Windows update constantly fails during the update process and rolls back.

T

Tinribs1

Hi,

Windows Server 2016 Datacentre being run as a VM Host using Hyper-V. One of three in a cluster. The third server refuses to update Windows. It will download the latest updates and install them. When the server is rebooted it starts the update process, gets to about 60% then displays "We couldn't complete the update. Undoing changes". After about 30-40 minutes the server returns to the logon screen. When I log on the server acts as if it crashed by displaying the "Windows shut down unexpectedly" dialogue box. Windows updates shows that the latest update fails and it automatically starts to redownload the latest update.

This is what I have done so far to try and fix it:

1) Windows update troubleshooter many, many times.

2) renamed/deleted the catroot2 and systemupdate folders several times

3) Ran the reset-register windows update script from here: https://gallery.technet.microsoft.com/scriptcenter/Reset-Windows-Update-Agent-d824badc

4) Manually downloaded the latest update (KB4507459) from the MS Update Catalog and installed that but that failed as well.

5) Run SFC


I can provide the CBS.log and the latest WindowsUpdate log files if required.

The errors on the update log are the same every time:

2019/08/09 09:10:34.2560964 2220 3364 SLS CoCreateInstance Failed: hr = 0x80040154

2019/08/09 09:10:34.2561023 2220 3364 Agent Failed to retrieve SLS response data for service 117cab2d-82b1-4b5a-a08c-4d62dbee7782, error = 0x80040154

2019/08/09 09:10:34.2561086 2220 3364 Agent Caller Service Recovery failed to opt in to service 117cab2d-82b1-4b5a-a08c-4d62dbee7782, hr=0X80040154


In every case the line preceeding the error is:

2019/08/09 09:09:59.9233228 2220 10472 Misc Got 9482F4B4-E343-43B6-B170-9A65BC822C77 redir SecondaryServiceAuth URL: 117cab2d-82b1-4b5a-a08c-4d62dbee7782""


We have recently changed ISPs and firewall etc so I thought it might be related to that but internet connectivity seems to be normal for this server and all the other servers/VMs that we have are updating fine.

We had similar issues with the other two VM hosts but in both of those cases resetting the systemupdate folder fixed the issue.

Any clues on what to look at next?

Continue reading...
 
Back
Top Bottom