New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 672963 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner: ----
Closed: Dec 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Android
Pri: 1
Type: Bug-Regression



Sign in to add a comment

smoothness.key_mobile_sites_smooth.reference failure on chromium.perf Android Nexus 9

Project Member Reported by charliea@chromium.org, Dec 9 2016

Issue description

Revision range first seen: 437476-437518 (although I suspect that this has been a source of flake before)
Link to failing step log: https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus9%20Perf%20%283%29/builds/3739/steps/smoothness.key_mobile_sites_smooth.reference/logs/stdio

During the benchmark run, we got a message along the lines of "error: cannot bind to socket: Address already in use".

rnephew@, jbudorick@, do either of you have any idea what this might be related to?

Traceback (most recent call last):
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome/android_browser_backend.py", line 110, in Start
    self._port, remote_devtools_port)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/telemetry/telemetry/internal/platform/android_platform_backend.py", line 503, in ForwardHostToDevice
    self._device.adb.Forward('tcp:%d' % host_port, device_port)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/android/sdk/adb_wrapper.py", line 620, in Forward
    self._RunDeviceAdbCmd(cmd, timeout, retries)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/android/sdk/adb_wrapper.py", line 282, in _RunDeviceAdbCmd
    check_error=check_error)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/android/decorators.py", line 57, in timeout_retry_wrapper
    retry_if_func=retry_if_func)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/utils/timeout_retry.py", line 159, in Run
    error_log_func=error_log_func)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/utils/reraiser_thread.py", line 186, in JoinAll
    self._JoinAll(watcher, timeout)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/utils/reraiser_thread.py", line 158, in _JoinAll
    thread.ReraiseIfException()
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/utils/reraiser_thread.py", line 81, in run
    self._ret = self._func(*self._args, **self._kwargs)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/utils/timeout_retry.py", line 152, in <lambda>
    child_thread = reraiser_thread.ReraiserThread(lambda: func(*args, **kwargs),
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/android/decorators.py", line 47, in impl
    return f(*args, **kwargs)
  File "/b/rr/tmptVEZgp/w/src/third_party/catapult/devil/devil/android/sdk/adb_wrapper.py", line 258, in _RunAdbCmd
    args, output, status, device_serial)
AdbCommandFailedError: (device: HT4CTJT04201) adb forward --no-rebind tcp:53425 localabstract:chrome_devtools_remote: failed with exit status 1 and output:
- error: cannot bind to socket: Address already in use
- error: cannot bind to socket: Address already in use


If the test is disabled, please downgrade to Pri-2.

 
Cc: -charliea@google.com jbudorick@chromium.org rnep...@chromium.org
(Woops - didn't CC either of you in the initial bug.)

rnephew@, jbudorick@, do either of you have any idea what this might be related to?
/bump

Do either of you know what might be happening here?
There appears to be only one instances of this, so my guess is that there is a very rare race condition that got triggered. That error message means that you are trying to use a socket that is already in use. Is there any instances of this error on other bots?

The older errors for that benchmark appear to be different than this one.
https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus9%20Perf%20%283%29?numbuilds=200


Status: WontFix (was: Untriaged)
Ah crap - thanks for looking into this Randy! I'm going to close this until we see this error again, if that works for you.

(I think at some point we should start doing postmortems for each of these "rare failures" that really do add up given the total number of benchmarks we have, but I think at this point we have higher-priority fish to fry.)

Sign in to add a comment