Here's a link to the flakiness dashboard for this platform: https://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=rendering.mobile&builder=chromium.perf%3AAndroid%20Nexus5X%20Perf

I'll try to kick off a bisect on techcrunch, which looks like it started failing around 5/29. There are some question marks in the range on the flakiness dashboard, but I'm not 100% sure what that means.

Youtube appears to have been failing for a really long time (as far back as we have history - 4/4), so I'm just going to disable that. 

Woops - I didn't notice the distinction between youtube and youtube desktop. youtube_desktop has been failing for a really long time. youtube appears to have been passing around 561968 but appears to have started failing by 562744 (around 5/25). techcrunch appears to have been passing around 563959 and started failing by r566933. I'll kick off bisects on youtube (non-desktop) and techcrunch and disable all 3.

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/11ede6a4a40000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1394351b240000

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/7b99bbe2d20898a5a9a415aba20cae0b743f8a40

commit 7b99bbe2d20898a5a9a415aba20cae0b743f8a40
Author: Charlie Andrews <charliea@chromium.org>
Date: Mon Jul 02 17:22:18 2018

Disable rendering.mobile stories failing on Nexus 5X

NOTRY=true
TBR=nednguyen@chromium.org

Bug:  859575 
Change-Id: I81e1c060bf93f8d30c56a4cc8563c6f828b524fb
Reviewed-on: https://chromium-review.googlesource.com/1122970
Commit-Queue: Charlie Andrews <charliea@chromium.org>
Reviewed-by: Charlie Andrews <charliea@chromium.org>
Cr-Commit-Position: refs/heads/master@{#571930}
[modify] https://crrev.com/7b99bbe2d20898a5a9a415aba20cae0b743f8a40/tools/perf/expectations.config

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/1394351b240000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/11ede6a4a40000

Finch: Add configs for waterfall testing. by farahcharab@google.com
https://chromium.googlesource.com/chromium/src/+/a60f7661bc3b3b2a9d8656e6b38d634c209538f1
0 → 0.12 (+0.12)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Farah, it looks like your config push to the waterfall significant increased the flakiness of a couple of our rendering tests (specifically, youtube and techcrunch). https://paste.googleplex.com/6537938573721600 is a link to the full failure callstack for a techcrunch failure. Could you please figure out why this led to an increase in flakiness, or route this bug to someone who can?

The NextAction date has arrived: 2018-07-06

It seems that it's started crashing. Is there a way to get a stacktrace out of this?

See comment 9 above: I think https://paste.googleplex.com/6537938573721600 should work

Re #13: Unfortunately, that doesn't seem to be a full failure callstack -- it has python one and it shows that the renderer has crashed, but it doesn't have the renderer   stacktrace for the crash.

I'm not really sure what next step might help debug the problem, especially given that I talked to altimin@ offline and he already tried unsuccessfully to reproduce the problem. nednguyen@, any ideas?

altimin@ can you reproduce the crash locally? 

I talked with him earlier and he said he was unable to :/

Did he tried to reproduce using the isolate? The local build maybe different from the bot build.

go/reproduce-telemetry

+Juan in case he has some thoughts about how to reproduce this problem

It looks that it's crashing. We need to look into this. farahcharab@, you mentioned that you were able to get a stacktrace for the crash?

rendering.mobile/youtube has been re-recorded as rendering.mobile/youtube_2018 and is now re-enabled.