Intermittent TEST-UNEXPECTED-TIMEOUT | dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html | application timed out after 330 seconds with no output

NEW
Unassigned

Status

()

defect
P5
normal
3 years ago
Yesterday

People

(Reporter: intermittent-bug-filer, Unassigned)

Tracking

({intermittent-failure, leave-open})

Firefox Tracking Flags

(Not tracked)

Details

(Whiteboard: [gfx-noted][stockwell disabled])

Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Whiteboard: [gfx-noted]
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Bulk priority update of open intermittent test failure bugs. 

P3 => P5

https://bugzilla.mozilla.org/show_bug.cgi?id=1381960
Priority: P3 → P5
These started to fail frequently on Windows 10 after bug 1403484 landed.
Blocks: 1403484
Flags: needinfo?(rthijssen)
Flags: needinfo?(jmaher)
yes, seeing intermittent failures here:
https://treeherder.mozilla.org/#/jobs?repo=autoland&filter-searchStr=mochitest-webgl windows10&group_state=expanded

error is the same as the intermittent for mochitest-webgl on windows 7 (bug 1397201), timeout at 330 seconds, screenshots on win 10 show a white screen (no desktop, no firefox window)

bug 1403484 was to enable the mochitest-webgl suite on windows 10. i suspect there might be an issue with the test (test_pixel_pack_buffer) since it's failing only intermittently but on both platforms (7, 10).
Flags: needinfo?(rthijssen)
I think we will track this as an intermittent and disable it if the test fails too frequently.
Flags: needinfo?(jmaher)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
this started failing I believe after migrating webgl tests from hardware to vm.  the screenshot is blank/white:
https://public-artifacts.taskcluster.net/PKf0GWGuTXGuxiAmbiFL3g/0/public/test_info/mozilla-test-fail-screenshot_bajn05.png

here is a recent log file:
https://treeherder.mozilla.org/logviewer.html#?repo=autoland&job_id=134855263&lineNumber=3521


here is a snippet from the log:
09:44:33     INFO -  485 INFO TEST-START | dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html
09:44:34     INFO -  GECKO(6580) | JavaScript warning: http://mochi.test:8888/tests/dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html, line 135: Error: WebGL warning: Exceeded 16 live WebGL contexts for this principal, losing the least recently used one.
09:44:34     INFO -  GECKO(6580) | WebGL(000002C298ED5800)::ForceLoseContext
09:44:38     INFO -  GECKO(6580) | ###!!! [Parent][MessageChannel] Error: (msgtype=0x15008F,name=PBrowser::Msg_UpdateNativeWindowHandle) Channel error: cannot send/recv
09:44:38     INFO -  GECKO(6580) | ###!!! [Parent][MessageChannel] Error: (msgtype=0x150083,name=PBrowser::Msg_Destroy) Channel error: cannot send/recv
09:44:38     INFO -  GECKO(6580) | ###!!! [Parent][MessageChannel] Error: (msgtype=0x1000B,name=PAPZ::Msg_Destroy) Channel error: cannot send/recv
09:50:08     INFO -  Buffered messages finished
09:50:08  WARNING -  TEST-UNEXPECTED-TIMEOUT | dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html | application timed out after 330 seconds with no output
09:50:08     INFO -  486 ERROR Force-terminating active process(es).
09:50:08     INFO -  Determining child pids from psutil
09:50:08     INFO -  Found child pids: [10280, 10224]
09:50:08     INFO -  Killing process: 10280
09:50:08     INFO -  TEST-INFO | started process screenshot
09:50:08     INFO -  TEST-INFO | screenshot: exit 0
09:50:08     INFO -  mozcrash Using Z:\task_1507109317\build\tests\bin\minidumpwriter.exe to write a dump to c:\users\genericworker\appdata\local\temp\tmp1kpmxa.mozrunner\minidumps\5035aba5-bc7e-4dc6-b071-e86b601c15af.dmp for [10280]
09:50:08     INFO -  Killing process: 10224
09:50:08     INFO -  Not taking screenshot here: see the one that was previously logged
09:50:08     INFO -  mozcrash Using Z:\task_1507109317\build\tests\bin\minidumpwriter.exe to write a dump to c:\users\genericworker\appdata\local\temp\tmp1kpmxa.mozrunner\minidumps\71ba8225-1620-4ff2-99dc-9c31e6ee2525.dmp for [10224]
09:50:08     INFO -  psutil found pid 10224 dead
09:50:08     INFO -  psutil found pid 10280 dead
09:50:08     INFO -  Killing process: 6580
09:50:08     INFO -  Not taking screenshot here: see the one that was previously logged
09:50:08     INFO -  mozcrash Using Z:\task_1507109317\build\tests\bin\minidumpwriter.exe to write a dump to c:\users\genericworker\appdata\local\temp\tmp1kpmxa.mozrunner\minidumps\c4f219e0-623f-4caa-a2d9-876446f1ad27.dmp for [6580]
09:50:09     INFO -  TEST-INFO | Main app process: exit 1
09:50:09     INFO -  Buffered messages finished
09:50:09    ERROR -  487 ERROR TEST-UNEXPECTED-FAIL | dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html | application terminated with exit code 1


:milan, would you know why this specific test is failing on the webgl tests when running on a windows10 vm?  it could be our configuration, the failure rate is very high, sO I want to move faster on this.
Flags: needinfo?(milan)
Whiteboard: [gfx-noted] → [gfx-noted][stockwell needswork]
Comment hidden (Intermittent Failures Robot)
Possibly the lower performance in VM gets us beyond the 16 contexts, we force dropping a WebGL context and things go bad?
Jeff, the max 16 per principal is hardcoded - is that something that is flexible or a hard limit?  Maybe send Joel a patch that changes that to a larger number and see if the intermittents go away?
Flags: needinfo?(milan) → needinfo?(jgilbert)
Comment hidden (Intermittent Failures Robot)

Comment 34

2 years ago
Pushed by jmaher@mozilla.com:
https://hg.mozilla.org/integration/mozilla-inbound/rev/18b0835d80c0
Disable dom/canvas/test/webgl-mochitest/test_pixel_pack_buffer.html on windows for frequent failures. r=me, a=test-only
please remember to enable this test when fixing the failure!
Keywords: leave-open
Whiteboard: [gfx-noted][stockwell needswork] → [gfx-noted][stockwell disabled]
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Comment hidden (Intermittent Failures Robot)
Flags: needinfo?(jgilbert)
You need to log in before you can comment on or make changes to this bug.