>>> [EMAIL PROTECTED] 11/15/02 04:48PM >>>
Okay ... here is a curved ball ...
This AM an application support person called and mentioned that we have a form that has become slow compared to 2 days ago. So we saw the slow behavior and then I did a 10046 trace on it.
After analysis, found that one query is taking about 90% of total time and isolated it to be worked on.
On our day-old instance (refreshed daily from prod), this form runs very fast. After comparing everything, I found that for some test we have db_file_multiblock_read_count = 4 whereas it is 32 (don't ask) on production.
So after changing this variable in my session to 4 (on production), I checked the explain plan and it seems to be okay, picking up the right indexes etc.
So, I asked the support person to add following line in pre-form trigger on the form and run a test
forms_ddl('alter session set db_file_multiblock_read_count=4');
ran the form, no change in the performance. So I requested following like to be addes as well
forms_ddl('alter session set events ''10046 trace name context forever, level 8''');
The form runs as fast as it can.
So we thought maybe flushing the shared pool might help, so we flushed the shared pool (again don't ask, this is prod env, lunch time, light load).
Still same, with 10046 event set, the form runs as fast as it can, but if I take it out, it slows down.
I am confused ... does anyone have a plausible explanation on why this is happening?
Oracle 9201, AIX 5l
Raj
______________________________________________________
Rajendra Jamadagni
MIS, ESPN Inc.
Rajendra dot Jamadagni at ESPN dot com
Any opinion expressed here is personal and
doesn't reflect that of ESPN Inc.
QOTD: Any clod can have facts, but having an opinion is an art!