You're in luck. http://thinkoracle.blogspot.com/2005/11/oracle-insights-tales-of-oak-table.html Chapter 7. If you need an instant remedy, this might fix you right up: http://books.google.com/books?id=b3DIkYO2gBQC&pg=PA9&lpg=PA9&dq=oracle+compulsive+tuning+disorder&source=bl&ots=yTXvCFe7xF&sig=yqsl1ZVbgG2VBtHLeldAvY0-6w8&hl=en&ei=upqqTN_0DsL78Aat8rSFBw&sa=X&oi=book_result&ct=result&resnum=5&sqi=2&ved=0CCkQ6AEwBA#v=onepage&q=oracle%20compulsive%20tuning%20disorder&f=false Seriously though, did the average waits get so large that they flipped to be too large for a 32 bit integer and simply appear to be negative? -bdbafh On Mon, Oct 4, 2010 at 11:19 PM, Shastry(DBA) <shastry17@xxxxxxxxx> wrote: > Hello Gurus, > > I would like to know if there is a book available to cover the below topics > in depth. Please suggest. > Does performance tuning books covers these areas or is there any different > pointers which needs to be hit. > Thanks in advance. > > EVENT COUNT(*) AVG(B.WAIT_TIME) > ---------------------------------------- ---------- ---------------- > library cache pin 1 -1 > SQL*Net more data to dblink 1 0 > gc cr multi block request 1 0 > SQL*Net message from dblink 1 0 > enq: TX - row lock contention 1 0 > SQL*Net more data from client 1 0 > SQL*Net more data to client 2 -1 > latch: session allocation 2 .5 > control file sequential read 2 -1 > db file scattered read 2 0 > latch: cache buffers chains 3 -1 > read by other session 4 0 > latch free 4 -1 > gc buffer busy 4 -.25 > gc cr request 4 -.5 > SQL*Net message to client 5 -1 > wait for unread message on broadcast cha 7 0 > nnel > > pipe get 10 0 > db file sequential read 37 -.21621622 > Streams AQ: waiting for messages in the 51 0 > queue > > PL/SQL lock timer 59 0 > SQL*Net message from client 2185 3.11716247 > > Regards, > /An >