All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=840768

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0022d38fc84ae99ed577495c5e82f3765910cd2949875800a9770b782f814ddb


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

android-nexus5
android-nexus5X
android-one

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/13de5bdbc40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/13de5bdbc40000

Implement process isolation for chrome-error:// documents. by nasko@chromium.org
https://chromium.googlesource.com/chromium/src/+/d83b571dc700010f144b68fe3d8628923c008773

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

This is worrisome that we regress benchmarks when we change error page behavior. It will be good to ensure we aren't testing navigations to non-existing pages expecting them to be successful.

For the time being, this behavior has been disabled, so I expect the metrics will get back to normal.

+tdresser: looks like we should take a look at long.running.tools.gmail.foreground/background stories, the bug indicates they may be producing error pages.

 Issue 862228  has been merged into this issue.

 Issue 862379  has been merged into this issue.

Issue 859126 has been merged into this issue.

There is not much for me to do on this bug. Error pages will be isolated going forward and the reason we are seeing a perf regression here is that the perf tests are hitting errors. Therefore I'm unassigning myself and marking it as available.

 Issue 846082  has been merged into this issue.

Issue 840765 has been merged into this issue.

Assigning to tdresser for triage: The problem here is that these benchmarks are showing error pages. We'll need to put this on the list of things to fix for system health benchmarks.

Ping! This bug has been flagged as stale by Chrome Speed Metrics bug triage, and it has an owner. Any update on this issue?