Hi Guys, We had a major problem over the weekend where a specific process seems to have gotten stuck for about 12 hours before I killed it and restarted it (and it works ok now). The process was doing a delete from one table, based on critieria in another table. The two tables were joined together and the table being joined in was HUGE (132 million rows). Normally the delete runs in about 15 minutes, even though the join results in a table scan -- however this time it appeared to never want to finish. Upon further investigation I saw that it was waiting on PAGEIOLATCH_SH. Never waiting more than maybe 100ms, but always with that wait type and always showing a different Page ID. Also no other processes were accessing anything in that database -- it was just the one process accessing the database entirely to itself for the entire duration. Does anyone have any ideas as to what might have happened? I'm completely stumped on this. One thing I did do was create an index on the 132 million row table so that the join it does is now a SEEK instead of a SCAN and that sped it up by quite a bit. But with that said, I still need to determine why the lockup occurred all of a sudden when doing that SCAN. Any thoughts or ideas would be greatly appreciated! Thanks Edit: I should also mention that only about 2000 rows were being deleted so it wasn't a hugely logged operation or anything like that.