironment.js | test_checkEnvironment - [test_checkEnvironment : 455] "undefined" == "google.com" This comes from https://dxr.mozilla.org/mozilla-central/rev/7e962631ba4298bcefa571008661983d77c3e652/toolkit/components/telemetry/tests/unit/test_TelemetryEnvironment.js#455 which was added about one year ago: https://hg.mozilla.org/mozilla-central/rev/0bf376ad9b85#l3.91 It only fails on Windows, not on Linux. Sorry, Mac currently broken, so I can't tell. I've first seen it on Fri Sep 22, 2017, 1:36:07 here: https://treeherder.mozilla.org/#/jobs?repo=comm-central&revision=e08911b56e26b2d7dbfd4d0cd4b1380555e70625&selectedJob=132611823 Georg (doesn't take NI) and Matt, can you shed some light onto this please.
Looks like it's bug 1400870 that broke this, but only by exposing an existing problem, not by doing anything wrong. The reason it's failing now is this line: https://hg.mozilla.org/releases/mozilla-beta/file/7e35ed4a8c47/toolkit/components/telemetry/tests/unit/test_TelemetryEnvironment.js#l894 which means that the data being looked for in the failing step isn't generated for Thunderbird (ironically enough as I recall that line was added to fix a different TB test failure). Prior to bug 1400870 this would have been silently ignored; now it generates a failure instead, probably correctly. To fix this we should skip running the steps that are failing if we did not run spoofAttributionData().
Thanks a lot for the quick attention!
(In reply to Matt Howell [:mhowell] from comment #1) > Looks like it's bug 1400870 that broke this, but only by exposing an > existing problem, not by doing anything wrong. Thanks for fixing this Matt!
Assignee: nobody → mhowell
Comment on attachment 8912081 [details] Bug 1403037 - Don't test attribution data in applications that don't use it. https://reviewboard.mozilla.org/r/183462/#review188632
Attachment #8912081 - Flags: review?(alessio.placitelli) → review+
Pushed by email@example.com: https://hg.mozilla.org/integration/autoland/rev/6f54019717c1 Don't test attribution data in applications that don't use it. r=Dexter
Status: NEW → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → Thunderbird 58.0
Comment on attachment 8912081 [details] Bug 1403037 - Don't test attribution data in applications that don't use it. Approval Request Comment [Feature/Bug causing the regression]: Bug 1400870. [User impact if declined]: None, but test failures on comm-beta [Is this code covered by automated tests?]: Yes. [Has the fix been verified in Nightly?]: Test only. [Is the change risky?]: No, just added a test for "browser". [Why is the change risky/not risky?]: See above. [String changes made/needed]: None.
Comment on attachment 8912081 [details] Bug 1403037 - Don't test attribution data in applications that don't use it. Test only change do not require relman approval.
Ryan, do you know the way forward here? How does it get uplifted? I push it myself?
Normally you'd just add [checkin-needed-beta] to the whiteboard, but no need to this time with the NI.
You need to log in before you can comment on or make changes to this bug.