Having retired a SQL Server Box I had which was clustered and replicated I had to get the databases back online ASAP to find some historical data. Problem is someone bent the pins on the SCSI card for one of the clustered box so I could not just start the cluster and get to work.
Now I had a single node of a cluster but not functioning.
Easy to fix emergency solution.
1. Install a named SQL Server Instance – saves time over uninstall & reinstall, after all I am not going to put this server back in production
2. Restore Backups or Force recovery of MDF & LDF
In this case I did not have the time to deal with restoring the database, as I have a DB thats over 100GB and on relatively slow disk (compared to the snazzy EMC SAN I have now). So my next option was to hack it.
I renamed the original MDF & LDF Files used by my clustered instance. Then I created databases from my scripts in source control with the exact file paths & setups of my old production databases. Caveat: Make sure you make the databases with small file sizes (deadlines…).
Shutdown the Datbase Server
Rename the New Files, and replace them with the Original Files.
Start the SQL Server Instance, and give it a few minutes to hiccup…. then voila… Database up and running. Instead of a 24 hour database restore, the system was dead slow for about 10 minutes figuring out what to do, and now everything works.