I have had recently the case where a huge SQL server had the “SQL Server VSS Writer” service not running because it was disabled...
This resulted in the Block Monitoring driver not being aware of the changes happening on the Source since the initial full replication completed.
After the cutover completed, the customer realized too late that one week of SQL data was missing on Target...and users were already performing some change on the database Target.
Now they need to choose between rolling back the workload and lose the data that has been changed since Cutover completed, or do not roll back the server and miss one week of data...HUGE BUSINESS IMPACT!
This could have been avoided by adding a pre-validation mechanism that would detect that the server is running SQL and that the SQL VSS writer was disabled (I could see everything in the diagnostic logs, so PlateSpin was aware of these).
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.