We got this one last night:
"Could not continue scan with NOLOCK due to data movement."
This came from a SQL command. I have never seen this one before. We did experience cluster problem however overnight. Can this be related to that or this is something that can be improved at the code level?