graphics_Idle failing with LoginException: Failed to restart browser in guest mode error. |
||||||
Issue descriptionhttps://wmatrix.googleplex.com/testrun/unfiltered?test_ids=366136635 https://wmatrix.googleplex.com/testrun/unfiltered?test_ids=366105962 10/19 04:08:52.301 DEBUG| cros_interface:0067| > stdout=[], stderr=[Traceback (most recent call last): File "/usr/local/autotest/bin/screenshot.py", line 21, in <module> image = crtcScreenshot(args.crtc) File "/usr/local/autotest/cros/graphics/drm.py", line 427, in crtcScreenshot "Unable to take screenshot. There may not be anything on the screen.") RuntimeError: Unable to take screenshot. There may not be anything on the screen. ] 10/19 04:08:52.302 ERROR| browser:0062| Failure while starting browser backend. Traceback (most recent call last): File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/browser/browser.py", line 55, in __init__ self._browser_backend.Start() File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_backend.py", line 157, in Start 'Failed to restart browser in guest mode (pid %d).' % pid) File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_backend.py", line 274, in _RaiseOnLoginFailure raise exceptions.LoginException(error) LoginException: Failed to restart browser in guest mode (pid 5670). 10/19 04:08:52.311 ERROR| chrome:0182| Timed out logging in, tries=2, error=LoginException('Failed to restart browser in guest mode (pid 5670).',) 10/19 04:08:52.323 DEBUG| base_utils:0185| Running 'logger "autotest finished iteration /usr/local/autotest/results/default/graphics_Idle/sysinfo/iteration.1"' 10/19 04:08:52.334 WARNI| test:0606| Autotest caught exception when running test: Traceback (most recent call last): File "/usr/local/autotest/common_lib/test.py", line 600, in _exec _call_test_function(self.execute, *p_args, **p_dargs) File "/usr/local/autotest/common_lib/test.py", line 810, in _call_test_function raise error.UnhandledTestFail(e) UnhandledTestFail: Unhandled LoginException: Failed to restart browser in guest mode (pid 5670). Traceback (most recent call last): File "/usr/local/autotest/common_lib/test.py", line 804, in _call_test_function return func(*args, **dargs) File "/usr/local/autotest/common_lib/test.py", line 461, in execute dargs) File "/usr/local/autotest/common_lib/test.py", line 347, in _call_run_once_with_retry postprocess_profiled_run, args, dargs) File "/usr/local/autotest/common_lib/test.py", line 376, in _call_run_once self.run_once(*args, **dargs) File "/usr/local/autotest/tests/graphics_Idle/graphics_Idle.py", line 38, in run_once arc_mode=arc_mode): File "/usr/local/autotest/common_lib/cros/chrome.py", line 174, in __init__ self._browser = browser_to_create.Create(finder_options) File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_finder.py", line 50, in Create self._credentials_path) File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_with_oobe.py", line 14, in __init__ backend, platform_backend, credentials_path) File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/browser/browser.py", line 55, in __init__ self._browser_backend.Start() File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_backend.py", line 157, in Start 'Failed to restart browser in guest mode (pid %d).' % pid) File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/cros_browser_backend.py", line 274, in _RaiseOnLoginFailure raise exceptions.LoginException(error) LoginException: Failed to restart browser in guest mode (pid 5670).
,
Oct 19 2016
This is affecting our M54 Stable candidate. Can we triage asap to determine if this should block the release?
,
Oct 19 2016
The problem seems to have started with R54-8743.1.0, aka at branch point. :-(
,
Oct 19 2016
,
Oct 20 2016
I tried briefly to repro this on samus (not quite matching build), but didn't have a repro. Best to install latest M54.
,
Oct 20 2016
,
Oct 20 2016
I believe this is the same as https://bugs.chromium.org/p/chromium/issues/detail?id=631640 You can read the bug for the details, but the summary is that the bug started happening on Jul 21, and the symptom is that there's occasionally a long delay (sometimes 30 sec) for chrome to restart in guest mode, because the DBUS restart signal is not received by session manager. We were also seeing this delay after enterprise enrollment, where the browser needs to be restarted by the session manager with new flags. This bug doesn't happen on TOT or R55, so somewhere along the line something landed to fix it. R55: https://wmatrix.googleplex.com/unfiltered?hide_missing=True&tests=login_CryptohomeIncognito&days_back=30&releases=55 R54: https://wmatrix.googleplex.com/unfiltered?hide_missing=True&tests=login_CryptohomeIncognito&days_back=30&releases=54 The rate of failure seems pretty low - 80 boards that ran for ~26 days produced 2 failures, so about 0.1%. We also bumped the timeout from 15 sec to 60 sec along the way; it's still 15 sec in R54, so increasing the timeout will improve the test failure rate. Unfortunately the timeout is in catapult, and we don't have a way of creating a R54 catapult branch.
,
Oct 20 2016
,
Oct 20 2016
For anyone planning to read through issue 631640, save yourself some trouble and start at http://crbug.com/631640#c98. I believe the underlying issue is that one or more events triggered by the login-prompt-visible Upstart signal are taking 30+ seconds to finish. If we still have complete Upstart logs from occurrences of this problem, it's hopefully possible to see what's taking so long.
,
Oct 20 2016
Would the fix for these issues be in test code or code shipping to users? The catapult change is test only. So basically sometimes we will let users wait for login. Maybe then that is acceptable and should not block release?
,
Oct 20 2016
Yup, that's what I think - sometimes there's a long delay before the guest session starts. And maybe that's not worth blocking the release over.
,
Oct 20 2016
+1, Removing RBS as per above comments Should we set as a Dup of issue 631640?
,
Oct 20 2016
|
||||||
►
Sign in to add a comment |
||||||
Comment 1 by ihf@chromium.org
, Oct 19 2016