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

Issue 878175 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Closed: Aug 28
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: ----



Sign in to add a comment

start_with_url.warm.startup_pages/about:blank failing in performance_test_suite failing on chromium.perf/Android Nexus5 Perf

Project Member Reported by sheriff-...@appspot.gserviceaccount.com, Aug 27

Issue description

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/10c4b34d640000
Components: Speed>Benchmarks
Owner: pasko@chromium.org
Status: Assigned (was: Available)
Bisect can't figure this out. The flaky error is with the trace events:

  ImportTraces at /b/swarming/w/ir/third_party/catapult/telemetry/telemetry/timeline/model.py:139
    importer.ImportEvents()
  ImportEvents at /b/swarming/w/ir/third_party/catapult/telemetry/telemetry/timeline/trace_event_importer.py:251
    self._ProcessInstantEvent(event)
  _ProcessInstantEvent at /b/swarming/w/ir/third_party/catapult/telemetry/telemetry/timeline/trace_event_importer.py:196
    args=event.get('args'))
  BeginSlice at /b/swarming/w/ir/third_party/catapult/telemetry/telemetry/timeline/thread.py:112
    'Slices must be added in increasing timestamp order')
ValueError: Slices must be added in increasing timestamp order

https://chrome-swarming.appspot.com/task?id=3f95c63e4f4b7210&refresh=10&show_raw=1

Pasko: can you look at this?
Status: WontFix (was: Assigned)
I'm not having enough cycles to investigate it fully.

What I know:
1. this error happened before with my tweaking of TBMv1 benchmarks
2. also got 'Fixed' once in  crbug.com/321529 
3. reappeared years later in another context: crbug.com/628347
4. we are moving to TBMv2, where this would not happen
5. the problem may appear because there is suddenly more concurrency somewhere (more processes with site isolation experiments? there was nothing obvious in fieldtrial_testing_config.json)

Given [4] I would prefer to focus on the new world of TBMv2. Another option: if pinpoint cannot grasp the flake, maybe increase repeat count? (I'm not familiar with the magic)
Doubledown on TMBv2 SGTM. I will disable these failing tests then.
Project Member

Comment 6 by bugdroid1@chromium.org, Aug 28

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/d4d7c171823633b83924e2702d89968d37e4d370

commit d4d7c171823633b83924e2702d89968d37e4d370
Author: nednguyen <nednguyen@google.com>
Date: Tue Aug 28 15:21:49 2018

Disable failing start_with_url.warm.startup_pages/* tests

NOTRY=true
TBR=pasko@chromium.org

Bug:  878175 
Change-Id: I399e3e0d905ade42d67985a1808f4b7a7ccd541a
Reviewed-on: https://chromium-review.googlesource.com/1194242
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#586713}
[modify] https://crrev.com/d4d7c171823633b83924e2702d89968d37e4d370/tools/perf/expectations.config

Project Member

Comment 7 by bugdroid1@chromium.org, Aug 28

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/4d366ff3657bd0a81f1e42787f1df1ffa863f619

commit 4d366ff3657bd0a81f1e42787f1df1ffa863f619
Author: nednguyen <nednguyen@google.com>
Date: Tue Aug 28 16:32:48 2018

Also disable failing cold startup pages

NOTRY=true

TBR=pasko@chromium.org

Bug:  878175 
Change-Id: I759a5d73e4ca9028e05dfba48771971416531680
Reviewed-on: https://chromium-review.googlesource.com/1194251
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#586737}
[modify] https://crrev.com/4d366ff3657bd0a81f1e42787f1df1ffa863f619/tools/perf/expectations.config

Issue 878900 has been merged into this issue.

Sign in to add a comment