Justin is already working on moving this into a new D-Bus service (tracked at  issue 692246 ). I'll let him decide if he wants to dupe this bug into that one, or keep this bug open to track moving the new service to a different process, or assign it back to me for that part, or what have you.

 Issue 703205  has been merged into this issue.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/4bb59edb9d55aa56f4e79236e86f081b92c686ef

commit 4bb59edb9d55aa56f4e79236e86f081b92c686ef
Author: teravest <teravest@chromium.org>
Date: Wed May 24 02:56:39 2017

dbus: Provide LivenessService.CheckLiveness

CheckLiveness is being moved to its own service so that it can be moved
outside the browser process for mus+ash.

BUG= 644322 
TEST=Manual dbus-send test on LibCrosService and LivenessService

Review-Url: https://codereview.chromium.org/2904763002
Cr-Commit-Position: refs/heads/master@{#474140}

[modify] https://crrev.com/4bb59edb9d55aa56f4e79236e86f081b92c686ef/chrome/browser/chromeos/chrome_browser_main_chromeos.cc
[modify] https://crrev.com/4bb59edb9d55aa56f4e79236e86f081b92c686ef/chromeos/dbus/services/liveness_service_provider.cc
[modify] https://crrev.com/4bb59edb9d55aa56f4e79236e86f081b92c686ef/chromeos/dbus/services/liveness_service_provider.h

I think that session_manager still needs to be updated to use LivenessService (see src/platform2/login_manager/liveness_checker_impl.cc), followed by the method being deleted from LibCrosService. Justin, feel free to reassign to me or Lann if you're busy with other work at the moment.

I'm updating session_manager.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/platform2/+/776bfc6faaa8c06fb29142331f5b767c9a453519

commit 776bfc6faaa8c06fb29142331f5b767c9a453519
Author: Daniel Erat <derat@chromium.org>
Date: Sat Mar 31 21:13:57 2018

login: Use org.chromium.LivenessService to ping Chrome.

Make session_manager call the org.chromium.LivenessService
D-Bus service rather than org.chromium.LibCrosService to
ping Chrome. The new service was added long ago;
session_manager just hasn't been updated to use it.

BUG= chromium:644322 
TEST=unit tests pass; also verified that pings are sent and
     responses received on a ToT build

Change-Id: I973f17180e6c0ea465043ee119efb28315fa97cd
Reviewed-on: https://chromium-review.googlesource.com/987204
Commit-Ready: Dan Erat <derat@chromium.org>
Tested-by: Dan Erat <derat@chromium.org>
Reviewed-by: Mattias Nissler <mnissler@chromium.org>

[modify] https://crrev.com/776bfc6faaa8c06fb29142331f5b767c9a453519/login_manager/session_manager_service.cc
[modify] https://crrev.com/776bfc6faaa8c06fb29142331f5b767c9a453519/login_manager/docs/chrome_hang_detection.md
[modify] https://crrev.com/776bfc6faaa8c06fb29142331f5b767c9a453519/login_manager/liveness_checker_impl.cc
[modify] https://crrev.com/776bfc6faaa8c06fb29142331f5b767c9a453519/login_manager/liveness_checker_impl.h
[modify] https://crrev.com/776bfc6faaa8c06fb29142331f5b767c9a453519/login_manager/liveness_checker_impl_unittest.cc

Deprecating Proj-Mustash-Mus-WS label in favor of Components.

I should probably move this into ash.

Hrm. Are you sure ash is the right place for this?

With OOPAsh it seems like either ash or chrome could hang. (Window server as a separate process doesn't exist in our current plans, ditto for separate service launcher.)

I wonder if session manager should watch ash, and ash should watch chrome. Or maybe session manager should watch both? Or just chrome and chrome watches ash? Not sure.

I think that it should live in whichever process is at the root of the process tree that gets kicked off by session_manager. It should furthermore be long-lived (i.e. if not needing to run Chrome-the-browser all the time is still a goal, it shouldn't be there). Does that make ash the best fit for this?

I'm going to mark this as fixed now that it's in its own D-Bus service. I'm using  issue 843392  to track moving these providers into the right places.