Currently, the pagination is performed upon the raw results returned in the searchBean, but only the items which pass the permissions checks and duplicate full name checks are displayed to the user. This is confusing, particularly if there is a large difference between the two results.

I don't see why duplicate full names should ever be excluded, and the pagination should be performed on the pruned result set, not the raw results obtained. This bug was not introduced in 2.7, but still exists on trunk.