High enqueue wait

Question

The oracle version is 9.2.0.7. The total memory of machine is 16g. I change sga max size from 2g to 7g. I changed the db cache size from 1.5g to 6g because i was getting law buffer hit.I increased processes parameter to 1000 from 700 and sessions parameter to 1000 from 700.Now I am seeing the high enqueue wait in the database like 90% in statspack.
What could be the reason for this

Answer 1

There are no rows return from those queries
But i got this
select * from v$enqueue_stat where cum_wait_time>0
order by cum_wait_time desc;
INST_ID EQ TOTAL_REQ# TOTAL_WAIT# SUCC_REQ#
FAILED_REQ# CUM_WAIT_TIME
1 TX 310179 8293 310175
4 76761677
TC 130 25 130 0
526
US 21407 7 21407 0
97
SQ 14093 217 14093 0
78
HW 9835 13 9835 0
40
CF 8901 2 8901 0
29Well, the result says that your enqueue waits were "TX" type.
Followings are genercal cases of TX lock contention:
- Row lock contention - Row update confliction, PK confliction, Bitmap Index confliction, ...
- ITL entry contention
- Index split contention
- Misc
Unfortunately, TX lock contention generally has no relationship with buffer cache size. I don't think your increased enqueue wait was caused by bigger buffer cache. TX lock contention means that multiple transactions are modifying/accessing same data.
You need to capture V$SESSION_WAIT, V$SESSION and V$SQL(as i said) just when the enqueue wait happens. More info will tell you more reasonable explanation.