Closed Bug 880739 Opened 11 years ago Closed 7 years ago

Intermittent browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403

Categories

(Firefox for Metro Graveyard :: Tests, defect)

Product:
Firefox for Metro Graveyard
Component:
Tests
Platform:
x86
Windows 8.1
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED INCOMPLETE

People

(Reporter: philor, Unassigned)

References

Details

(Keywords: intermittent-failure, Whiteboard: [test disabled][leave open] [defect] p=0)

Attachments

(3 files)

fix
(deleted), patch
bbondy
: review+
Details | Diff | Splinter Review
debug code
(deleted), patch
bbondy
: review+
Details | Diff | Splinter Review
disable test
(deleted), patch
fryn
: review+
Details | Diff | Splinter Review
May just wind up being a "sometimes started by" for bug 880703, dunno. https://tbpl.mozilla.org/php/getParsedLog.php?id=23903633&tree=Mozilla-Inbound WINNT 6.2 mozilla-inbound pgo test metro-immersive on 2013-06-07 08:45:31 PDT for push aa9a38e68dcf slave: t-w864-ix-095 08:49:19 INFO - TEST-PASS | chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/browser_context_menu_tests.js | menu item visible 08:49:19 INFO - Download started - Disallowing suspend 08:49:19 WARNING - TEST-UNEXPECTED-FAIL | chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403 08:49:19 INFO - wfo_canceller@chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:423 08:49:19 INFO - Stack trace: 08:49:19 INFO - JS frame :: chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js :: <TOP_LEVEL> :: line 697 08:49:19 INFO - JS frame :: resource://gre/modules/Task.jsm :: TaskImpl_run :: line 193 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: reject :: line 135 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: then :: line 54 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: resolve :: line 187 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: reject :: line 210 08:49:19 INFO - JS frame :: resource://gre/modules/Task.jsm :: TaskImpl_run :: line 223 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: reject :: line 135 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: then :: line 54 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: resolve :: line 187 08:49:19 INFO - JS frame :: resource://gre/modules/commonjs/sdk/core/promise.js :: reject :: line 210 08:49:19 INFO - JS frame :: chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js :: <TOP_LEVEL> :: line 403 08:49:19 INFO - JS frame :: chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js :: wfo_canceller :: line 423 08:49:19 INFO - native frame :: <unknown filename> :: <TOP_LEVEL> :: line 0
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → DUPLICATE
Status: RESOLVED → REOPENED
Resolution: DUPLICATE → ---
Assignee: nobody → jmathies
Attached patch fix (deleted) — — Splinter Review
The default timeout here is 2 seconds. If I bring that down .5 seconds, I can reliably reproduce locally on a majority of runs. At 1.5 seconds I get it occasionally. So good guess here is that the timeout is too short. Lets try bumping this up to 5 seconds and see if this goes away.
Attachment #760983 - Flags: review?(netzen)
Comment on attachment 760983 [details] [diff] [review] fix Review of attachment 760983 [details] [diff] [review]: ----------------------------------------------------------------- Seems like we'll still get it but less often, but that's better I guess :).
Attachment #760983 - Flags: review?(netzen) → review+
https://hg.mozilla.org/mozilla-central/rev/473252f1819d
Status: REOPENED → RESOLVED
Closed: 11 years ago → 11 years ago
Resolution: --- → FIXED
Not fixed yet!
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Whiteboard: [leave-open]
Attached patch debug code (deleted) — — Splinter Review
Comment on attachment 762014 [details] [diff] [review] debug code I'd like to land this on inbound to see what events we're getting.
Attachment #762014 - Flags: review?(netzen)
Comment on attachment 762014 [details] [diff] [review] debug code Review of attachment 762014 [details] [diff] [review]: ----------------------------------------------------------------- ::: browser/metro/base/tests/mochitest/browser_context_menu_tests.js @@ +352,5 @@ > } > }); > > +var observeLogger = { > + _recvEvents: [], I don't think you need to keep track of these since you just print them out as you go.
Attachment #762014 - Flags: review?(netzen) → review+
Interesting. Nothing seems out of the ordinary here, we get a dl-start, dl-scanning, and then the time out. Plus that last patch bumped the time out to ten seconds, so something is clearly going wrong here, but it's not showing up in the logs.
Assignee: jmathies → nobody
Attached patch disable test (deleted) — — Splinter Review
requested by tree managers, #2 top orange.
Attachment #767183 - Flags: review?(fyan)
Comment on attachment 767183 [details] [diff] [review] disable test Review of attachment 767183 [details] [diff] [review]: ----------------------------------------------------------------- ::: browser/metro/profile/metro.js @@ +9,5 @@ > pref("nglayout.debug.disable_xul_cache", true); > pref("nglayout.debug.disable_xul_fastload", true); > pref("devtools.errorconsole.enabled", true); > #endif > +pref("layout.css.devPixelsPerPx", "1.0"); Ignore this, removed from my local patch.
Attachment #767183 - Flags: review?(fyan) → review+
Fun fact: this is PGO only, so rather than a flaky test, it's probably a sign that you've got some code that becomes busted when PGO gets its hands on it (or, the flakiness is a problem of things happening too quickly rather than too slowly).
And, fun question, since with PGO spaced out every 3 or 6 hours depending on the tree I'd never notice if it was: is this actually permaorange on PGO builds?
Shame, since permaorange is always easier to deal with, but no: this hits right around 50% of runs on PGO builds.
(In reply to Phil Ringnalda (:philor) from comment #136) > Fun fact: this is PGO only, so rather than a flaky test, it's probably a > sign that you've got some code that becomes busted when PGO gets its hands > on it (or, the flakiness is a problem of things happening too quickly rather > than too slowly). My money is on the latter, which is why I don't feel too bad about requesting it be disabled... :-)
Whiteboard: [leave-open] → [test disabled][leave open]
Blocks: metro-testing, 865451, metrov1defect&change
No longer blocks: 880298
Priority: -- → P3
Whiteboard: [test disabled][leave open] → [test disabled][leave open] feature=defect c=testing u=developer p=0
Blocks: 880298
No longer blocks: metro-testing
Summary: Intermittent browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403 → Defect - Intermittent browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403
I filed bug 895873 on that last one, this test has been disabled.
No longer blocks: 865451
(for too many intermittent failures)
(In reply to Ed Morley [:edmorley UTC+1] from comment #176) > Disabled for real: > https://hg.mozilla.org/integration/mozilla-inbound/rev/a49ebc80a97f This disables all of our context menu tests! What failures are you concerned about?
It looks like the most recent set of intermittent oranges (from 7/30 - 7/31) resulting in the disabling of the testsuite are actually a representation of bug 895873. Discussion on re-enabling the test suite (related to fixing a few race conditions in the context menu code as a part of bug 856264) will happen in bug 895873. As for the orange represented by the error in the title of this bug, that entire test code path was commented out in jimm's patch above. Given that it relates to downloads behavior (which we've been changing) we may need to drop that test entirely or do some significant reworking. We can resolve this bug as necessary once steps forward on that are clear.
No longer blocks: metrov2defect&change
Summary: Defect - Intermittent browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403 → Intermittent browser_context_menu_tests.js | runTests: Task failed - Error: dl-done event timeout at @chrome://mochitests/content/metro/browser/metro/base/tests/mochitest/head.js:403
Whiteboard: [test disabled][leave open] feature=defect c=testing u=developer p=0 → [test disabled][leave open] [defect] p=0
Priority: P3 → --
OS: Windows 8 Metro → Windows 8.1
Mass close of bugs in obsolete product https://bugzilla.mozilla.org/show_bug.cgi?id=1350354
Status: REOPENED → RESOLVED
Closed: 11 years ago → 7 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: