First failure build:
https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Trusty%20Leak/builds/9660

Failed tests:
http/tests/navigation/navigation-interrupted-by-fragment.html
virtual/mojo-loading/http/tests/navigation/navigation-interrupted-by-fragment.html 
virtual/stable/http/tests/navigation/navigation-interrupted-by-fragment.html

Camille, please triage, thanks.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/4e89de7f5dafce2bcd9b862ddb545c4f160912e8

commit 4e89de7f5dafce2bcd9b862ddb545c4f160912e8
Author: Owen Min <zmin@chromium.org>
Date: Fri Sep 15 19:53:40 2017

Disable failed leak test navigation-interrupted-by-fragment.html

Bug: 765721
Change-Id: I8494f976c2c5984fd8dc6352a0364ea4afb5b292
Tbr: jam@chromium.org
Reviewed-on: https://chromium-review.googlesource.com/668695
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Reviewed-by: Owen Min <zmin@chromium.org>
Commit-Queue: Owen Min <zmin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#502334}
[modify] https://crrev.com/4e89de7f5dafce2bcd9b862ddb545c4f160912e8/third_party/WebKit/LayoutTests/LeakExpectations

-Sheriff-Chromium as test is disabled.

I looked at the test, and I'm pretty sure that what it is testing doesn't happen with PlzNavigate. It tests that a fragment navigation will cancel an ongoing slow navigation. However this is not the case if the fragment navigation is not user initiated like in the test. I suspect the test finishes too early, making it look like the navigation stopped even though it did not.

Any updates? The comment in third_party/WebKit/LayoutTests/LeakExpectations says "Not revert suspected CL as jam@ request, expected to be fixed soon." but it's been 7 months.