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

Issue 827077 link

Starred by 1 user
Status: Fixed
Owner:
█ phoglund@chromium.org
Last visit > 30 days ago
Closed: Apr 2018
Cc:
█ sakal@chromium.org
nisse@chromium.org
oprypin@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 828310


Sign in to add a comment

1%-146.4% regression in webrtc_perf_tests at 22582:22637

Project Member Reported by nisse@chromium.org, Mar 29 2018 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Mar 29 2018 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=827077

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=852b7111c91f31359b96207ac746e1cd24bba48b02f203ccf7d791b20ad96c7e


Bot(s) for this bug's original alert(s):

webrtc-android-tests-nexus5-kitkat

Comment 2 by nisse@chromium.org, Mar 29 2018

Cc: sakal@chromium.org
Owner: phoglund@chromium.org
These alerts show some performance changes on webrtc-android-tests-nexus5-kitkat, between commit #22584 and #22636 (and we also have some earlier change at #22520).

I see no obvious changes to cause this. Have there been any changes on the test devices?

Comment 3 by phoglund@chromium.org, Apr 3 2018

Cc: oprypin@chromium.org
Hmm, it looks like our perf waterfall is misconfigured. The bot you point to has five attached devices, and it should be one (otherwise results are random on which device happens to get selected). I will follow up with the lab engineers. As far as we know they have plugged in devices for us, which is nice in general but not what we want for these bots.

Comment 4 by phoglund@chromium.org, Apr 3 2018

Blockedon: 828310

Comment 5 by phoglund@chromium.org, Apr 5 2018

Status: Fixed (was: Untriaged)
The waterfall configuration has been corrected.
►

Sign in to add a comment