We have this very busy oltp system running sql 2008 sp3 cu2. The server has 32 cores allocated to it. We are frequently encountering deadlocks between multiple processes. The deadlock graph shows two processes deadlocked between different lock partitions on the same object . I have attached the deadlock graphs, replaced the table names in this case.
Since the select needs schema stability locks, and ddl(disable tigger) needs schema modification locks we can understand blocking but since we cannot disable lock partitioning we end up having a deadlock. Please suggest what to do in this case. We have seen many deadlocks like this in the past few months.