locked
Issue With Deploying Windows 10 Feature Update 1709 To 1607 Clients Using WSUS RRS feed

  • Question

  • We're currently running Windows 10 Enterprise x64 1607 on our PCs.  We have a Windows 2012 R2 WSUS server which has successfully been deploying Windows updates to the Windows 10 machines.  We now would like to test deployment of the 1709 feature update to a test group.  We enabled feature updates within WSUS, approved the 1709 update for a test group, and added the .ESD MIME type to IIS.

    We're seeing an odd behavior where the initial update to 1709 will fail but then will succeed three days later.  When it initially fails, we can see a C:\$WINDOWS.~BT directory is created but it's clearly incomplete as its only 60 mb in size and the only subdirectory is Sources. The random folder created for the 1709 ESD is in SoftwareDistribution\Download and it contains the following files:

    16299.125.171213-1220.rs3_release_svc_refresh_CLIENTBUSINESS_VOL_X64FRE_en-us.esd - 3.2 gb

    state - 1 kb

    WindowsUpdateBox - 19.2 mb

    The bluebox log shows SetupHost.exe reporting a 0x0 result code:

    2018-03-05 22:21:28: Searching for ESD files...
    2018-03-05 22:21:28: Launching process: [C:\$WINDOWS.~BT\Sources\SetupHost.Exe] with command-line [/PreDownload /Update /Quiet  /progressCLSID 23b9ae65-0cf4-48d8-a0b2-0af232d7e403 /ReportId {71854F91-C7C7-4CDE-9394-9DDB5CCFF443}.200 "/ClientId" "0d2bb21c-8505-47d2-8f86-5c200b25ba9b" "/CorrelationVector" "GGAb5u8QHk2NJIDf.50.1.2" ] in Session: [-1]
    2018-03-05 22:21:28: Launching: [C:\$WINDOWS.~BT\Sources\SetupHost.Exe] [/PreDownload /Update /Quiet  /progressCLSID 23b9ae65-0cf4-48d8-a0b2-0af232d7e403 /ReportId {71854F91-C7C7-4CDE-9394-9DDB5CCFF443}.200 "/ClientId" "0d2bb21c-8505-47d2-8f86-5c200b25ba9b" "/CorrelationVector" "GGAb5u8QHk2NJIDf.50.1.2" ] from [C:\$WINDOWS.~BT\Sources]!
    2018-03-05 22:21:28: Waiting for process events...
    2018-03-05 22:21:48: Process exit code: [0x0]
    2018-03-05 22:21:48: LaunchProcessInSession returned: [0x0]
    2018-03-05 22:21:48: Checking cleanup registry value...
    2018-03-05 22:21:48: Cleanup value missing... assuming no cleanup.
    2018-03-05 22:21:48: Checking SetupHost result value...
    2018-03-05 22:21:48: SetupHost result value: [0x0]
    2018-03-05 22:21:48: Skipping cleanup.

    However the update fails to install and the Windows Update Client reports the following error to the event log:

    Installation Failure: Windows failed to install the following update with error 0x8024200D: Feature update to Windows 10 (business editions), version 1709, en-us.

    I looked up the 0x8024200D error and found the following:

    0x8024200D WU_E_UH_NEEDANOTHERDOWNLOAD The update handler did not install the update
    because it needs to be downloaded again.

    So it appears the initial download fails or is incomplete, and perhaps this explains why a few days later the download retries and successful.  Once this happens, the upgrade can proceed.  The question is how can I go about troubleshooting why this is happening as I've only seen the problem with deployment of a feature update?



    • Edited by jdbst56 Tuesday, March 6, 2018 11:37 PM
    Tuesday, March 6, 2018 11:36 PM

All replies

  • Hi,

    As you said, 0x8024200D means:The update handler did not install the update because it needs to be downloaded again.
    It may means that you only need to try again to solve it.

    I did not find a specific reason from your log.

    Maybe you could refer to the similar issue to get more ideas:
    https://www.urtech.ca/2016/11/solved-windows-update-error-31-windows-update-error-8024200d-feature-update-windows-10-pro-sits-downloading-0/.


    Please remember to mark the replies as answers if they help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com.

    Wednesday, March 7, 2018 9:10 AM
  • Thanks for your reply.  I had already added the .esd MIME type to IIS.  What I don't understand is the consistency in that the first time that the download is attempted, it is apparently incomplete so the install fails.  Yet it can try it again the next day and it works.  The behavior is problematic because we don't have certainty that the update will deploy when we expect it to.  Any ideas on other logs to check or other diagnostic steps to take?  I have a Premier support case open for this issue as well and we've haven't made any progress either.
    Wednesday, March 7, 2018 11:05 PM
  • I am experiencing the same issue with a 2016 WSUS server with all the latest updates too.  The majority of our systems are on 1703 and 1709.  I have tried updgrading both 1703 (to 1709) and 1709 (to 1803) and experience the same results.  Seems the first time the workstation sees the update is there it will try to download it, it appears to download the update to 100% (over a T1) in like 2 minutes and then fails to install resulting in error code 0x80024200D.  This error indicates the download was not complete.  I then appears to try again and takes longer to download but then fails again.  Seems like it will do this a few times and eventually complete.  Not sure why but it never seems to download successfully the first time.

    I currently have the Delivery Optimization Max download speed set to 50 KB/S in GPO and also enabled DoNotConnectToWindowsUpdateInternetLocations in GPO too.  Seems like if I don't add these to my GPO the download kills our T1 and it will bypass wsus and go to windows update servers for updates.

    Hopefully there is some reason as to why this happening and if there is hopefully MS will address it.

    If anyone else has a resulution for us, please let us know.

    Thanks!

    Thursday, May 10, 2018 5:37 PM