11:11:22.85 spid15s waiter id=processfe9888 mode=S requestType=wait 11:11:22.85 spid15s waiter id=process30a53d8 mode=X requestType=wait 11:11:22.85 spid15s DELETE FROM RECORD_LOCK WHERE LOCKEE_SID = 14208235 AND LOCKEE_TYPE = 99 AND LOCK_TYPE = 2 11:11:22.85 spid15s SELECT count(*) REC_CNT FROM RECORD_LOCK WHERE LOCK_END_TIME IS NULL AND LOCKEE_SID = 14208235 AND LOCKEE_TYPE = 99 AND LOCKEE_SUBTYPE = 99 AND LOCK_TYPE = 2 AND LOCKER_ID = '885FEDB0FA86FF0A2BD1895EB36B5F6E' 11:11:22.85 spid15s select count ( * ) REC_CNT from RECORD_LOCK where LOCK_END_TIME is null and LOCKEE_SID = and LOCKEE_TYPE = and LOCKEE_SUBTYPE = and LOCK_TYPE = and LOCKER_ID = 11:11:22.85 spid15s frame procname=adhoc line=1 sqlhandle=0x0200000000cc4e1d11ac8630a25c26d6bec7e09e96ba396d Here is one instance of a deadlock showing a 'select statement', having shared lock 'S' I have not tried the snapshot isolation, but in my limited tests that also behaves the same. Underlying fact is read committed snapshot is stillign letting shared locks to be acquired. I am running out of ideas and adding nolocks in some cases. Thanks everyone for your valuable ideas and spending time to look at my post, I appreciate it.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |