Post reply

Disaster recovery - master db concerns

  • September 21, 2011 at 12:11 pm

    #248345

    I'm implementing a planned disaster recovery. The goal is to recovery completely from scratch. i.e. build hardware, reimage machines, pull files from tape.

    From my perspective, I will have to reinstall SQL Server applying patches and restore databases, resyncing users. My concern is restoring the master database. I'm looking through the failure recovery and moving the master database sections of this article:

    http://msdn.microsoft.com/en-us/library/ms345408.aspx

    from a windows perspective, the DR server will be almost identically set up to the production server. DNS name, drive mappings, etc. The only thing that may be different is ram and drive size. If the drive mappings and installations will be identical, would I still need to remap the file locations and alter databases?

  • September 21, 2011 at 12:23 pm

    #1384730

    I'm not sure i would even consider restoring master at all.

    the only items of any importanceto me are the following:

    1. Server Logins and their permissions.

    2. Linked servers. definitions and their remote logins.

    3. some administrative stored procedures and functions in the master database.

    4.-edit- any trace definitions running ont he server...forgot about that. also scriptable.

    instead of restoring master, I know I script out the logins with sp_help_revlogin on a regular basis;

    any linked servers are created by me, and their definitions are in TFS.

    I also script out all the user tables, procs and functions in master as well every once in a while , but the administrative procs are all in a suite of scripts that are also in TFS for change control.

    I'd be afraid i'd restore master with bad data that kills the server; much easier to run some scripts i've tested, and start attaching or restoring databases on a system that I know works.

    Lowell

    --help us help you! If you post a question, make sure you include a CREATE TABLE... statement and INSERT INTO... statement into that table to give the volunteers here representative data. with your description of the problem, we can provide a tested, verifiable solution to your question! asking the question the right way gets you a tested answer the fastest way possible!

  • September 21, 2011 at 12:45 pm

    #1384743

    I agree with you. These are application driven databases. I can't be sure that they won't have special configurations. The restore would come from the latest backup. We're not creating a disaster, just implementing steps that we would take in case.

    I understand the steps (and have execute) for restoring the master database on the same server, but I'm not 100% about this one.

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

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