RE: Permissions block database creation

2006-03-14 Thread Marciano [Intercol]
You need to set GRANTs to create on the host.

http://dev.mysql.com/doc/refman/5.0/en/grant.html


-Original Message-
From: Doug Pinkerton [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, March 14, 2006 1:26 PM
To: mysql@lists.mysql.com
Subject: Permissions block database creation

I'm a total newbie working through the tutorial in DuBois's _MySQL_.  
I've got MySQL running on my PowerBook. In Terminal, I can use the  
mysql client to get responses to things like SELECT NOW(). But the  
command CREATE DATABASE sampdb; results in the following error.

ERROR 1044 (42000): Access denied for user ''@'localhost' to database  
'sampdb'
mysql

I am unable to determine its location, which causes me to suspect  
that it is in the portion of the system not accessible from the  
Finder. Can anyone offer a suggestion?

Mac OS 10.4.5

Thanks,
dp


-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:
http://lists.mysql.com/[EMAIL PROTECTED]



-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]



RE: Changing the binlog dir

2006-03-06 Thread Marciano [Intercol]
Well,

I supposed that your slave is broken from now, because your bin-log
is not the exact situation of your databases. So:

1) no. If your slave is stopped for a abrupt lost of connection to your
master, the old binlogs cannot do anything about. You need to copy the
databases from your master to your slave before start it, and, with this,
have a real database replication.

2) no. When you started, mysqld will create a new for you.

3) yes. Stop the slaves to copy the databases and rebuild your replication.


-Original Message-
From: Jeff [mailto:[EMAIL PROTECTED] 
Sent: Monday, March 06, 2006 12:00 PM
To: mysql@lists.mysql.com
Subject: Changing the binlog dir

Hello all,

I had a problem last night where my Master server filled up the /var/
partition and stopped logging to the bin log.  This caused all sorts of
havok on my slaves and replication. My bad for not watching this but now
what I'd like to do is move where MySQL writes the binlog to.  Currently
it's in the default /var/lib/mysql but I'd like to move it where I
actually have the database files which is on a much larger partition.

From what I've read I can put --log-bin=/data/hostname-bin into the
my.cnf and restart the mysql server.  

Questions:

1) do I have to move the old binlogs to the /data/ prior to restarting
mysql
2) should I move the binlog index as well?
3) will moving the binlog location throw the slaves off?

Thanks,

Jeff



-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:
http://lists.mysql.com/[EMAIL PROTECTED]



-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]



RE: 'Into outfile' doesn't include the column names. How can it be done?

2006-03-06 Thread Marciano [Intercol]
Maybe, if you make a file with the SELECT content and call them
from the OS shell, like this:

example of the content of the my_file
SELECT table1.a, tabela1.b, table1.c FROM mydatabase.table;

After, call this file from the OS shell:

shellmysql -A mydatabase  my_file

This will result on the screen. You can direct the result to a file:

shellmysql -A mydatabase  my_file  my_table.sql

If you need the tabs, do with -t on the options:

shellmysql -A -t mydatabase  my_file  my_table.sql



-Original Message-
From: Ariel Sánchez Mora [mailto:[EMAIL PROTECTED] 
Sent: Monday, March 06, 2006 5:12 PM
To: mysql@lists.mysql.com
Subject: 'Into outfile' doesn't include the column names. How can it be
done?

When using select into outfile, I can only get the table data, but I can't
find how to include the column names. I haven't been able to include the
column names into the actual return of the query, and they don't get stored
in the .csv

Anyone know how to include the column names, table name, other info, as an
actual answer from MySQL? I realize that when I use the command line I can
see the column names, but this is not the case when using, for example,
Delphi, and my real problem, when the outfile is made.

I use SQLyog for exporting to csv and they do include the columns, my bet is
its some SQL option. The manual doesn't specify, at least in the select
syntax part.

Ariel



--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]



RE: mysqldump backup on filters

2006-03-02 Thread Marciano [Intercol]
Your table is missing. Try this:

?mysqldump --single-transaction -u root clientdb table
--where=FLD_CLIENT_ID=1  client1_dbbackup.sql

-Original Message-
From: Rithish Saralaya [mailto:[EMAIL PROTECTED] 
Sent: Thursday, March 02, 2006 1:23 PM
To: MySQL general mailing list
Subject: mysqldump backup on filters

Hello.

We have a database driven system that serves multiple clients. We have a
single database for this purpose. The data(rows) in some of the tables are
specific to individual clients. In all such tables, we have a field
FLD_CLIENT_ID whose value depicts the client to whom that row of information
corresponds to. All other tables in the system, are agnostic to client
information; but are required for the system to function as a whole.

I would want to take seperate backups for individual clients. I try to use
mysqldump to generate a dump file by filtering on the FLD_CLIENT_ID column.
However, since not all tables contain the FLD_CLIENT_ID column, mysqldump
fails.

I use mysqldump from the commandline as


---
?mysqldump --single-transaction -u root clientdb --where=FLD_CLIENT_ID=1
 client1_dbbackup.sql


---

The error that is thrown is


---
mysqldump: Couldn't execute 'SELECT /*!40001 SQL_NO_CACHE */ * FROM
`ACCOUNT` WHERE FLD_CLIENT_ID=1': Unknown column FLD_CLIENT_ID' in 'where
clause' (1054)


---
which means thats the dump can't be created as the table ACCOUNT does not
contain the column FLD_CLIENT_ID.

Is it possible to write the where clause such that it should apply the
FLD_CLIENT_ID filter only if the column exists in the table; and if not,
dump the data anyways.

I am on RHEL - MySQL 4.1.11 - Storage engine INNoDB

Regards,
Rithish.


-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]



RE: Want mysql to return tablename.fieldname format

2006-03-02 Thread Marciano [Intercol]
Yes.

You can add on SELECT syntax the AS operation, like this:

SELECT
ads.id AS 'ads.id',
track.id AS 'track.id'

FROM .


-Original Message-
From: Ryan Stille [mailto:[EMAIL PROTECTED] 
Sent: Thursday, March 02, 2006 2:42 PM
To: mysql@lists.mysql.com
Subject: Want mysql to return tablename.fieldname format

I am working with an existing compilcated query someone wrote years ago.
When I dump the data from the query to try to figure out why I'm getting
unexpected data, I have three fields named id.  Is there anyway to
tell mysql to name the fields with the table name when they are
returned, so they show up as ads.id, track.id, etc?

-Ryan


-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:
http://lists.mysql.com/[EMAIL PROTECTED]



-- 
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]