>I manage several production SQL servers, one with many SharePoint databases on it (about 200), and more are being migrated to this same server. Because of a continuing problem with fragmentation, I have a maintenance job that runs nightly that rebuilds the indexes on the user databases on the server. Even this has been unsuccessful and the level and number of fragmented indexes continues to climb. > >I have tried to implement the MS recommended method of SharePoint database de-fragmentation, but this had also not worked. > >REF: http://support.microsoft.com/kb/943345 > >If anyone has knowledge or can help with this, it would be greatly appreciated.
shrinkfile still causes fragmentation on disk and countless shrinkfiles can become an issue. It is of course the best course if you have a file growth issue and you need to do something about it. I thought of throwing a party when I saw truncate removed in 2008. ;-) Just last week I had to do a shrinkfile on a log that went out there due to a index maintenance and backup issue that made the log go to around 70GB. It's not really a problem but did cause a problem with my mirror and it was a mistake on my part and should have been avoided.
I highly recommend reading a few things and concerns on anything that goes with shrink though so understand what you really are doing to things and performance.
