Windows updates might have corrupted our install

  • Hi all

    We've had a new VM built for a clean install of SQL Standard 2016 (this was about 4 weeks ago).
    We got that up and running and restored some backups to make sure everything was working and it was all good.
    The above was all done off-line (using VMWare).

    When we came to finally make the new server live, there were some windows updates that needed to be installed (I'm getting the details from our server team as I type). as I couldn't RDP across to the server to install some more recent backups.

    When I could RDP across, SQL wouldn't start as (apparently) the master database was corrupt.

    I followed the instructions on this page:-
    https://docs.microsoft.com/en-us/sql/relational-databases/databases/rebuild-system-databases?view=sql-server-2017
    (although it did say it was SQL2017) and ran the following command:-
    Setup /QUIET /ACTION=REBUILDDATABASE /INSTANCENAME=InstanceName /SQLSYSADMINACCOUNTS=accounts

    That didn't help so I tried to run a repair install from the install media but that didn't complete successfully.
    The error on that was something to do with a wait on databases services (from memory).

    Anyone any ideas why the master database would suddenly become corrupt after installing windows updates?
    I can't see any other changes that were made to the server between the initial restore about 3 weeks ago and Saturday just gone.

  • Without knowing what updates were applied, we can only speculate.

    I will say that where I work our anti-virus / anti-malware is exceptionally aggressive and has "broken" my SQL installs when I've (tried to) apply SQL Service Packs and Cumulative Updates.  Possibly you're in the same boat and one of the updates applied was a CU?

    In my case I recovered by rolling my VM back on a snapshot taken just prior to installing updates, if you have a snapshot it might be the way to go, roll back on it, then work forward from there.

  • I'm still waiting for the list of updates that were applied.
    I've also asked the same team to take a clone of one of current (working) servers and apply the same updates to see if it breaks).
    At least that way, we might be able to narrow down the culprit.

Viewing 3 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply