Hi All,

We use nbackup to pull live copies of customer's production databases back to our office to play with and help with support. We run a level0 backup every weekend, and a level1 every day which are both copied back to the office over night. That just leaves us to run a level2 when we want a live copy. Generally it does exactly what it says on the tin and we are all happy at the speed we can grab a customer's 5GB database without the server really noticing - as opposed to a full gbak stealing all the disk IO.

However, occasionally we end up tearing our hair out chasing down bizarre bugs not directly related to the issue we are really interested in, and it always ends up being a problem with the indexes that nbackup has restored. There doesn't seem to be any pattern to the customers this affects, the size of the level2, the time of day, or anything. Sometimes a random index seems to just get corrupted. Queries still use it, but get no results. If we deactivate/activate an offending index then everything is great again.

Has anyone else come across something similar? 

Thanks

Ian