Windows 10 Fast ring 14986 was this actually tested at all?

Seems that the builds even the fast ring releases need to be tested more. I have been installing Fast Ring builds for years and the newly released build 14986 is the first one which refuses to install properly.

As I write this there is a 24 page thread with various suggestions from end users including registry hacks to disable Unified Update Platform (UUP) which is apparently in this build. Some have had success while other like myself simply see this build go through the process of installing the update only to have it revert back out. No error messages are displayed.

So far except for the initial post by a Microsoft Forum Moderator/ Support Engineer no followups over the last 3 days from anyone from Microsoft. They have asked for specific information regarding the existing boxes and configuration but that is as far as it has gotten.

I just hope that if this installation issue is related to UUP which appears at least in some cases it is, since registry hacks that bypass UUP, to allow the upgrade to work does not become another AUP.

I have seen this same thing happen with AUP on some systems where the upgrade fails, reverts back to the prior build and without any indication to the end user what the issue is. The system then tries to install the same update over and over again. 

Conversation 17 comments

  • 2015

    10 December, 2016 - 4:54 am

    <p>This is the first build to use an entirely new updating mechanism, of course there were bound to be issues. You are in the fast ring mate, and testing this stuff on a wide scale is <strong>exactly</strong> what the fast ring is for. &nbsp;With limited internal testing rings, it’s impossible to test on every configuration of PC/software. &nbsp;That’s what we’re here for.</p>
    <p>If it offends you, perhaps Fast isn’t the right ring for you. &nbsp;In fact, MS and media commentators specifically warn you not to run Fast ring on your primary PC or "production" machines. &nbsp;If you have an issue, just blow it away and re-install. Simple. No ifs, no buts.</p>
    <p>Me specifically, I accept the risk and run it on my primary gaming PC. I can’t get 14986 to even be offered to my PC, even after a re-install last night. I’m stuck on 14959 and Windows Update says I’m up to date &nbsp;Clearly they’re busy squashing bugs from all the feedback they’re getting, so I’m betting it’ll sort itself out soon.</p>

    • 3385

      11 December, 2016 - 1:22 am

      <blockquote><em><a href="#29745">In reply to </a><a href="../../../../users/Ryster">Ryster</a><a href="#29745">:</a></em></blockquote>
      <p>There is no way in heck I would put Fast Ring on a production machine. The machine is strictly for testing Fast Ring builds. Its a generic install of Windows 10 Fast Rings and that is all it is used for. The issue is that there are people over in the Insider Build MS Communities and they are without direction from any one from Microsoft. We are going around in circles. I don’t expect Microsoft to have a fix but at least have some sort of presence in the Insider Community.</p>
      <p>People keep saying run the Windows Update Troubleshooter. Well I have and each time I have run it finds the same problem and says it fixes it. Yet right after that I run it again and again it finds the same problem.</p>
      <p>I have just reset this box and wiped out all my files. If 14986 reverts back to 14971 on this box then there is definitely something in this build that needs fixing and hopefully Microsoft conveys to us Insiders over on their communities that they do indeed care and are aware of this issue.</p>
      <p>&nbsp;</p>

    • 1377

      Premium Member
      11 December, 2016 - 1:17 pm

      <p><em><a href="#29745">In reply to </a><a href="../../../../users/Ryster">Ryster</a><a href="#29745">:</a></em></p>
      <p>In my own case, 14986 doesn’t work well in VirtualBox VMs nor, from what I’ve read, in VMWare VMs. That seems to be wide spread, which does raise some concerns that MSFT didn’t test 14986 much in VMs.</p>
      <p>I can live with this. There was another flaky build sometime last Spring, and my response this time will be the same as then: give feedback (as long as the app don’t crash the VM), then cease using the VM until the next build.</p>

  • 2

    Premium Member
    10 December, 2016 - 8:46 am

    <p>I guess I’m of two minds here. On the one hand, this is the point of the Insider program, to test new things. On the other, yes, there are a lot of new features and changes in this one, and it’s a mess. I guess it all comes down to tolerance. But if you’re&nbsp;in the Fast ring, you should expect this: Fast ring testers see the builds before Microsoft employees, so the testing is happening. We’re doing it.</p>

    • 3385

      11 December, 2016 - 1:25 am

      <blockquote><em><a href="#29755">In reply to </a><a href="../../../../users/paul-thurrott">Paul Thurrott</a><a href="#29755">:</a></em></blockquote>
      <p>It would not be so bad if the installer / upgrade actually generates an error but it does not. Its the same issue I have seen with the Anniversary Update. They system balks restores the prior build and tells the end user nothing about what is happening. &nbsp;It then attempts to install the same update over and over again. It even shows that it has successfully installed the update in the Update History list when in fact it never did of course.</p>
      <p>&nbsp;</p>
      <p>&nbsp;</p>

      • 8630

        11 December, 2016 - 6:07 am

        <blockquote><em><a href="#29867">In reply to </a><a href="../../../../users/PcGuy8088">PcGuy8088</a><a href="#29867">:</a></em></blockquote>
        <p>You need to read the logs in the PANTHER FOLDER</p>

        • 250

          Premium Member
          11 December, 2016 - 7:12 am

          <blockquote><em><a href="#29880">In reply to </a><a href="../../../../users/jgnetworksecurity">jgnetworksecurity</a><a href="#29880">:</a></em></blockquote>
          <p>Thanks. I posted to this topic before seeing your comment but will check this out.</p>

        • 3385

          11 December, 2016 - 11:21 am

          <blockquote><em><a href="#29880">In reply to </a><a href="../../../../users/jgnetworksecurity">jgnetworksecurity</a><a href="#29880">:</a></em></blockquote>
          <p>Thanks will take a look there and see if I can find any indication what the heck is going on. Will take a look and see what happens on a clean install of 14986 after I did a reset of Windows10 on the test box and deleted all of my files on it.</p>

        • 1377

          Premium Member
          12 December, 2016 - 3:18 am

          <p><em><a href="#29880">In reply to </a><a href="../../../../users/jgnetworksecurity">jgnetworksecurity</a><a href="#29880">:</a></em></p>
          <p>Perhaps if it fails to install, but there’s nothing useful in C:\$Windows~BT\Sources\Panther about crashes after installation completes. What’s in the Event log is nearly useless.</p>

    • 1377

      Premium Member
      11 December, 2016 - 1:28 pm

      <p><em><a href="#29755">In reply to </a><a href="../../../../users/paul-thurrott">Paul Thurrott</a><a href="#29755">:</a></em></p>
      <p>I tolerate a lot in Insider builds, but I cease seeing the point to using Insider builds when they crash minutes apart.&nbsp;Because I don’t use Insider builds in production machines (or VMs), I don’t have to use it at all, and in this case I’ve now ceased using it, so also ceased being able to provide feedback, until the next build.</p>
      <p>When Insider builds are as flaky as 14986, I figure the build is so much of a PITA to use that MSFT gets much less useful feedback than they do from more robust builds. Maybe MSFT gets some value I can’t comprehend from 10 times as many IT CRASHES feedback posts and 1/10 as many features-related feedback posts. If not, 14986 couldn’t have been much use to MSFT.</p>

  • 5496

    10 December, 2016 - 12:47 pm

    <p>Insiders are BETA.</p>

    • 1377

      Premium Member
      11 December, 2016 - 1:19 pm

      <p><em><a href="#29785">In reply to </a><a href="../../../../users/lordbaal1">lordbaal1</a><a href="#29785">:</a></em></p>
      <p>Granted, but 14986 has an alpha-like stench. I can reliably crash it in a VirtualBox VM just by clicking the mouse inside the password box in the login screen.</p>

  • 5664

    Premium Member
    11 December, 2016 - 12:27 am

    <p>I had to go back to Release Preview honestly.</p>
    <p>I’m going to miss some the UI niceties that the current builds have, but I’m not going to deal with apps randomly starting in the background lol</p>

    • 124

      Premium Member
      12 December, 2016 - 4:21 pm

      <blockquote><em><a href="#29862">In reply to </a><a href="../../../../users/jimchamplin">jimchamplin</a><a href="#29862">:</a></em></blockquote>
      <p>Had that problem last build this build it went away.</p>

  • 250

    Premium Member
    11 December, 2016 - 7:09 am

    <p>This release installed on a Surface Book with no problems other than unexpectedly slow progress, but an old Surface Pro and a Dell XPS15 seemingly froze and refused to initiate the download process for hours. Those two managed to update themselves overnight, and I got the "Hi" screen when I woke them up in the morning. An old HP Stream 8 tablet that I almost never use but keep updating just to see if it has a breaking point ended up in a peculiar hybrid state. The "About" page shows that it is still on 14971, and that is also the last reported installed update. But in the Installed Updates section labeled "Feature Updates," a term I don’t completely understand, more than a dozen successful and failed installation attempts are recorded for 14986. On one occasion I saw a brief message about (I think) insufficient RAM and recovery media, but it blanked before I could find my glasses and read it in detail.</p>
    <p>I agree there is a lot going on in this update, but on the machines where I have&nbsp; explored it a little it seems pretty stable. The update process itself was the most irritating part of the experience.I suspect the 15000 and up releases will install more smoothly when we begin to see them.</p>

    • 124

      Premium Member
      12 December, 2016 - 4:24 pm

      <blockquote><em><a href="#29881">In reply to </a><a href="../../../../users/MacLiam">MacLiam</a><a href="#29881">:</a></em></blockquote>
      <p>Interesting I am running Fast Ring on my&nbsp;Surface Pro 1 and it seems to be fine.&nbsp;</p>

  • 5447

    11 December, 2016 - 11:15 pm

    <p>Weird how this goes so differently for different people;&nbsp;This build installed overnight without me even seeing it, and my SP4 has been&nbsp;awesome ever since.</p>

Windows Intelligence In Your Inbox

Sign up for our new free newsletter to get three time-saving tips each Friday

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Thurrott © 2024 Thurrott LLC