b2g_ubuntu64_vm mozilla-inbound opt test gaia-unit on 2014-06-19 15:47:40 PDT for push dcf85de7c8cc slave: tst-linux64-spot-045 https://tbpl.mozilla.org/php/getParsedLog.php?id=42093487&tree=Mozilla-Inbound { 16:39:34 INFO - gaia-unit-tests TEST-START | system/test/unit/bootstrap_test.js | at boot, if NOFTU is defined (i.e in DEBUG mode) 16:39:35 INFO - gaia-unit-tests TEST-PASS | system/test/unit/bootstrap_test.js | system/Bootstrap check for updates setting at boot, if NOFTU is defined (i.e in DEBUG mode) should be enabled 16:39:35 INFO - JavaScript error: http://system.gaiamobile.org:8080/js/internet_sharing.js?time=1403221174249, line 156: AirplaneMode is not defined 16:39:35 INFO - gaia-unit-tests TEST-UNEXPECTED-FAIL | system/test/unit/bootstrap_test.js | system/Bootstrap check for updates setting at boot, if NOFTU is defined (i.e in DEBUG mode) should be enabled | ReferenceError: AirplaneMode is not defined (http://system.gaiamobile.org:8080/js/internet_sharing.js?time=1403221174249:156) 16:39:35 INFO - gaia-unit-tests INFO | stack trace: 16:39:35 INFO - Error: ReferenceError: AirplaneMode is not defined (http://system.gaiamobile.org:8080/js/internet_sharing.js?time=1403221174249:156) 16:39:35 INFO - at onerror (http://system.gaiamobile.org:8080/common/vendor/mocha/mocha.js:4959:7) 16:39:35 INFO - gaia-unit-tests TEST-UNEXPECTED-FAIL | system/test/unit/bootstrap_test.js | "after each" hook | this.sinon is null 16:39:35 INFO - gaia-unit-tests INFO | stack trace: 16:39:35 INFO - TypeError: this.sinon is null 16:39:35 INFO - at (anonymous) (http://system.gaiamobile.org:8080/common/test/sinon_helper.js:8:5) 16:39:35 INFO - at wrapper (http://system.gaiamobile.org:8080/common/test/mocha_generators.js:62:13) 16:39:35 INFO - at run (http://system.gaiamobile.org:8080/common/vendor/mocha/mocha.js:3709:7) 16:39:35 INFO - at next (http://system.gaiamobile.org:8080/common/vendor/mocha/mocha.js:3973:5) 16:39:35 INFO - at (anonymous) (http://system.gaiamobile.org:8080/common/vendor/mocha/mocha.js:3984:5) 16:39:35 INFO - at (anonymous) (http://system.gaiamobile.org:8080/common/vendor/mocha/mocha.js:4932:28) 16:39:35 INFO - gaia-unit-tests INFO | suite results (pass/fail): 2/2 16:39:35 INFO - gaia-unit-tests TEST-END | system/test/unit/bluetooth_transfer_test.js | at boot, if NOFTU is defined (i.e in DEBUG mode) }

Hei Gregor, this is causing some reds in travis errors lately: https://travis-ci.org/mozilla-b2g/gaia/jobs/29034834 Also we are having a pretty red tree cause of this bug and bug 1033213, together they are killing us :( Could you find someone to take a look? Thanks!

Is this only happening on 2.0?

Probably caused by bug 1020867.

I'll look.

Comment on attachment 8450432 [details] Pull request - master This is pretty simple - wondering if one of you guys could take a quick look and throw a R? on it so we can reopen the tree. Thanks!

Comment on attachment 8450432 [details] Pull request - master This has to be reviewed by a peer of the system module. The apparent simplicity of a patch does not warrant an exception.

(In reply to Anthony Ricaud (:rik) from comment #13) > Comment on attachment 8450432 [details] > Pull request - master > > This has to be reviewed by a peer of the system module. The apparent > simplicity of a patch does not warrant an exception. This was called out as a tree-blocker here, and having limited peers across timezones causes problems. If it's not a tree-closing bug, then I guess it's fine to wait. FWIW - I'm totally fine for any devs to use r=me for something simple like this on any of my modules.

Comment on attachment 8450432 [details] Pull request - master Hey guys - anyone got time for a review here to make the tree a lot more green? Thanks!

There was a communication mismatch obviously, sorry. Gregor told me on IRC that it was because of bug 1020917 which was already reverted. You say it's probably bug 1020867. I also didn't notice it was one of the bugs that we said was a tree-blocker, sorry. I would have said r+ otherwise. Next time, may I suggest asking review to someone available (to re-open the tree quickly) but also asking review to a peer so that they can assess if it was the proper fix once they are available again? (On another note, a unit test including real code is not a unit test. All those should be mocked.)

No worries, sounds good to me - will make sure to follow these steps in the future. TBH I'm not sure how valuable this test is anyway. Seems like integration tests should be able to cover almost all of what bootstrap is doing. Including the entire system world in the test is a bit frustrating - but it's also caught a few issues in the past.

To make things even more confused, there are 2 different errors... "AirplaneMode is not defined" and "window.AppWindowManager.init is not a function". If they have the same cause, fine, but otherwise maybe we'll need a separate issue. Thanks anyway !

Master: https://github.com/mozilla-b2g/gaia/commit/ed27d1fbfbbdc72089e50e86ca1a69d6e9514053 2.0: https://github.com/mozilla-b2g/gaia/commit/8193ad514e27440154684f112a5796e65f13fd93 (In reply to Julien Wajsberg [:julienw] from comment #21) > To make things even more confused, there are 2 different errors... > "AirplaneMode is not defined" and "window.AppWindowManager.init is not a > function". If they have the same cause, fine, but otherwise maybe we'll need > a separate issue. I have not seen this, but it should be another simple include. Are we seeing it often? In any case, file a bug and CC me and I can take a look if no one fixes it.

(In reply to Kevin Grandon :kgrandon from comment #17) > No worries, sounds good to me - will make sure to follow these steps in the > future. TBH I'm not sure how valuable this test is anyway. Seems like > integration tests should be able to cover almost all of what bootstrap is > doing. Including the entire system world in the test is a bit frustrating - > but it's also caught a few issues in the past. If we end up throwing everything under system into bootstrap.js, I prefer to mock again :/ Don't need to be real mock files but build them in the bootstrap_test only.

(In reply to Kevin Grandon :kgrandon from comment #22) > Master: > https://github.com/mozilla-b2g/gaia/commit/ > ed27d1fbfbbdc72089e50e86ca1a69d6e9514053 > 2.0: > https://github.com/mozilla-b2g/gaia/commit/ > 8193ad514e27440154684f112a5796e65f13fd93 > > (In reply to Julien Wajsberg [:julienw] from comment #21) > > To make things even more confused, there are 2 different errors... > > "AirplaneMode is not defined" and "window.AppWindowManager.init is not a > > function". If they have the same cause, fine, but otherwise maybe we'll need > > a separate issue. > > I have not seen this, but it should be another simple include. Are we seeing > it often? In any case, file a bug and CC me and I can take a look if no one > fixes it. It was one of the failure that was reported here by TBPL today... we'll see :)

(In reply to Alive Kuo [:alive][NEEDINFO!] from comment #23) > If we end up throwing everything under system into bootstrap.js, I prefer to > mock again :/ > Don't need to be real mock files but build them in the bootstrap_test only. Sure, I'm more than happy to support whatever you want to do. I still kind of like loading *real* files where possible, but maybe I'm not a unit test purist :) It might also be worth exploring some other options for component registration and start hooks.