FYI, I have read multiple Google search threads regarding disabling HW acceleration and various other hacks to no avail.  Again, this worked fine before so this isn't some new hardware issue.  Either a Windows KB is preventing Chrome from launching or Chrome is doing something odd with the Windows Enterprise SKU..?

Please let me know if there's anything else you need to help debug this issue. 

To be clear about the dump file:  I set my Chrome browser to a default web page (e.g. http://www.cnn.com) and per the status bar (at the bottom) it starts to load up (usually ad crap) but that's when it stalls out... so I broke into the Chrome process at that point.  I've done this repro several times and the stalled stack trace is the same each time.  

PS.  I've disabled our corporate AV software with no effect.  I've tried on the corporate WiFi (internet proxy) and my home WiFi with no luck.  Pretty certain it's a software bug at this point.  One thing I'd be curious to try is a clean install of Win 8.1 retail to see if it's a configuration/group policy issue with our Enterprise/Corporate image.

In addition to the above comment # 3, could you please try deleting the profiles if it might have corrupted and try it without having any extension in it(if you have extensions in your account, try disabling them where u have the access to it).
If you have a crash Id, would help us to further triage it.

The problem is that DsBind is taking a terribly long time to timeout. This was happening on some of our bots in issue 555131, but it was resolved via configuration fix rather than a code fix. If the call to DsBind is truly needed, then it must be done on a thread in the blocking pool rather than done on the FILE thread (see ChromeBrowserPolicyConnector::CreatePlatformProvider).

Thank you for providing such detailed info, cn.

Thanks George.  It's about time my 19 years of testing experience comes in handy. ;)

Let me know if you guys need any more assistance or with testing a fix.  

PS.  I had a feeling it had something to do with our Enterprise SKU. ;)

I would suggest we stop collecting these UMA stats for now. I think we extracted all the needed info from it by now. This was useful to guide our decisions on transitioning to Registry reading again provided the right environment is present and now we don't really gain any more information out of this.

What do you think Matt, Saswat?

CL is in review : https://codereview.chromium.org/2157143003

#7 - SGTM

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f34984bbeee68c231b5ad17464ad8ea41f1f0d09

commit f34984bbeee68c231b5ad17464ad8ea41f1f0d09
Author: pastarmovj <pastarmovj@chromium.org>
Date: Tue Jul 19 20:36:41 2016

Remove slow test for connecting to the AD DC server.

DsBind was slowing down Chrome startup blocking the file thread for
up to a few minutes on some machines. This check is not really delivering
any more useful information to us so it can be safely removed.

BUG= 621270 
TEST=none

Review-Url: https://codereview.chromium.org/2157143003
Cr-Commit-Position: refs/heads/master@{#406373}

[modify] https://crrev.com/f34984bbeee68c231b5ad17464ad8ea41f1f0d09/components/policy/core/common/policy_loader_win.cc
[modify] https://crrev.com/f34984bbeee68c231b5ad17464ad8ea41f1f0d09/tools/metrics/histograms/histograms.xml

I don't think there is much to be verified here. However if Chrome still loads slow in some cases we should open up another bug and see what hangs then. At least the call to DsBind is gone now.

I think we should merge that back to 53 as it is practically zero risk but has the potential to make chrome start faster for up to 1mil daily users (this is the number of failing DsBinds which I assume can take long time to timeout for example when the user is on a corp laptop outside of the corp network).

pastarmovj, i'll definitely try the 54.x dev build when it's available.  Regarding #12, I was I also seeing issues while on corp network as well. My debugging was off the corporate network however.  Thanks for the quick fix!

Your change meets the bar and is auto-approved for M53 (branch: 2785)

Unable to repro this issue on Windows 7 for Google Chrome Canary Version - 54.0.2803.0 

Tested on Corp Network.

Screen-recording is attached.

Thank you.

Deleted: 621270.mp4 2.0 MB

	621270.mp4 2.0 MB View Download

So far so good on 54.0.2803 Canary 64-bit.  Will test tomorrow on the corporate network and then back at home.  Thanks guys!

Thanks for all the testing cnick66qn0! 

This change is now merged into 53 as well so it won't take so long until it hits the stable channel as well.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f8af32b566032c4b59ddfc97508d3232cf8aacd4

commit f8af32b566032c4b59ddfc97508d3232cf8aacd4
Author: Julian Pastarmov <pastarmovj@chromium.org>
Date: Thu Jul 21 08:32:05 2016

Remove slow test for connecting to the AD DC server.

DsBind was slowing down Chrome startup blocking the file thread for
up to a few minutes on some machines. This check is not really delivering
any more useful information to us so it can be safely removed.

BUG= 621270 
TEST=none

Review-Url: https://codereview.chromium.org/2157143003
Cr-Commit-Position: refs/heads/master@{#406373}
(cherry picked from commit f34984bbeee68c231b5ad17464ad8ea41f1f0d09)

Review URL: https://codereview.chromium.org/2167103002 .

Cr-Commit-Position: refs/branch-heads/2785@{#260}
Cr-Branched-From: 68623971be0cfc492a2cb0427d7f478e7b214c24-refs/heads/master@{#403382}

[modify] https://crrev.com/f8af32b566032c4b59ddfc97508d3232cf8aacd4/components/policy/core/common/policy_loader_win.cc
[modify] https://crrev.com/f8af32b566032c4b59ddfc97508d3232cf8aacd4/tools/metrics/histograms/histograms.xml