Closed Bug 1221081 Opened 9 years ago Closed 9 years ago

Intermittent browser_ProcessHangNotifications.js | Notification Displayed. - Got 2, expected 1

Categories

(Firefox :: General, defect)

Product:
Firefox
Component:
General
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED

People

(Reporter: cbook, Unassigned)

References

(
URL
)

Details

(Keywords: intermittent-failure) 

https://treeherder.mozilla.org/logviewer.html#?job_id=2591651&repo=mozilla-central

 05:12:26     INFO -  47 INFO TEST-UNEXPECTED-FAIL | browser/modules/test/browser_ProcessHangNotifications.js | Notification Displayed. - Got 2, expected 1
 05:12:26     INFO -  Stack trace:
 05:12:26     INFO -      chrome://mochikit/content/browser-test.js:test_is:940
 05:12:26     INFO -      chrome://mochitests/content/browser/browser/modules/test/browser_ProcessHangNotifications.js:active:13
 05:12:26     INFO -      chrome://global/content/bindings/notification.xml:appendNotification:179
 05:12:26     INFO -      resource:///modules/ProcessHangMonitor.jsm:ProcessHangMonitor.showNotification:327
 05:12:26     INFO -      resource:///modules/ProcessHangMonitor.jsm:ProcessHangMonitor.updateWindow:291
 05:12:26     INFO -      resource:///modules/ProcessHangMonitor.jsm:ProcessHangMonitor.updateWindows:273
 05:12:26     INFO -      resource:///modules/ProcessHangMonitor.jsm:ProcessHangMonitor.reportHang:406
 05:12:26     INFO -      resource:///modules/ProcessHangMonitor.jsm:ProcessHangMonitor.observe:195
 05:12:26     INFO -      chrome://mochitests/content/browser/browser/modules/test/browser_ProcessHangNotifications.js:terminateScriptTest:101
 05:12:26     INFO -      Tester_execTest@chrome://mochikit/content/browser-test.js:754:9
 05:12:26     INFO -      Tester.prototype.nextTest</<@chrome://mochikit/content/browser-test.js:676:7
 05:12:26     INFO -      SimpleTest.waitForFocus/waitForFocusInner/focusedOrLoaded/<@chrome://mochikit/content/tests/SimpleTest/SimpleTest.js:735:59
 05:12:26     INFO -  48 INFO TEST-PASS | browser/modules/test/browser_ProcessHangNotifications.js | proper number of buttons -
 05:12:26     INFO -  49 INFO TEST-PASS | browser/modules/test/browser_ProcessHangNotifications.js | was the correct method call made on the hang report object? -
 05:12:26     INFO -  50 INFO Leaving test terminateScriptTest
 05:12:26     INFO -  51 INFO Entering test waitForScriptTest
jimm: regression from the landing of bug 1119442 ?
Flags: needinfo?(jmathies)
Yeah that's the one, the test is telling us that for some reason there are two notifications present in the notification bar list at the start of the test. This is why bug 1119442 was backed out originally. If this is really bad lets just disable the test until someone has a chance to find the source of that notification.
Flags: needinfo?(jmathies)
If it helps you discover the source of the notification, this failure isn't actually even the least bit intermittent, it's permaorange. It's just that because it is permaorange on tests run on nightly builds, and we don't actually label tests on nightlies in any way, the only two ways of knowing a test is on a nightly are to know from the time of day when it ran on a particular m-c cset or to look at https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1221081 and think "um, huh, this has never once failed on any tree other than m-c."
(In reply to Phil Ringnalda (:philor) from comment #5)
> If it helps you discover the source of the notification, this failure isn't
> actually even the least bit intermittent, it's permaorange. It's just that
> because it is permaorange on tests run on nightly builds, and we don't
> actually label tests on nightlies in any way, the only two ways of knowing a
> test is on a nightly are to know from the time of day when it ran on a
> particular m-c cset or to look at
> https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1221081 and
> think "um, huh, this has never once failed on any tree other than m-c."

I tried debugging this on try, it didn't fail.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=44ba17a0a2c8
https://treeherder.mozilla.org/#/jobs?repo=try&revision=f4291d814262

if the test is perma fail just disable it.
Well, the two pieces of information I wanted to convey were

* you won't ever hit this with a try push or a local build or an on-push build on m-i or fx-team, because it's actually only happening on nightly builds

* other than the hassle of figuring out how to run packaged tests on a packaged build locally, it should be easy to see failing, since running the test on a nightly build fails (not quite every time like I thought, but close enough that running it twice should be certain to let you see it fail)

But it turns out there's an even easier way, push to try on top of the patches in https://treeherder.mozilla.org/#/jobs?repo=try&revision=fc8a1d7327a6&group_state=expanded&selectedJob=13988160 since whatever notification nightly builds trigger than regular builds do not trigger, apparently when we merge to aurora in three weeks we'll trigger them there every time.
(In reply to Phil Ringnalda (:philor) from comment #8)
> Well, the two pieces of information I wanted to convey were
> 
> * you won't ever hit this with a try push or a local build or an on-push
> build on m-i or fx-team, because it's actually only happening on nightly
> builds
> 
> * other than the hassle of figuring out how to run packaged tests on a
> packaged build locally, it should be easy to see failing, since running the
> test on a nightly build fails (not quite every time like I thought, but
> close enough that running it twice should be certain to let you see it fail)
> 
> But it turns out there's an even easier way, push to try on top of the
> patches in
> https://treeherder.mozilla.org/#/
> jobs?repo=try&revision=fc8a1d7327a6&group_state=expanded&selectedJob=13988160
> since whatever notification nightly builds trigger than regular builds do
> not trigger, apparently when we merge to aurora in three weeks we'll trigger
> them there every time.

Thanks, I have a debug patch for this. I'll try the aurora patch in a try push to see.
Flags: needinfo?(jmathies)
Depends on: 1227515
Flags: needinfo?(jmathies)
should be fixed now. please reopen if you still see it.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.