The deeply-hated (bug 449763) test_handlerApps.xhtml, which launches iCal on OS X and leaves it running (which in the case of our 10.7+ slaves, means leaves it running across reboots so that it's still running when the slave later runs talos, bug 743594), is permaorange on 10.8 because it launches /Applications/iCal.app, having failed to anticipate that Apple would decide to switch the name of the app to match the name of the iPhone app. This would be a marvelous opportunity to STOP LAUNCHING ICAL, but because this test is holding all of mochitest-5 on 10.8 hostage (it's hidden only because of this), we need to do *something* about it.
Oh yeah, file handling is totally unowned. I'm not willing to be the one to test launching Calendar.app, nor am I willing to be responsible for the havoc that causing it to be constantly running on every single test slave may cause, and the hack that SeaMonkey used to just open /bin/echo instead in bug 749872 totally does not work (it just causes the test to fail on 10.5-10.7 and 10.8 opt, but on 10.8 debug it crashes rather spectacularly, in Launch Services), so I just disabled the test on 10.8 in http://hg.mozilla.org/integration/mozilla-inbound/rev/f7c89de3ab43
Whiteboard: [leave open][qa-]
Whiteboard: [leave open][qa-] → [qa-]
Hi, it's me again. https://hg.mozilla.org/integration/mozilla-inbound/rev/b3b9d66e27d5
Summary: test_handlerApps.xhtml is permaorange on 10.8 from trying to launch iCal.app, which no longer exists → test_handlerApps.xhtml is permaorange on 10.8 and 10.9 from trying to launch iCal.app, which no longer exists
Whiteboard: [qa-] → [qa-][test disabled][leave open]
Summary: test_handlerApps.xhtml is permaorange on 10.8 and 10.9 from trying to launch iCal.app, which no longer exists → test_handlerApps.xhtml is permaorange on 10.8 and 10.9 and 10.10 from trying to launch iCal.app, which no longer exists
Apparently the way I was disabling it lead to intermittently calling finish twice, which I don't understand, so I just disabled it completely on OS X in https://hg.mozilla.org/integration/mozilla-inbound/rev/0e25e47b597c.
You need to log in before you can comment on or make changes to this bug.