Closed Bug 1671539 Opened 4 years ago Closed 3 years ago

Intermittent [Fission] dom/base/test/test_bug1406102.html | intersectionRatio should be greater than zero

Categories

(Core :: Layout, defect)

defect

Tracking

()

RESOLVED WORKSFORME
Fission Milestone M8

People

(Reporter: intermittent-bug-filer, Unassigned)

References

Details

(Keywords: intermittent-failure, Whiteboard: [layout:triage-discuss])

Filed by: nbeleuzu [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer.html#?job_id=318800933&repo=mozilla-central
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/GjVfgEtQQmCGE_F_axjCFg/runs/0/artifacts/public/logs/live_backing.log


[task 2020-10-16T04:17:40.140Z] 04:17:40     INFO - TEST-START | http://mochi.test:8888/tests/dom/base/test/test_bug1406102.html
[task 2020-10-16T04:17:40.231Z] 04:17:40     INFO - GECKO(1518) | [Child 4918, Main Thread] WARNING: Extra shutdown CC: 'i < NORMAL_SHUTDOWN_COLLECTIONS', file /builds/worker/checkouts/gecko/xpcom/base/nsCycleCollector.cpp:3359
[task 2020-10-16T04:17:40.289Z] 04:17:40     INFO - GECKO(1518) | [Child 4918, Main Thread] WARNING: NS_ENSURE_TRUE(Preferences::InitStaticMembers()) failed: file /builds/worker/checkouts/gecko/modules/libpref/Preferences.cpp:4287
[task 2020-10-16T04:17:40.330Z] 04:17:40     INFO - GECKO(1518) | [5026, Unnamed thread 7f509195d5e0] WARNING: XPCOM objects created/destroyed from static ctor/dtor: file /builds/worker/checkouts/gecko/xpcom/base/nsTraceRefcnt.cpp:202
[task 2020-10-16T04:17:40.330Z] 04:17:40     INFO - GECKO(1518) | ### XPCOM_MEM_BLOAT_LOG defined -- logging bloat/leaks to /tmp/tmpGqOk9L.mozrunner/runtests_leaks.log
[task 2020-10-16T04:17:40.330Z] 04:17:40     INFO - GECKO(1518) | [5026, Unnamed thread 7f509195d5e0] WARNING: XPCOM objects created/destroyed from static ctor/dtor: file /builds/worker/checkouts/gecko/xpcom/base/nsTraceRefcnt.cpp:202
[task 2020-10-16T04:17:40.330Z] 04:17:40     INFO - GECKO(1518) | [5026, Main Thread] WARNING: XPCOM_MEM_BLOAT_LOG is set, disabling native allocations.: file /builds/worker/checkouts/gecko/tools/profiler/core/platform.cpp:247
[task 2020-10-16T04:17:40.355Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: QuotaManager failure: 'OkIf(outputStream)', file ActorsParent.cpp:2522
[task 2020-10-16T04:17:40.356Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: QuotaManager failure: 'GetBinaryOutputStream(*file, kUpdateFileFlag)', file ActorsParent.cpp:8340
[task 2020-10-16T04:17:40.356Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: 'NS_FAILED(rv)', file /builds/worker/checkouts/gecko/dom/quota/ActorsParent.cpp:8204
[task 2020-10-16T04:17:40.365Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: 'NS_FAILED(rv)', file /builds/worker/checkouts/gecko/dom/quota/ActorsParent.cpp:8114
[task 2020-10-16T04:17:40.365Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: QuotaManager failure: 'OkIf(outputStream)', file ActorsParent.cpp:2522
[task 2020-10-16T04:17:40.366Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: QuotaManager failure: 'GetBinaryOutputStream(*file, kUpdateFileFlag)', file ActorsParent.cpp:8340
[task 2020-10-16T04:17:40.366Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: 'NS_FAILED(rv)', file /builds/worker/checkouts/gecko/dom/quota/ActorsParent.cpp:8204
[task 2020-10-16T04:17:40.366Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, QuotaManager IO] WARNING: 'NS_FAILED(rv)', file /builds/worker/checkouts/gecko/dom/quota/ActorsParent.cpp:8114
[task 2020-10-16T04:17:40.443Z] 04:17:40     INFO - GECKO(1518) | [Child 5026, Main Thread] WARNING: could not set real-time limit in CubebUtils::InitLibrary: file /builds/worker/checkouts/gecko/dom/media/CubebUtils.cpp:626
[task 2020-10-16T04:17:40.499Z] 04:17:40     INFO - TEST-INFO | started process screentopng
[task 2020-10-16T04:17:40.850Z] 04:17:40     INFO - TEST-INFO | screentopng: exit 0
[task 2020-10-16T04:17:40.850Z] 04:17:40     INFO - TEST-UNEXPECTED-FAIL | dom/base/test/test_bug1406102.html | intersectionRatio should be greater than zero 
[task 2020-10-16T04:17:40.850Z] 04:17:40     INFO -     SimpleTest.ok@SimpleTest/SimpleTest.js:417:16
[task 2020-10-16T04:17:40.850Z] 04:17:40     INFO -     observer<@dom/base/test/test_bug1406102.html?currentTestURL=dom%2Fbase%2Ftest%2Ftest_bug1406102.html&closeWhenDone=1&showTestReport=false&expected=pass:27:7
[task 2020-10-16T04:17:40.851Z] 04:17:40     INFO - GECKO(1518) | MEMORY STAT | vsize 2558MB | residentFast 158MB | heapAllocated 9MB
[task 2020-10-16T04:17:40.851Z] 04:17:40     INFO - TEST-PASS | dom/base/test/test_bug1406102.html | intersectionRatio should be greater than zero 
[task 2020-10-16T04:17:40.852Z] 04:17:40    ERROR - Command structuredLogger.error not found
[task 2020-10-16T04:17:40.852Z] 04:17:40    ERROR -       in xOriginDispatchMap
[task 2020-10-16T04:17:40.852Z] 04:17:40     INFO - TEST-UNEXPECTED-ERROR | http://mochi.test:8888/tests/dom/base/test/test_bug1406102.html | called finish() multiple times
[task 2020-10-16T04:17:40.853Z] 04:17:40     INFO - TEST-INFO took 468ms
[task 2020-10-16T04:17:40.853Z] 04:17:40     INFO - TEST-OK | http://mochi.test:8888/tests/dom/base/test/test_bug1406102.html
[task 2020-10-16T04:17:40.854Z] 04:17:40     INFO - GECKO(1518) | [Parent 1518, Main Thread] WARNING: IPC message discarded: actor cannot send: file /builds/worker/workspace/obj-build/dist/include/mozilla/ipc/ProtocolUtils.h:295
[task 2020-10-16T04:17:40.855Z] 04:17:40     INFO - GECKO(1518) | ###!!! [Parent][RunMessage] Error: Channel closing: too late to send/recv, messages will be lost```
Summary: Intermittent [Tier2] dom/base/test/test_bug1406102.html | intersectionRatio should be greater than zero → Intermittent [Fission] dom/base/test/test_bug1406102.html | intersectionRatio should be greater than zero

ahal hit this test failure multiple times when trying to enable mochitest-plain in fission-xorigin mode on more platforms in bug 1694835:

https://treeherder.mozilla.org/jobs?repo=try&revision=a4813c46eff128482994870bf8a06677022cee24&selectedTaskRun=OrW1ws-wTH6pAfP9kTTeKg.0

Blocks: 1700781
Fission Milestone: --- → M7a
Priority: P5 → P3

test_bug1406102.html is a test for IntersectionObserver, so move to Layout.

Component: DOM: Core & HTML → Layout

Sending this bug to Layout triage. The Fission team hit this test failure multiple times when trying to enable mochitest-plain in fission-xorigin mode on more platforms in bug 1694835, so Fission and/or xorigin mode might make this infrequent intermittent failure (about 5-15 failures per week) more frequent.

https://treeherder.mozilla.org/jobs?repo=try&revision=a4813c46eff128482994870bf8a06677022cee24&selectedTaskRun=OrW1ws-wTH6pAfP9kTTeKg.0

See also bug 1694536 about intermittent IntersectionObserver test test_bug1399605.html failing more frequently in fission-xorigin mode.

Severity: normal → --
Priority: P3 → --
See Also: → 1694536
Whiteboard: [layout:triage-discuss]

As in bug 1694536, it looks like we've gone weeks without any failure reports here.

Here's a graph of failures since the start of the year -- the most recent report here was on March 25, nearly 6 weeks ago:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1671539&startday=2021-01-01&endday=2021-05-04&tree=all

Maybe this is fixed? (Seems like it'd either be that, or this could be due to something like us disabling/changing the affected platform/configuration around March 25 -- cpeterson, do you know of any changes like that which could indicate that this is still an issue & we're just not getting the signal?)

Flags: needinfo?(cpeterson)

(In reply to Daniel Holbert [:dholbert] from comment #27)

Maybe this is fixed? (Seems like it'd either be that, or this could be due to something like us disabling/changing the affected platform/configuration around March 25 -- cpeterson, do you know of any changes like that which could indicate that this is still an issue & we're just not getting the signal?)

The sudden disappearance of the test failure in that graph is quite interesting! I don't see any bugs that landed around that time which might have fixed this, but I will resolve this bug as "fixed" for now.

Status: NEW → RESOLVED
Closed: 3 years ago
Flags: needinfo?(cpeterson)
Resolution: --- → FIXED

Reopening this bug because this failure was reproduced in Fission+xorigin mode for bug 1716404:

https://treeherder.mozilla.org/logviewer?job_id=342000217&repo=try&lineNumber=5795

This test_bug1406102.html test is now disabled for fission+xorigin mode on all platforms, so we won't see more failures:

[test_bug1406102.html]
skip-if =
  fission && xorigin  # Bug 1716404 - New fission platform triage

https://searchfox.org/mozilla-central/rev/5e955a47c4af398e2a859b34056017764e7a2252/dom/base/test/mochitest.ini#600-602

Status: RESOLVED → REOPENED
Resolution: FIXED → ---

Frank, could you please assign this to someone in the Layout team? Thanks!

Flags: needinfo?(fgriffith)

Will be assigned on today's Layout Triage call..

Flags: needinfo?(fgriffith)
Assignee: nobody → dholbert

(Just setting expectations: I probably won't be able to seriously look at this until mid-July, since I'll be out for most of the next few weeks. I suspect we can live with this until then, given that the test is disabled & not causing spam.)

(In reply to Daniel Holbert [:dholbert] from comment #32)

(Just setting expectations: I probably won't be able to seriously look at this until mid-July, since I'll be out for most of the next few weeks. I suspect we can live with this until then, given that the test is disabled & not causing spam.)

Thanks for the heads up. In that case, I will move this bug from our current Fission Milestone (M7a) to our next (M8).

Fission Milestone: M7a → M8
See Also: → 1678089

I'd highly suspect this is a case where we haven't received any transform matrix which is representing the position of the iframe in question from APZ side.

Okay, now I am also 100% sure this is not related to the test failures on fis-xorig runs (bug 1716404), since even with --enable-fission the IntersectionObserver in the test doesn't run in an OOP iframe process, it runs in the same process of the top level content document.

I've fixed the failures on fis-xorig runs in bug 1716404.

No longer blocks: 1716404

(In reply to Hiroyuki Ikezoe (:hiro) from comment #35)

Okay, now I am also 100% sure this is not related to the test failures on fis-xorig runs (bug 1716404), since even with --enable-fission the IntersectionObserver in the test doesn't run in an OOP iframe process, it runs in the same process of the top level content document.

I've fixed the failures on fis-xorig runs in bug 1716404.

Hiro, what do we need to fix for this bug after you fix bug 1716404 and re-enable test_bug1406102.html for fission && xorigin?

This test is currently disabled for fission && xorigin, not all Fission. And I don't see any recent Fission failures without xorigin for this test:

https://treeherder.mozilla.org/intermittent-failures/bugdetails?startday=2021-03-15&endday=2021-07-13&tree=trunk&bug=1671539

Flags: needinfo?(hikezoe.birchill)

I'd say we can close this bug. To be honest, I don't know the reason why this test fails on Fission runs, I only know the reason why it fails on Fission&&XOrig runs. Is it possible that we had run mochitest-fission with xorig setup in the past?

Anyway, I tried to run 1000 times this test locally with "--enable-fission" and MOZ_CHAOSMODE=255 without the change for bug bug 1716404, I didn't see any failures.

Flags: needinfo?(hikezoe.birchill)

(In reply to Hiroyuki Ikezoe (:hiro) from comment #37)

I'd say we can close this bug. To be honest, I don't know the reason why this test fails on Fission runs, I only know the reason why it fails on Fission&&XOrig runs. Is it possible that we had run mochitest-fission with xorig setup in the past?

Yes. We enabled some fission && xorigin tests for just Linux debug about a year ago (bug 1647720). Last month, we enabled fission && xorigin tests on Windows, macOS, and more Linux configurations (bug 1700781). So this bug might have filed for an old test failure that was been fixed before we started filing new bugs for fission && xorigin failures found by bug 1700781.

Anyway, I tried to run 1000 times this test locally with "--enable-fission" and MOZ_CHAOSMODE=255 without the change for bug bug 1716404, I didn't see any failures.

Thanks for verifying! In that case, I'll close this bug.

Status: REOPENED → RESOLVED
Closed: 3 years ago3 years ago
Resolution: --- → WORKSFORME
Assignee: dholbert → nobody
You need to log in before you can comment on or make changes to this bug.