Re: grant select command so slow
- From: "Riyaj Shamsudeen" <riyaj.shamsudeen@xxxxxxxxx>
- To: john.kanagaraj@xxxxxxxxx
- Date: Thu, 22 May 2008 20:36:24 -0500
Joan
It might be easier to turn on 10046 event in your session, run grant
statement, tkprof the trace file and see which action is consuming time.
Cheers
Riyaj Shamsudeen
The Pythian Group www.pythian.com
orainternals.wordpress.com
On Thu, May 22, 2008 at 11:41 AM, John Kanagaraj <john.kanagaraj@xxxxxxxxx>
wrote:
> > If that was the case, how did the grant run very fast the first time,
> > according to the OP.
>
> Not sure if that question was addressed to me, but here's a stab at
> the answer (Disclaimer: I am a member of the BAAG party, so guessing
> violates my agreement 8-)
>
> The OP wasn't clear *when* the grants were first run. Was this when
> the number of tables/views were less? What was the other load at that
> time? Tracing the event and providing some stats would give us the
> right clues. Otherwise we are just guessing here.
>
> In general, the principles that I explained above applies: The
> execution of thousands of GRANT statements translates to thousands of
> hard parses and every SQL statement will get a different hash value,
> chopping up your shared pool into fine bits.
>
> --
> John Kanagaraj <><
> DB Soft Inc
> http://www.linkedin.com/in/johnkanagaraj
> http://jkanagaraj.wordpress.com (Sorry - not an Oracle blog!)
> ** The opinions and facts contained in this message are entirely mine
> and do not reflect those of my employer or customers **
> --
> //www.freelists.org/webpage/oracle-l
>
>
>
--
Cheers
Riyaj Shamsudeen
The Pythian Group - www.pythian.com
orainternals.wordpress.com
Other related posts:
