Hi Experts,

Just wondering if this is A possible and B the right thing to do if it is.
Or what is best practice when you have lots of small (count) variable constraints.

I would like to use a constraint on a table field, as a lookup from another table, tblConstraints.

So for example see code attachment
Code Snippet:

–Table for All Constraints
CREATE TABLE tblConstraints (                    intConstraintID int IDENTITY(1,1) NOT NULL,
vchConstraintName varchar(50) NOT NULL,
vchConstraintValue varchar(100) NOT NULL,
intOrder int NULL
)
GO
–Insert a Constraint
INSERT INTO tblConstraints (vchConstraintName, vchConstraintValue)
SELECT ‘TreeSpecies’, ‘ACA’ UNION ALL
SELECT ‘TreeSpecies’,  ‘BAN’

–Select Constraint
SELECT vchConstraintValue
FROM tblConstraints
WHERE vchConstraintName = ‘TreeSpecies’

–Add Constraint to a table??
CREATE TABLE [dbo].[TreeSpecies] (
[Species] char (255)  NOT NULL )

ALTER TABLE [TreeSpecies]
ADD CONSTRAINT [CK_TreeSpecies_Species] CHECK ( (
[Species] IN (
SELECT vchConstraintValue
FROM tblConstraints
WHERE vchConstraintName = ‘TreeSpecies’
)  )  )  ) ;

Solution:

The column that a foreign key references does not have to be a primary key, it just has to have a unique constraint on it.

I don’t think SQL is simply ‘looking’ in sys.constraints, rather that the constraints are a piece of table metadata the existence of which can be viewed using this system view.
Posted via EE Mobile

When I try to run SQL Loader, it has an error…
Anyone knows how to solve it?

Solution:

Hi,

You need to set the ORACLE_BASE in path.

This link should help you

http://www.orafaq.com/forum/t/51572/2/

is there a way/method to identify or remove redundant/unused tables in a production database?
would there be any harm of leaving those tables there.

Solution:

1 – if you only turn auditting on for tables you think are good candidates then no, it shouldn’t affect performance,  if you are right and they aren’t used, then it will have 0-performance because you only audit when the object is touched.  No touch=no audit=no performance

“drop any not referenced”  – remember auditting records are windows in time
if you audit everything today and then I touch your table tomorrow, I won’t appreciate it if you drop the table.

So, your audit window must be large enough to capture all possible (and desirable) activity

And “referenced” using dba_dependencies ONLY returns objects that are in that database.
If you have references in other databases through db links, those won’t be seen
If you have references in dynamic sql in that or other db’s, those won’t be seen
If you have references in applictions (c#, java, vb, fortran, python, whatever) that aren’t in the database those won’t be seen.

1 and 2 give you a good start,  then you have to use your own knowledge of the database and all possible entry points to it to determine how far you need to search

Does anyone know the pros/cons of bouncing a production instance every weekend?
Is this a normal or bad practice for oracle databases.

My understanding it is not good. however, our DBA does it every weekend to take a cold backup of the database. Oracle then seems to lose all the cache advice statistics after that.

Solution:

1st  – you are forcing an outage, your system is unavailable while you do the backup.

2nd – any “cached” information is lost when you bounce the database
some examples…
- advice views as you already determined
- automatic memory management  pool sizes will need to readjust
- all statements will need to be reparsed
- all data in buffer caches will need to be reloaded
- all pinned code will need to be reloaded

3rd – schedules will be thrown off
- any scheduled processes will either be skipped during the down time
or they will need to play “catchup”

My db has a mysql backend and MSAccess frontend. Currently I am using a temporary table (created from query of multiple tables with some concatenate and calculated fields) as the record source for a search form. Problem is that it takes about 10 sec to create the temporary table and this delay is set to increase as the db grows.

I am know thinking of turning the temporary table into a permanent table. My delema is in deciding the best way to up date that table so that the search result are always current.

I believe that my options are:

1) After entering a new record or editing an exiting record – Delete related records in the search table and then insert the updated related record.

2)After entering a new record Insert the record in the search table. After updating an existing record run and update query on the search table with the primary key of the updated record as the where condition.

I would prefer to use the first option as the code will be far easier. However, I am worried that errors may pop – up if a record is deleted in the search table while another user is looking at it.

Need some advise for MSAcceQL experts.

Cheers

SB9

Solution:

Well, I imagine that it all comes down to how you handle it in your code.

If you try to pull details up for a particular record and the record isn’t there then I would suggest displaying to the user that the record does not exist and the possible reason for it.

I just installed phpmyadmin version 3.2.2 with mysql 5.1.40 and Apache 2.2 on Windows. When I goto the index page, I get the normal phpmyadmin login screen. I put my username and password in and it hangs for about a minute and then I get a blank page. I tried an invalid password and it does the same thing. I created a config file by going to the Setup folder in the root directory of phpmyadmin and typing in all the correct info. I then clicked Save and it shows the server I just created in the server list. If I refresh the page, I then get a message that says “There are no configured servers”, even though I just configured one and clicked save. I then created another server, saved it and then before refreshing it, I clicked Download and saved it to phpmyadmin\config, which is where the documentation says the config file should be. If I refresh the setup page again, it still is not listed there and the phpmyadmin login page still does the same thing. I tried using different login methods such as cookie, http and config, but all do the same. I have included the config.inc.php file, which is in the phpmyadmin\config folder. Also, I connected via command line and it works fine, so obviously there is not a firewall issue and my ip, username and password must be correct. I must be missing something here, I have done this a few times and never had any trouble.
Code Snippet:

<?php
/*
* Generated configuration file
* Generated by: phpMyAdmin 3.3.2 setup script by Piotr Przybylski <piotrprz@gmail.com>
* Date: Sat, 24 Apr 2010 12:00:21 -0700
*/

/* Servers configuration */
$i = 0;

/* Server: mysql [1] */
$i++;
$cfg['Servers'][$i]['verbose'] = ‘mysql’;
$cfg['Servers'][$i]['host'] = ‘192.168.0.31′;
$cfg['Servers'][$i]['port'] = 3306;
$cfg['Servers'][$i]['socket'] = ”;
$cfg['Servers'][$i]['connect_type'] = ‘tcp’;
$cfg['Servers'][$i]['extension'] = ‘mysql’;
$cfg['Servers'][$i]['auth_type'] = ‘http’;
$cfg['Servers'][$i]['user'] = ‘root’;
$cfg['Servers'][$i]['password'] = ”;

/* End of servers configuration */

$cfg['DefaultLang'] = ‘en-utf-8′;
$cfg['ServerDefault'] = 1;
$cfg['UploadDir'] = ”;
$cfg['SaveDir'] = ”;
?>

Solution:

I just reread the documentation and must have missed the fact that you are suppose to move the config file out of the config folder. I did that and it works fine.

I know this must be very simple, but I cannot make this work – I am trying to store a simple file path in a MySQL database from my PHP web page. But the backslashes get lost when I write the update statement (it’s an existing record).

$sPath = “c:\\mypath\myfile.jpg”;
$sSQL = “UPDATE mytable SET filepath=’” . $sPath . “‘ WHERE id = 9″;
mysql_query($sSQL);

I end up with a filepath that is c:mypathmyfile.jpg

If I first call addslashes($sPath) then I end up with double the slashes in the database.

Can someone help this poor soul?

Thanks

Solution:

Or use mysql_real_escape_string()

http://www.php.net/manual/en/function.mysql-real-escape-string.php

Regards

What is the basic difference between Dimension/Facts relationship and PK/FK realtionships?

Thanks

Solution:

No difference that I can think of.

OLAP doesn’t necessarily imply “dimensional”. If by “OLAP” you just mean a database used for business intelligence / decision support purposes then that could also be a normal form model. But either way a foreign key means the same thing in each case.

Hi everybody.

I have an application that send emails to subscribers every week, every 14 days and every month, I select this information using an sql script every day.
Now i need to implement the two months subscription.
I like to make an sql select that retrieves this information and send the newsletter (for months with more than 28 days i’m sending the month email on 28th day)
Thanks for all your help.

Byron H.

Solution:

how about using modulo?

like….

select ………..
from ……….
where …………
and day(current_date)=28
and month(current_date) % 2 = 0

That will return true for all even months
while the

and month(current_Date) % = 1
will return true for all odd months

please suggest the method to back up my sql automatic ly to flash drive or another server

Solution:

http://dev.mysql.com/doc/refman/5.1/en/mysqldump.html

If you are in a *nix environment, write a cronjob to run at the times you want, get it to mysqldump the databases you need, tar them up (optional) then either scp them to another server or cp them to the flash drive.