It appears that you are using AQ, as that is what 'unread message' waits are for.
Apparently also DBMS_PIPE, as per the 'pipe put'. I don't know if AQ makes
any use of DBMS_PIPE.
The numbers provided aren't enough to tell if this might be a problem. Your
database has spent almost 2 hours waiting on these two events. If your database
has been up for 3 hours, then these probably need some attention.
If it's been up for a month, accumulating 2 hours of waits over that period may
not represent any problem.
I do find it interesting that the 'pipe put' waits are about 5 seconds each on average,
as that's a pretty long wait.
You may want to discuss this with the app folks.
A good buy would be "Optimizing Oracle Performance" by Cary Millsap and Jeff Holt.
It goes into some detail about how to use Oracle traces to pinpoint problems.
HTH
Jared
Zabair Ahmed <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED] 11/06/2003 05:44 AM
|
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]> cc: Subject: oracle events... |
Does anyone know what these oracle events are and how the wait times can be reduced. This is an extract from statspack output on 9.2.0.4 on HP 11.11
Top 5 Timed Events
~~~~~~~~~~~~~~~~~~ % Total
Event Waits Time (s) Ela Time
-------------------------------------------- ------------ ----------- --------
pipe put 780 3,904 44.65
wait for unread message on broadcast channel 3,016 3,034 34.70
PX Deq: Signal ACK 33,937 678 7.76
db file scattered read 6,575,215 499 5.70
PX qref latch 222 200 2.29
-------------------------------------------------------------
Want to chat instantly with your online friends? Get the FREE Yahoo! Messenger