testSecurity/testGreenLarry.js failed in the Firefox 15.0b3 CI testruns on all Windows versions, seems like Mac and Linux were unaffected. There are also no previous instances of this failure on previous Beta builds, nor did it happen with Firefox 16/17 testruns. This could be a one-off connection glitch but I'm filing it anyway. Here is the stack: TimeoutError@resource://mozmill/modules/utils.js:447 waitFor@resource://mozmill/modules/utils.js:485 @resource://mozmill/modules/controller.js:685 @resource://mozmill/modules/controller.js:1328 @resource://mozmill/modules/frame.js -> file:///c:/docume~1/mozilla/locals~1/temp/tmpytmwzz.mozmill-tests/tests/functional/testSecurity/testGreenLarry.js:25 @resource://mozmill/modules/frame.js:591 @resource://mozmill/modules/frame.js:661 @resource://mozmill/modules/frame.js:707 @resource://mozmill/modules/frame.js:540 @resource://mozmill/modules/frame.js:719 @resource://jsbridge/modules/server.js:179 @resource://jsbridge/modules/server.js:183 @resource://jsbridge/modules/server.js:283
The page being loaded is https://addons.mozilla.org/licenses/5.txt which currently gives me a 404. My guess is that this was temporarily causing long load times. We need to change this test to use the mozqa.com content. What do you think about marking this bug as a duplicate of bug 760411?
(In reply to Dave Hunt (:davehunt) from comment #1) > The page being loaded is https://addons.mozilla.org/licenses/5.txt which > currently gives me a 404. My guess is that this was temporarily causing long > load times. We need to change this test to use the mozqa.com content. What > do you think about marking this bug as a duplicate of bug 760411? IMHO, its definitely a dupe of bug 760411 since its the same failure message and testGreenLarry is a security test. We will address the migration to the mozqa.com pages in bug 661121 which is currently assigned to Alex.
Going ahead and marking as DUPE -- thanks guys.
Just one more note. We can't have this testcase located at mozqa.com because there we do not have an EV cert. So if this is failing again, we can assume that something is wrong with AMO, e.g. a push to production is currently happening.