 | | | Oracle List | | Juan
This appears to be a message that was sent to you privately.
If so it is very bad form to include it in a public mailing list
without the permission of the person that sent it to you.
Juan Carlos
Please note that I did not suggest "joining " the two SELECT statements
together into a single SELECT statement. What I suggested was running the
two independent SQL statements UNION 'dHi I got from tom kyte their good hints (hint suggested to be used to tune
in normal situations by every one (not hints in exceptional or too complex
querys) )
this was interesting because some Ron
Do you mean SQL*Loader? Why would you want to use the wrong SQL*Loader?
Or do you mean export or import?
Dennis Williams
DBA
Lifetouch Inc.
dwilliams@(protected)
-----Original MesThis was a mistake an I am sorry if you see the correct email was in the
subject
but I think this don 't give you the right to insultme
cjgait@(protected)
Even there is very few what I can to edThanks Tim
that was only one example to show the problem.
the insert is of thousands of records not only 5.
The problems is that I can 't do a join because some circumstances can cause
another If you have 9i you can turn on index monitoring. It goes something like this
ALTER INDEX index NOMONITORING USAGE
Results should be in V$OBJECT_USAGE.
You can only do it from your own scheContent-Type text/plain charset "us-ascii " format flowed
In the database the user owns it 's own plan_table which is
Version of Oracle is 9.2.0.4.0
SQL > desc plan_table
Name Juan Carlos
First I have to ask why you are bothering with APPEND when you are only
adding 5 rows at a time? The APPEND feature (a.k.a. direct-path) as well as
parallel execution is intended for
On 04/02/2004 10 39 55 AM April Wells wrote
> Hey! You might could (oh yeah... I 'm talkin Texan) loosen up some of those
> SLAs right about now...
I wouldn 't misunderestimate (now I 'm talkiWell perl is truly portable truly open source language while Java is
owned by SUN overly complex and fairly slow monster which tries to be
everything to everybody. Java is going to lose against .It 's not so resource demanding. I have it turned on on all my databases.
It isn 't the same thing as STATISTICS_LEVEL ALL which also collects
the OS statistics and is fairly expensive. Speaking oHi!
> I welcome any discussion from the larger list regarding one or more of the
> following
> 1. Using BLOBs as opposed to BFILE. (Although I 'm not sure whether BFILE
> would work with the AHi All
I was going through Oracle Doc 's and certain things just stuck me I am
jotting them down plz let me knw your opinions on these
1) What is the difference between physical reads and physic
If you look in v$lock for the row of type TM
and check the CTIME column - this is roughly
the time (in seconds to the nearest 3) that the
table has been locked or a request has been
waiting.
The difference between shard and exclusive locks is very well described
in Andrew Tannenbaum 's "Modern Operating Systems ". Oracle uses share lock
on the table level when the foreign key column isn 't
When transaction B attempts to lock the row
it sees the lock byte on the row is set and checks
the ITL in the block to discover all it can about
the transaction in that ITL position. (e.g. hasSQL > SELECT *
2 FROM v$oracle-l
3 WHERE UPPER(location) 'VANCOUVER '
4 AND UPPER(country) 'CANADA '
from v$oracle-l
*
ERROR at line 2
ORA-00933 SQL command not p
For those interested Bookpool got it in stock now.
Oracle Database 10g New Features
1 Copy ISBN 0072229470 $18.95
Subtotal $ 18.95
US Postal Media Mail $Learning Perl (Randall Schwartz) is the best book to start with. For more complex stuff
I recommend Damian Conway book "Object Oriented Perl " and O 'Reilly book "Advanced Perl Programming ".
(author
On 04/01/2004 08 08 04 PM Connor McDonald wrote
> Has anyone come up with a reason why you 'd want the concept of a KEEP INDEX option? Consider the
> scenarios
Connor I have to confess I loHi!
> I believe Oracle is enforcing this to guarantee consistency of the
> database. I realise you are probably using the APPEND hint for
performance
> reasons but is it really necessary?
Ye
I believe Oracle is enforcing this to guarantee consistency of the
database. I realise you are probably using the APPEND hint for performance
reasons but is it really necessary?
OtherwisOh NOW I know what 10g stands for
But now I wonder does the g stand for acceleration or mass?
At 04 21 PM 4/1/2004 you wrote
>_accelerate_with_1g ( 9.81 m/sec*sec in metric units - ideal fo
On 04/01/2004 04 16 29 PM Juan Pacheco wrote
> Hi I have a serious problem
Try committing among the two. It will help.
--
Mladen Gogala
Oracle DBA
Srinivas
My theory as a DBA is to perform tasks in a manner that is least likely
to cause bad consequences. Is there any way export/import would work for
you? I have a much warmer feeling about Hi I have a serious problem
I can 't do two append insret to a table in a same transaction and I can
join in only one.
because some times the same process( a function) must be runs more than once
Yes it probably would.
But depending on the circumstances it might be an
even better job for the initialisation parameter
db_file_name_convert.
Depending on what the original poster is trying to
Technically it 's a cost-based decision.
For 'offline rebuilds the optimizer could choose
to use the index or table. For online rebuilds it
looks as if the optimizer HAS to use the table.
Ju--- Ben <poelsb@(protected) > wrote
> Hi
> Are there any documents or books anyone can
> recommend that discuss how
> best to handle security in such an environment.
> Thanks
> Ben
|
|
|
