Closed Bug 945667 Opened 11 years ago Closed 6 years ago

Intermittent browser_aboutHome.js | Health Reporter instance available.

Categories

(Firefox :: General, defect, P3)

Product:
Firefox
Component:
General
Platform:
x86
Linux
Type:
defect

Tracking

()

Status:
RESOLVED WORKSFORME
Tracking Flags:
Tracking Status
firefox27 --- disabled
firefox28 --- disabled
firefox29 --- disabled
firefox-esr24 --- disabled
b2g-v1.2 --- disabled

People

(Reporter: cbook, Unassigned)

References

(
URL
)

Details

(Keywords: intermittent-failure, Whiteboard: [test disabled on Linux][leave open])

Attachments

(1 file, 1 obsolete file)

Patch - Disable failing portion of browser_aboutHome.js and enable dependent tests
(deleted), patch
Details | Diff | Splinter Review 
Rev3 Fedora 12 mozilla-inbound debug test mochitest-browser-chrome on 2013-12-02 18:24:23 PST for push 59672fad0777

slave: talos-r3-fed-033

https://tbpl.mozilla.org/php/getParsedLog.php?id=31357155&tree=Mozilla-Inbound

TEST-UNEXPECTED-FAIL | chrome://mochitests/content/browser/browser/base/content/test/general/browser_aboutHome.js | Health Reporter instance available.
Given the way this failure adds around 100 minutes to a test run, I'm starting to question whether we really want to keep this test running on Linux.
This is basically the same as bug 924307. It's also hitting an exception introduced to track it down:

http://mxr.mozilla.org/mozilla-central/source/services/datareporting/DataReportingService.js#230
I liked that one better, it only added 30 minutes to the run so I hardly noticed the slow; adding 100 minutes is a lot more obvious.
(In reply to Tim Taubert [:ttaubert] from comment #89)
> This is basically the same as bug 924307. It's also hitting an exception
> introduced to track it down:
> 
> http://mxr.mozilla.org/mozilla-central/source/services/datareporting/
> DataReportingService.js#230

This is the #3 top orange on trunk at the moment. We need to fix or disable this test ASAP.
(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #168)
> And browser_findbar.js for the same reason. Why do these tests depend on
> another test running again?

Ryan, mnozilla-central still shows https://tbpl.mozilla.org/php/getParsedLog.php?id=32140982&tree=Mozilla-Central after your push, does this also need to disabled for Linux asan especially ?
Flags: needinfo?(ryanvm)
(In reply to Carsten Book [:Tomcat] from comment #171)
> Ryan, mnozilla-central still shows
> https://tbpl.mozilla.org/php/getParsedLog.php?id=32140982&tree=Mozilla-
> Central after your push, does this also need to disabled for Linux asan
> especially ?


and linux x64 -> https://tbpl.mozilla.org/php/getParsedLog.php?id=32140318&tree=B2g-Inbound error after merge back from central to b2g-inbound
(In reply to Carsten Book [:Tomcat] from comment #172)
> and linux x64 ->
> https://tbpl.mozilla.org/php/getParsedLog.php?id=32140318&tree=B2g-Inbound
> error after merge back from central to b2g-inbound

filed bug 951581 now
(In reply to Carsten Book [:Tomcat] from comment #171)
> Ryan, mnozilla-central still shows
> https://tbpl.mozilla.org/php/getParsedLog.php?id=32140982&tree=Mozilla-
> Central after your push, does this also need to disabled for Linux asan
> especially ?

(In reply to Carsten Book [:Tomcat] from comment #174)
> https://hg.mozilla.org/mozilla-central/rev/228a1f83ccee

Wes didn't merge comment 168, per the cset revs.
Flags: needinfo?(ryanvm)
(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #168)
> And browser_findbar.js for the same reason. Why do these tests depend on
> another test running again?

The really shouldn't, and losing all of this test coverage isn't really acceptable. If FHR is flaky, disabling that part of the test is a much better solution.

Steven: can you look into doing that, so that we can re-enable the majority of these tests?
Flags: needinfo?(smacleod)
Cute, apparently the FHR bits also think that performing a Google search and then calling gBrowser.stop() will not hit the network and cause flakiness. That assumption seems to not be the case.
This patch re-enables the tests, and disables only the failing test case on linux.

Gavin, can you sanity check this and let me know if it's what you had in mind?
Attachment #8356156 - Flags: review?(gavin.sharp)
Flags: needinfo?(smacleod)
Comment on attachment 8356156 [details] [diff] [review]
Patch - Disable failing portion of browser_aboutHome.js and enable dependent tests

looks good, assuming it doesn't cause a lot of intermittent orange :)
Attachment #8356156 - Flags: review?(gavin.sharp) → review+
https://hg.mozilla.org/integration/fx-team/rev/e8f3be56612d
Keywords: checkin-needed
Whiteboard: [test disabled on Linux][leave open] → [test disabled on Linux][leave open][fixed-in-fx-team]
https://hg.mozilla.org/mozilla-central/rev/e8f3be56612d
Whiteboard: [test disabled on Linux][leave open][fixed-in-fx-team] → [test disabled on Linux][leave open]
Summary: Intermittent TEST-UNEXPECTED-FAIL | chrome://mochitests/content/browser/browser/base/content/test/general/browser_aboutHome.js | Health Reporter instance available. → Intermittent browser_aboutHome.js | Health Reporter instance available.
Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.
Priority: -- → P3
This test was removed, as was healthreport, so gonna assume this can be closed.
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: