"[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:
> hi,
> i have problem with automatic setting of variable wait_timeout.
> mysql version 4.0.20 (and 4.0.18) on slackware linux.
> i've tried to set it in several ways:
>
> 1) in startup script as a parameter to mysql_safe:
> -O wait_timeout=30
>
> 2a
on 05/20/2004 10:18 PM, David Griffiths at [EMAIL PROTECTED] wrote:
>>
>> I can use any I like, I wont have not even built the DB yet, so I am open to
>> any suggestions.
>>
>> I don't see how I can do it all in one transaction, what does that mean?
>>
>>
>>
> What that means is you do 100,0
You didn't mention the table-type (storage-engine) you were planning on
using, but if you use InnoDB, and do it all in one transaction (ie turn
off auto-commit and commit just once at the endi), then any failure
during the mass-insert will cause a rollback, so you don't have to worry
about the int
Jonathan Villa wrote:
I'm administering a Red Hat machine which is used soley for MySQL. The
person working with the db has more than 15 million records in various
tables...they were trying to run some queries to create one table from
these many tables. When this was happening, they ran out of di
on 05/20/2004 09:36 PM, David Griffiths at [EMAIL PROTECTED] wrote:
> You didn't mention the table-type (storage-engine) you were planning on
> using, but if you use InnoDB, and do it all in one transaction (ie turn
> off auto-commit and commit just once at the endi), then any failure
> during the
Assuming you insert 100,000 rows, you also have to consider that any
indexes on the table will need to be re-analyzed to fix the statics.
Also, the inserts will be slower due to any indexes.
You didn't mention the table-type (storage-engine) you were planning on
using, but if you use InnoDB, an
> [EMAIL PROTECTED] wrote:
> > Another perspective on the subject of BLOB vs.
> Links.
> >
> > Links are easier to implement and may be an OK way
> to start. However, a file system is really a crude
> database, and I emphasize "crude". It's not very
> good at handling high transaction rates, acces
You might want to read it again. He is saying the performance and consistancy exists
in the database whereas it doesn't in a file system.
--ja
On Thu, 20 May 2004, David Blomstrom wrote:
> > [EMAIL PROTECTED] wrote:
> > > Another perspective on the subject of BLOB vs.
> > Links.
> > >
> >
> [EMAIL PROTECTED] wrote:
> > Another perspective on the subject of BLOB vs.
> Links.
> >
> > Links are easier to implement and may be an OK way
> to start. However, a file system is really a crude
> database, and I emphasize "crude". It's not very
> good at handling high transaction rates, acces
Hi everybody,
I'm experiencing that the command TRUNCATE TABLE with InnoDB takes an
extraordinary amount of time on large (> 400MB) tables - ie about 3-4
minutes. In comparison, performing a DROP TABLE then recreating it
takes about 2 seconds. When the truncate operation is performed
I have this same problem on 4.1.1 as well.
Donny
> -Original Message-
> From: Steven Roussey [mailto:[EMAIL PROTECTED]
> Sent: Thursday, May 20, 2004 7:33 PM
> To: [EMAIL PROTECTED]
> Subject: thread stack issues
>
> Since going from 4.0.18 to 4.0.20 (or 4.0.19) I now receive these warni
Faced with some larger than usual for me data requirements, I thought I
would ask some others what they think about my logic.
MySql 4
In short, I will have a file upload on a web server that will digest a file,
I will be able to dictate the format of this file. There may be a few
formats, the mai
On Thu, May 20, 2004 at 05:45:58PM -0400, [EMAIL PROTECTED] wrote:
> Another perspective on the subject of BLOB vs. Links.
>
> Links are easier to implement and may be an OK way to start. However, a file system
> is really a crude database, and I emphasize "crude". It's not very good at handling
Since going from 4.0.18 to 4.0.20 (or 4.0.19) I now receive these warnings
on startup:
040520 14:55:21 mysqld started
040520 14:55:21 Warning: Asked for 196608 thread stack, but got 126976
/usr/sbin/mysqld: ready for connections.
Version: '4.0.20-standard' socket: '/tmp/mysql.sock' port: 3306
On May 20, 2004, at 2:45 PM, [EMAIL PROTECTED] wrote:
Another perspective on the subject of BLOB vs. Links.
Links are easier to implement and may be an OK way to start. However,
a file system is really a crude database, and I emphasize "crude".
It's not very good at handling high transaction rate
MERGE tables have a table_name.MRG that contains pathed names to the
original table location. You can carefuly hand edit these or run an
ALTER command to fix them.
ALTER TABLE merge_table_name UNION=(table_1,table_2...)"
Ed
-Original Message-
From: Ed Reed [mailto:[EMAIL PROTECTED]
Sen
Another perspective on the subject of BLOB vs. Links.
Links are easier to implement and may be an OK way to start. However, a file system is
really a crude database, and I emphasize "crude". It's not very good at handling high
transaction rates, access from multiple machines, or volume.
If your
Hi Sasha,
I've tried writing a python wrapper but It's an hard work because I've
to wrap every type of queries, join, where, order, group and any other
MySQL 4.0 clause.
Have you ever written or used a wrapper like the one I need?
Anyone knows a simple way for developing it?
This is not just a SE
innodb and dbd
-Original Message-
From: Ed Reed
To: [EMAIL PROTECTED]
Sent: 5/20/04 3:52 PM
Subject: RE: Rename database
Is there anything wrong with just stopping the server and renaming the
database's directory in the DATA directory? I've used that method
without any problems. It also
Hi -
I'm running 4.0.18, so I can't use subqueries. I need to run a query to
get the sum of costs from multiple tables for a given master table.
What I have right now which is clearly not going to work is:
SELECT conveyor.serial, SUM(conveyorsupport.cost),
SUM(conveyoraccessories.cost) from (
Hi -
I'm running 4.0.18, so I can't use subqueries. I need to run a query to get the sum
of costs from multiple tables for a given master table.
What I have right now which is clearly not going to work is:
SELECT conveyor.serial, SUM(conveyorsupport.cost), SUM(conveyoraccessories.cost) from
Is there anything wrong with just stopping the server and renaming the
database's directory in the DATA directory? I've used that method
without any problems. It also works very well for making a copy of the
database to a new differently named database. Are there any pitfalls
that I haven't encount
http://dev.mysql.com/doc/mysql/en/RENAME_TABLE.html
-Original Message-
From: Ngim Ngau - Kzresults [mailto:[EMAIL PROTECTED]
Sent: Thursday, May 20, 2004 4:50 PM
To: [EMAIL PROTECTED]
Subject: Rename database
Hi, Is there a way I can rename a database? or at least copy an old database
wi
Hi, Is there a way I can rename a database? or at least copy an old database
with existing tables into
a new one?
Thanks.
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
hi,
i have problem with automatic setting of variable wait_timeout.
mysql version 4.0.20 (and 4.0.18) on slackware linux.
i've tried to set it in several ways:
1) in startup script as a parameter to mysql_safe:
-O wait_timeout=30
2a) in configuration file /etc/my.cnf, section [mysqld]:
set-variable
Jonathan Villa wrote:
>
>in /usr/local/mysql/data there is one dir named the same as the db
>they are working on, then there is this other file called ibdata1
>which is 2.1G in size.I take a look at the first few lines and all
>I get is garbled data, as if it were a binary file or something.
Our application needs to support both solaris 8 and solaris 9. I was
wondering if mysql has a backwards compatibility in terms of operating
systems. Can I use mysql-pro-4.0.20-sun-solaris2.9-sparc.tar file to
install mysql on Solaris 8?
Thanks
Aysun
"Leonardo Francalanci" <[EMAIL PROTECTED]> wrote:
>> Could you provide structure of tables and some data for testing?
>
> How? I tried to post a couple of zip files (2 mails, 10k each) but it
> doesn't work.
Don't send attachments to the mailing list. You can upload file to the
ftp://support.mys
I'm administering a Red Hat machine which is used soley for MySQL. The
person working with the db has more than 15 million records in various
tables...they were trying to run some queries to create one table from
these many tables. When this was happening, they ran out of disk space.
They had abo
On May 20, 2004, at 11:08 AM, Sasha Pachev wrote:
Greg Willits wrote:
I have two mysql apps running on the same machine (OS X 10.3.3). A
mysql 3.23.54 on port 14551, and a mysql 4.0.16 on 3306. Each has a
config file specifying the port and a unique socket name in /tmp.
They have coexisted just
On Thu, 2004-05-20 at 07:13, Lewick, Taylor wrote:
> Hi all, I am having a little trouble deciding how to relate some of my
> tables together.
>
>
>
> I was wondering about creating one big lookup table to relate 3 or 4
> tables together, but wasn't sure if that was a good id, or should I have
[EMAIL PROTECTED] wrote:
I remember reading about someone inquiring about including the user's
information in each query logged to the binlog. Neither "mysqlbinlog" or
"show binlog events" shows which user performed which changes to the
database (at least for me:4.1.1a-alpha-nt-log / binary distro
Daniel Cummings wrote:
We want to convert a 200+ meg table to a heap table. We've gotten pretty
good performance converting from InnoDb to MyIsam table but converting to a
heap table would give an even bigger performance boost.
You think it would, or it already has in your testing? Heap vs. MyISAM
[EMAIL PROTECTED] wrote:
I remember reading about someone inquiring about including the user's
information in each query logged to the binlog. Neither "mysqlbinlog" or
"show binlog events" shows which user performed which changes to the
database (at least for me:4.1.1a-alpha-nt-log / binary distro
You can't do efficient ranging on a HASH indexed column without a table
scan...
Select * from heap_table where indx_col between 247 and 258
This would table scan your 200meg table even if "indx_col" is a PRIMARY
KEY, using the default hash key. Hash key was the only choice prior to
4.1, but now y
To simplify, is It better to have many smaller lookup tables or one big
one?
Traylor:
You can create three entity tables (organization,contact,event) + the relation
tables (org_contact, org_event, contact_event). Small lookup tables are usually
better, and also give you points for sticking to
Marco Lazzeri wrote:
Yes, I know. Perhaps, I'm searching for workarounds.
Il gio, 2004-05-20 alle 17:58, Victor Pendleton ha scritto:
You can grant those permissions on columns but not on individual rows.
Marco:
Your choices are limited to creating a wrapper for your users. If they have
direct ac
Greg Willits wrote:
I have two mysql apps running on the same machine (OS X 10.3.3). A mysql
3.23.54 on port 14551, and a mysql 4.0.16 on 3306. Each has a config
file specifying the port and a unique socket name in /tmp. They have
coexisted just peachy for a very long time.
Now however, w/o any
Daniel Cummings wrote:
We have one query which takes approximately 2 minutes. MySql seems to be
unresponsive to any other threads until this query has completed.
On some platforms, the thread library has a limitation/bug that does not permit
thread pre-emption, which means that if a thread does
Jay Fitzgerald <[EMAIL PROTECTED]> wrote:
> Does anyone know how or where to set the maximum size limit on a per
> user basis for their db? This is not for a table, but for their entire db.
>
> example:
>GRANT ALL ON usera.* TO [EMAIL PROTECTED] IDENTIFIED BY "password"
> WITH max_db_size=5M
We want to convert a 200+ meg table to a heap table. We've gotten pretty
good performance converting from InnoDb to MyIsam table but converting to a
heap table would give an even bigger performance boost.
Does making a heap file this large make any sense?
Are there size limitations to heap
"Eve Atley" <[EMAIL PROTECTED]> writes:
> Not specifically a MySQL question, but my boss has a lending library, and I
> found a book here titled "Principles of Data-Base Management" by James
> Martin, publish date of 1976. I thought I might peruse this to get a
> background on DB Mangement, but th
Yes, I know. Perhaps, I'm searching for workarounds.
Il gio, 2004-05-20 alle 17:58, Victor Pendleton ha scritto:
> You can grant those permissions on columns but not on individual rows.
>
> -Original Message-
> From: Marco Lazzeri
> To: [EMAIL PROTECTED]
> Sent: 5/20/04 10:04 AM
> Subjec
You can grant those permissions on columns but not on individual rows.
-Original Message-
From: Marco Lazzeri
To: [EMAIL PROTECTED]
Sent: 5/20/04 10:04 AM
Subject: Limit operations by condition
I would like to grant SELECT/UPDATE/INSERT/DELETE privileges _only_ on
particular set of rows.
Not specifically a MySQL question, but my boss has a lending library, and I
found a book here titled "Principles of Data-Base Management" by James
Martin, publish date of 1976. I thought I might peruse this to get a
background on DB Mangement, but the publishing date concerns me. Has anyone
read t
Does anyone know how or where to set the maximum size limit on a per
user basis for their db? This is not for a table, but for their entire db.
example:
GRANT ALL ON usera.* TO [EMAIL PROTECTED] IDENTIFIED BY "password"
WITH max_db_size=5M;
Therefore, usera would be limited to 5 megs of enti
I have a database that stores information in Japanese characters stored in
a blob formatted fields. I am having trouble copying these fields from one
database to another.
I tried doing a mysqldump on the table and then copyng the Insert statement
generated but that failed; the inserted filed
>I want to compose a query that will pull out each
>distinct instance of a form (in other words, no
>duplicates) and then list which procedures use that
>form
Something like...
SELECT DISTINCT formName
FROM forms INNER JOIN procedures USING (formName)
PB
- Original Message -
From: Ro
I would like to grant SELECT/UPDATE/INSERT/DELETE privileges _only_ on particular set
of rows.
The GRANT syntax I'm dreaming could be:
GRANT SELECT, UPDATE, INSERT, DELETE
ON table_a TO user_a WHERE table_a.foo = 'bar'
I've tried using TEMPORARY TABLEs as follow:
CREATE TEMPORARY TABLE table_
I know my emial; accounts been getting quite alot of spam, though it's been like that
for a while now for my self. All I know my email was harvested from other lists too;
which may have been the case for myself.
Chris
>
> From: Daniel Kasak <[EMAIL PROTECTED]>
> Date: 2004/05/20 Thu AM 08:14:0
Can you post the explain plan and the query?
-Original Message-
From: Boyd E. Hemphill
To: [EMAIL PROTECTED]
Sent: 5/20/04 9:32 AM
Subject: Very large query text?
I am having a tough time of it in production today ... Any help would be
appreciated.
I am executing a query of about 10k in
I remember reading about someone inquiring about including the user's
information in each query logged to the binlog. Neither "mysqlbinlog" or
"show binlog events" shows which user performed which changes to the
database (at least for me:4.1.1a-alpha-nt-log / binary distro ) . Is that
information
Since your on 4.1, give this a try...
select *
from tbl as a
where a.recdate=(select max(b.recdate)
from tbl as b
where b.id=a.id and b.mount=a.mount)
Ed
-Original Message-
From: Duncan Hill [mailto:[EMAIL PROTECTED]
Subject: Re: Query help with groupin
I am having a tough time of it in production today ... Any help would be
appreciated.
I am executing a query of about 10k in size.
When I do this it takes about 15 seconds.
If I remove a bunch of case, sum and if statements but get the same
explain plan the query runs in 5 seconds (it size is l
Hi all, I am having a little trouble deciding how to relate some of my
tables together.
I was wondering about creating one big lookup table to relate 3 or 4
tables together, but wasn't sure if that was a good id, or should I have
a look up table
For each pair of tables.
Here is a simple ex
On Thursday 20 May 2004 12:49, Egor Egorov might have typed:
> Duncan Hill <[EMAIL PROTECTED]> wrote:
> > I have a table that has data that looks like:
> > ++-+---+-+
> >
> > | id | recdate | mount | perused |
> >
> > ++-+-
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
- --
~ |...|
~ | _ _|Victor Medina M |
~ |\ \ \| | _ \ / \ |Linux - Java - MySQL |
~ | \ \ \ _| | |_) / _ \ |Dpto. Sistemas - Ferreteria
Hi all.
Over the last 2 days, both my home and work email addresses have been
hit with a sudden surge in spam, and the content in both cases is the
same. Another strange thing is that my home address got basically none -
until now.
I'm assuming my addresses have been harvested from this list, as
Egor Egorov <[EMAIL PROTECTED]> wrote:
> Duncan Hill <[EMAIL PROTECTED]> wrote:
>> I have a table that has data that looks like:
>> ++-+---+-+
>> | id | recdate | mount | perused |
>> ++-+---+-+
>> | 1 | 2004-05-20
Duncan Hill <[EMAIL PROTECTED]> wrote:
> I have a table that has data that looks like:
> ++-+---+-+
> | id | recdate | mount | perused |
> ++-+---+-+
> | 1 | 2004-05-20 10:46:12 | QUAR | 80 |
> | 2 | 2004-05
I have a table that has data that looks like:
++-+---+-+
| id | recdate | mount | perused |
++-+---+-+
| 1 | 2004-05-20 10:46:12 | QUAR | 80 |
| 2 | 2004-05-20 11:05:11 | QUAR | 99 |
| 2 | 2004-05-20
had no mysql list mail for a while
just testing please ignore
mysql
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
62 matches
Mail list logo
