Sometimes the wait interface is useless.
joel garry
Mark D Powell
> http://jamesmorle.wordpress.com/2009/11/09/the-oracle-wait-interface-...
>
> jg
> --
> @home.com is bogus.http://www.infoq.com/news/2009/11/oracle-sun-palns
Interesting article. Knowing the proper use and limitations of your
tool set is an important attribute for all developers/DBA's. This is
true with the Oracle Wait Interface and equally with whatever Java/C+
+/.Net Development environment one works in.
Thanks for the link.
IMHO -- Mark D Powell --
Mladen Gogala
> http://jamesmorle.wordpress.com/2009/11/09/the-oracle-wait-interface-is-
useless-sometimes-pt/
>
> jg
Of course, all that James Morle says is that there are situations in which
the wait interface is not nearly precise enough to do diagnostics. His
query spends lots of CPU time, without doing any major waits. Of course,
there are tools to deal with that, too: V$SQL monitors SQL by the spent
CPU time, there are EXPLAIN PLAN command and 10046 trace which would
quickly locate the offending CPU. All that James Morle says is that wait
interface is not and must not be the only tool in the DBA arsenal. But
then again, Cary Millsap sort of said that in his book, as Morle has
mentioned.