Closed Bug 1046635 Opened 11 years ago Closed 10 years ago

Intermittent Gaia ui test fail with timed out after 1000 seconds of no output

Categories

(Firefox OS Graveyard :: Gaia, defect)

Product:
Firefox OS Graveyard
Component:
Gaia
Platform:
x86
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED WORKSFORME

People

(Reporter: cbook, Unassigned)

References

(
URL
)

Details

(Keywords: intermittent-failure)

b2g_macosx64 mozilla-inbound opt test gaia-ui-test on 2014-07-31 02:43:48 PDT for push 0893f2ef4b9a slave: talos-mtnlion-r5-046 https://tbpl.mozilla.org/php/getParsedLog.php?id=44958163&tree=Mozilla-Inbound Automation Error: mozprocess timed out after 1000 seconds running ['/builds/slave/talos-slave/test/build/venv/bin/python', '-u', '/builds/slave/talos-slave/test/gaia/tests/python/gaia-ui-tests/gaiatest/cli.py', '--restart', '--timeout=10000', '--type=b2g', '--testvars=/builds/slave/talos-slave/test/gaia/tests/python/gaia-ui-tests/gaiatest/testvars.json', '--profile=/builds/slave/talos-slave/test/gaia/profile', '--symbols-path=https://ftp-ssl.mozilla.org/pub/mozilla.org/b2g/tinderbox-builds/mozil... [exceeded max length] timed out after 1000 seconds of no output
Hi Zac, When dealing with bug 1016184, I encounter the error "mozprocess timed out after 1000 seconds" on try server [1]. My patches somehow trigger about 20% random errors on Gip on "B2G Desktop OS X Opt", but it is fine on "B2G Desktop Linux x64 Opt". This timeout happens at different test cases in different Gip runs randomly. Here are my questions. 1) Is it possible gecko crashed at previous test so that marionette cannot connect to server in the next test? I saw the success gecko log print "accepted connection on 127.0.0.1:49698", but the failed log does not have this line. Will marionette show error message correctly if gecko is crashed? 2) Is it possible to dump gecko log for every test cases instead of only the last one to get more information? I cannot reproduce this problem locally, so I am stuck here. Any advice will be appreciated. Thank you. [i] https://tbpl.mozilla.org/?tree=Try&rev=fbc6a3ec7809
Flags: needinfo?(zcampbell)
Hi, it sounds less like a crash and more like it froze altogether. If it crashes then usually there'll be more attempts to run tests. but when the b2g process freezes it just hangs like this. I'm not sure what to do aside from downloading the binary and running locally - hopefully you have a mac! Do you think your patch might cause a race or failure in the startup of b2g? Re 2), that is an outstanding problem with the logs ( https://bugzilla.mozilla.org/show_bug.cgi?id=1028202 )
Flags: needinfo?(zcampbell)
Zac, thanks for your prompt reply. My patches only turn on the touch caret preference with fix to existing test cases. I've tried to only flip the preference. It still has chances to fail [1]. The issue must already in the codebase. https://tbpl.mozilla.org/?tree=Try&rev=8d622c3d3fd6
Here's more Gip fail with some touch caret debug log. Before "timed out after 1000 seconds of no output" happening, here's always an "SSLError: The read operation timed out" in the setup phase. I'm not sure they are relevant. https://tbpl.mozilla.org/?tree=Try&rev=cbd262539996
(In reply to Ting-Yu Lin [:TYLin] (UTC+8) from comment #9) > Here's more Gip fail with some touch caret debug log. > > Before "timed out after 1000 seconds of no output" happening, here's always > an "SSLError: The read operation timed out" in the setup phase. I'm not sure > they are relevant. > > https://tbpl.mozilla.org/?tree=Try&rev=cbd262539996 Mmm... the SSLError also happens in many green Gip.
Depends on: 1050090
I'm out of luck today. Still cannot reproduce this issue in talos-mtnlion-r5 slave.
Comment #22 is for a completely different test suite.
Inactive; closing (see bug 1180138).
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.