Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> catio.sql question

catio.sql question

From: SightBlinder <no email>
Date: Fri, 29 Oct 1999 10:45:39 GMT
Message-ID: <38197a82.37817955@cnews.newsguy.com>


After I posted my initial question about finding hot tables, I saw a thread about this script. I have been messing around with it for most of the day and am having two issues.

First is when you run the sample_io procedure, I am getting errors on the 21 line where is uses a hextoraw conversion to initialize tail_of_lru.

tail_of_lru := hextoraw('4000000000'); -- init to very large value

If I take out two zeros, making the init value 40000000, sample_io will run. Or if I change the datatype of tail_of_lru from raw(4) to something larger like raw(6) it will work. Should I have to do that inorder for that proc to run? Is
there an init.ora setting I am missing?

The other issue is on another database, 8.0.5. After the creation of the table, extent_to_object_tab, the script tries to create a unique composite index on file#, lowb, for this particular database it fails because there is two records where file# and lowb are the same. Should this be occuring? One of the records is a temp object. Should I delete this row from the table then create the unique index, or just create a non-unique index?

My last question, has anyone had good data come back out of using catio.sql? What were the parms you used for sample_io? did you cron it to run a different intervals throughout the day, with the same parms? Is there any significant overhead in running this?

anyone's experience with this would be of great help. ~Jim Received on Fri Oct 29 1999 - 05:45:39 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US