This is perfectly normal, with the constant flood of "urgent" sets of 200 files from CERNVM people, the 'base' X-LUPD job from IRLEARN has not yet had a chance to cross the CEARN-MOP line. Since the people in IRLEARN have a lot of lists and update them very often, a lot of 'update' files get generated and you get a lot of checksum errors. The base 1.5n code will then ask for a refresh every day. I have a fix for that, which will wait until the refresh arrives before ever requesting a new one, but I'm waiting for the queue on CEARN-MOP to get down to a reasonable value before doing it (which probably means waiting 1 month or so - 1,800,000 right now). Eric