Hi,
I have executed ANALYZE TABLE for myisam tables, but still myisam is showing
more scanning of rows as compared to innodb. What does ANALYZE TABLE command
exactly do for myisam storage engine.
Thanks
Krishna
On Wed, Apr 2, 2008 at 9:48 PM, Rob Wultsch [EMAIL PROTECTED] wrote:
On Wed, Apr
Hi,
On myisam storage system
mysql explain select ui.user_id, ucp.user_id,ucp.payment_date from
user_info ui, user_course_payment ucp where ui.user_id=ucp.user_id;
On Wed, Apr 2, 2008 at 11:32 PM, Krishna Chandra Prajapati
[EMAIL PROTECTED] wrote:
Hi,
On myisam storage system
mysql explain select ui.user_id, ucp.user_id,ucp.payment_date from
user_info ui, user_course_payment ucp where ui.user_id=ucp.user_id;
Please actually read my reply before asking the same question. As I
stated, InnoDB outputs *estimated* row counts in EXPLAIN, whereas MyISAM
outputs *accurate* row counts.
-jay
Krishna Chandra Prajapati wrote:
Hi,
On myisam storage system
mysql explain select ui.user_id,
On Wed, Apr 2, 2008 at 5:06 AM, Krishna Chandra Prajapati
[EMAIL PROTECTED] wrote:
Horribly ugly stuff
I know I sure as heck am not going to spend half an hour to turn those
queries into something understandable, and I expect no one else will
either. If you want help please remove all
The MyISAM isn't scanning more rows. It's that the InnoDB rows output
in EXPLAIN is an estimate and the MyISAM one is accurate...
-jay
Krishna Chandra Prajapati wrote:
Hi All,
I have same table configuration, every thing same except the storage engine.
Explain result on innodb system
On Wed, Apr 2, 2008 at 8:52 AM, Jay Pipes [EMAIL PROTECTED] wrote:
The MyISAM isn't scanning more rows. It's that the InnoDB rows output in
EXPLAIN is an estimate and the MyISAM one is accurate...
-jay
Also, if he was testing one storage engine vs another he might have
dumped the table and
Hi Saravanan,
Please check http://forums.mysql.com/read.php?97,18003,18003
-Raj.
-Original Message-
From: Saravanan [mailto:[EMAIL PROTECTED]
Sent: Friday, March 21, 2008 1:59 PM
To: mysql@lists.mysql.com
Subject: innodb transaction not works
Hi lists,
We are running database with
Oh, and a followup question that I forgot to ask--what if the two
systems have different db schemas? Is it possible to do some sort of
mapping between the two?
On Fri, Mar 21, 2008 at 1:48 PM, Waynn Lue [EMAIL PROTECTED] wrote:
I'm thinking of buying a license for this tool to do a migration
Check the permission on mysql database realted files, it shoud be owned by
mysql user.
regards
anandkl
On Feb 13, 2008 4:42 PM, Saravanan [EMAIL PROTECTED] wrote:
Hi List,
I got crash when i was creating procedure and it listed the following
error
*** glibc detected ***
I have used the command below to change the file system permissions
chown -R mysql:mysql /data/mysql
On Jan 17, 2008 5:01 PM, Baron Schwartz [EMAIL PROTECTED] wrote:
Have you checked filesystem permissions where InnoDB needs to create
its files? And you're sure you've removed ALL of
I did the same as you have written, but innodb storage engine is not
available now. Even the skip-innodb is commented in my.cnf
On Jan 16, 2008 9:43 PM, Rolando Edwards [EMAIL PROTECTED] wrote:
1) 'mysqldump' all databases to an SQL file
2) Drop all databases
3) Shutdown mysqld
4) Delete
Have you checked filesystem permissions where InnoDB needs to create
its files? And you're sure you've removed ALL of InnoDB's previous
data and log files? And there's nothing in the server's error logs?
Are you looking in the right error logs? (cause an error deliberately
and look for it to be
:04 AM
To: Rolando Edwards
Cc: MySql
Subject: Re: Innodb gets disabled
I did the same as you have written, but innodb storage engine is not available
now. Even the skip-innodb is commented in my.cnf
On Jan 16, 2008 9:43 PM, Rolando Edwards [EMAIL PROTECTED]mailto:[EMAIL
PROTECTED] wrote:
1
Hi,
On Jan 16, 2008 7:52 AM, Krishna Chandra Prajapati
[EMAIL PROTECTED] wrote:
Hi all,
In order to reclaim the free space from mysql innodb storage engine. I have
stopped the mysql server, remove all the things from data (to create new
datadirectory and log files) directory, added
no errors are there in log file
On Jan 16, 2008 6:38 PM, Baron Schwartz [EMAIL PROTECTED] wrote:
Hi,
On Jan 16, 2008 7:52 AM, Krishna Chandra Prajapati
[EMAIL PROTECTED] wrote:
Hi all,
In order to reclaim the free space from mysql innodb storage engine. I
have
stopped the mysql
1) 'mysqldump' all databases to an SQL file
2) Drop all databases
3) Shutdown mysqld
4) Delete the ibdata1, ib_logfile0, ib_logfile1
5) Add innodb_file_per_table to my.cnf (which you already did)
6) Make sure you gave this setting in [mysqld] group of my.cnf
Hi all,
There is no errors. the log_error file is blank
Thanks,
Prajapati
On Jan 4, 2008 5:46 PM, Moon's Father [EMAIL PROTECTED] wrote:
You should paste all your error messages here.
On Jan 4, 2008 7:29 PM, Vitaliy Okulov [EMAIL PROTECTED] wrote:
You must copy all files from old
Hi Krishna,
Did you try this option to start mysql
cd mysql-enterprise-gpl-5.0.40-linux-x86_64-glibc23/bin
./mysqld_safe --user=mysql
On 1/4/08, Krishna Chandra Prajapati [EMAIL PROTECTED] wrote:
Hi All,
I was trying to change the data directory of mysql on debian. I have added
two thins
You must copy all files from old location, probably /var/lib/mysql to
/data/mysqldata/.
Also open your mysql error log file read it.
2008/1/4, Krishna Chandra Prajapati [EMAIL PROTECTED]:
Hi All,
I was trying to change the data directory of mysql on debian. I have added
two thins in my.cnf
You should paste all your error messages here.
On Jan 4, 2008 7:29 PM, Vitaliy Okulov [EMAIL PROTECTED] wrote:
You must copy all files from old location, probably /var/lib/mysql to
/data/mysqldata/.
Also open your mysql error log file read it.
2008/1/4, Krishna Chandra Prajapati [EMAIL
Hi Ben,
On Dec 12, 2007 8:14 AM, Ben Clewett [EMAIL PROTECTED] wrote:
Dear MySql,
Using 5.0.41 I had a single innodb table which would not unlock. I
wonder if this might be a bug, or an issue that is known to be fixed in
later versions?
Any DML like this example:
UPDATE ws_queue SET
It locks the table for both.
regards
anandkl
On 11/20/07, Thomas Raso [EMAIL PROTECTED] wrote:
Hi all,
just a simple question :
Does the query ANALYZE position reads and/or writes locks ?
I read these two pages but I didn't find the answer...
Thanks everyone for the responses. Will put me on the right track
here..something that was rolling through my head but I couldn't really
define. I will be blogging about this later as I think it is fairly
important, but often not understood by beginning/mid-level dbas.
thank again,
Keith
At 02:05 PM 11/16/2007, you wrote:
How do you import the data?
Load data from file is faster thought
so better to export myisam - file and then you do load data from file
make sure you set autocommit=0 to make it faster
Ady,
Sure but won't the entire Load Data will still be wrapped in a
Use smaller transactions that don't have 140 million rows. When attempting an
action with important data, make sure you can survive the actions failure. If
you can't, then you need to think of a different way of doing it that will
allow a recoverable failure.
- Original Message
How do you import the data?
Load data from file is faster thought
so better to export myisam - file and then you do load data from file
make sure you set autocommit=0 to make it faster
On Nov 17, 2007 12:29 AM, B. Keith Murphy [EMAIL PROTECTED] wrote:
I have something to throw out. I just
Marten,
Yeah, my experience has been that InnoDB is great when it's working
but a complete nightmare when it stops working. I have scripts to deal
with this which I'm actually hoping to release to the public in the
near future. Essentially, what you need to do is edit your my.cnf to
bring MySQL
Mariella,
Mariella Petrini wrote:
Hi All,
I have been using MySQL 5.1.x with InnoDB and Raw
Devices.
[mysqld]
innodb_data_home_dir=
innodb_data_file_path=/dev/hdd1:3Gnewraw;/dev/hdd2:2Gnewraw
...
[mysqld]
innodb_data_home_dir=
innodb_data_file_path=/dev/hdd1:5Graw;/dev/hdd2:2Graw
Is
To have a good understanding on the show innodb status output checkout
http://www.mysqlperformanceblog.com/2006/07/17/show-innodb-status-walk-through/
One area you can look at is the LATEST DETECTED DEADLOCK. But in most
cases have found calculations on the status variables more helpful.
--
Alex
Thanks a lot Alex.
regards
anandkl
On 8/29/07, Alex Arul Lurthu [EMAIL PROTECTED] wrote:
To have a good understanding on the show innodb status output checkout
http://www.mysqlperformanceblog.com/2006/07/17/show-innodb-status-walk-through/
One area you can look at is the LATEST DETECTED
Alex Arul Lurthu wrote:
To have a good understanding on the show innodb status output checkout
http://www.mysqlperformanceblog.com/2006/07/17/show-innodb-status-walk-through/
One area you can look at is the LATEST DETECTED DEADLOCK. But in most
cases have found calculations on the status
Christian Parpart wrote:
Hi all,
recently someone said to know alot about mysql told us that InnoDB is about to
be removed from the mySQL server. however, InnoDB seems to be the fastest
storage engine in our case, as myisam take a hell longer to insert new rows
e.g.
so is it true, that
Even if it is removed it would only be removed from future versions .. say 6.0.
Even version 5.1 has been feature-frozen, if I am not mistaken, meaning that
features will not be added or removed. But, as others have said, Falcon is what
I think MySQL has in mind to replace Innodb This
On Tue, Jul 24, 2007 at 11:48:38AM +0200, Christian Parpart wrote:
so is it true, that innodb is to be removed?
No, not in the forseeable future.
Jim Winstead
MySQL Inc.
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:
Jim Winstead wrote:
No, not in the forseeable future.
if you are going to use that kind of Orwellian newspeak, then expect the
followup question to be, how far into the future does MySQL see? do
you have a number in mind, like 1 year, 3 years, 3 months, 3 days?
Les Schaffer
--
MySQL General
On Tue, Jul 24, 2007 at 11:17:11AM -0400, Les Schaffer wrote:
Jim Winstead wrote:
No, not in the forseeable future.
if you are going to use that kind of Orwellian newspeak, then expect the
followup question to be, how far into the future does MySQL see? do
you have a number in mind, like 1
At 10:48 AM 7/24/2007, Jim Winstead wrote:
On Tue, Jul 24, 2007 at 11:17:11AM -0400, Les Schaffer wrote:
Jim Winstead wrote:
No, not in the forseeable future.
if you are going to use that kind of Orwellian newspeak, then expect the
followup question to be, how far into the future does
On Tue, Jul 24, 2007 at 02:49:55PM -0500, mos wrote:
Since you're here, maybe you can answer a question for me. Will
Falcon eventually replace InnoDb? Or do you have another engine in mind?
There is no effort underway to replace InnoDB. Last I heard, Oracle
intends to keep developing
... we are using file-per-table.
also, it occurred to me after sending my first email: perhaps innodb is
already doing something under the hood to make sure the tables are in
the same state on startup as they were left on shutdown
Les
--
MySQL General Mailing List
For list archives:
Hi Julien,
Do you see any InnoDB: using the same InnoDB data or log files.
InnoDB: Unable to lock /var/lib/mysql/ibdata1, error: 11
InnoDB: Check that you do not already have another mysqld process
InnoDB: using the same InnoDB data or log files.
of these error in the error log file.
: Julien Marchand [EMAIL PROTECTED]
Cc : mysql@lists.mysql.com
Envoyé le : Vendredi, 22 Juin 2007, 13h45mn 12s
Objet : Re: InnoDB: Assertion failure - MySQL keeps crashing
Hi Julien,
Do you see any InnoDB: using the same InnoDB data or log files.
InnoDB: Unable to lock /var/lib/mysql/ibdata1, error
No, I don't have this error :/ And not any file system full issue...
- Message d'origine
De : Ananda Kumar [EMAIL PROTECTED]
À : Julien Marchand [EMAIL PROTECTED]
Cc : mysql@lists.mysql.com
Envoyé le : Vendredi, 22 Juin 2007, 13h45mn 12s
Objet : Re: InnoDB: Assertion failure - MySQL
@lists.mysql.com
Envoyé le : Vendredi, 22 Juin 2007, 16h31mn 15s
Objet : Re: Re : InnoDB: Assertion failure - MySQL keeps crashing
I had this issue, so just wanted to know, even if you have the same error
message.
regards
anandkl
On 6/22/07, Julien Marchand [EMAIL PROTECTED] wrote:
No, I don't have
Did you see this part of the stack trace?
It is possible that mysqld could use up to key_buffer_size +
(record_buffer + sort_buffer)*max_connections = 182271 K bytes of memory
Hope that's ok, if not, decrease some variables in the equation
How much memory (MyISAM and InnoDB) are you allocating
PROTECTED]
À : Julien Marchand [EMAIL PROTECTED]; mysql@lists.mysql.com
Envoyé le : Vendredi, 22 Juin 2007, 21h18mn 01s
Objet : RE: InnoDB: Assertion failure - MySQL keeps crashing
Did you see this part of the stack trace?
It is possible that mysqld could use up to key_buffer_size +
(record_buffer
me know how it goes.
David
-Original Message-
From: Julien Marchand [mailto:[EMAIL PROTECTED]
Sent: June 22, 2007 12:43 PM
To: David Griffiths
Cc: mysql@lists.mysql.com
Subject: Re : InnoDB: Assertion failure - MySQL keeps crashing
Yes, and
I also saw InnoDB: Out of memory in additional
:)
- Message d'origine
De : David Griffiths [EMAIL PROTECTED]
À : Julien Marchand [EMAIL PROTECTED]
Cc : mysql@lists.mysql.com
Envoyé le : Vendredi, 22 Juin 2007, 23h31mn 27s
Objet : RE: Re : InnoDB: Assertion failure - MySQL keeps crashing
It doesn't look like a memory issue, but only you
Hi All,
If you specify one file per table, these files would be created under the
database directory of that particular database . So, the benifit with
respect to IO is negative. To have these files placed in different file
system to get IO benifit, you need to use symbolic links.
Please correct
Hi all,
Are there any reasons why one would NOT use separate ibd files for each
table
Fragmentation for one.
A single file can re-use empty space from deleted rows for any added
rows. A single file can only re-use space from that one file.
Therefore the sum table size will be larger
Olaf Stein wrote:
Hi all,
Are there any reasons why one would NOT use separate ibd files for each
table (--innodb_file_per_table). It seems logical to me to separate what
does not belong together logically (different databases), but I as the
shared tablespace is the default I wonder if it has
In the last episode (Jun 15), Ben Clewett said:
Are there any reasons why one would NOT use separate ibd files for
each table
Fragmentation for one.
A single file can re-use empty space from deleted rows for any added
rows. A single file can only re-use space from that one file.
Found the problem. After searching for a while, I found where someone
deleted the ib_logfile*.* in the data directory. I did that, and that
cleaned it up.
Jesse
- Original Message -
From: Jesse [EMAIL PROTECTED]
To: MySQL List mysql@lists.mysql.com
Sent: Saturday, May 19, 2007
Hi !
Jon Ribbens wrote:
On Tue, May 15, 2007 at 06:39:21PM -0500, Kenneth Loafman wrote:
Interesting... guess the intent was a disconnect that would break code
trying to work on MySQL, regardless of engine selected. That decision
makes it two products, MySQL/MyISAM and MySQL/InnoDB with
Folks,
Thanks for all the help. Not only is the code working, but my
understanding of the issue has improved, thanks to this list.
...Ken
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Hi Kenneth -
it appears that you need to use an explicit 'commit' command when using
InnoDB tables and Python.
Something like this:
try:
cursor.execute(INSERT INTO Test1 (s1, i1) VALUES ('Now is the
time', 5))
db.commit()
Found this on http://www.serpia.org/mysql
Thanks for the tip, that worked.
Sounds like InnoDB is still borked though. You should not have to use a
commit unless you have started a transaction, as I understand it. The
semantics for non-transaction access should be identical.
...Ken
Dan Buettner wrote:
Hi Kenneth -
it appears
Kenneth Loafman [EMAIL PROTECTED] wrote:
Sounds like InnoDB is still borked though. You should not have to use a
commit unless you have started a transaction, as I understand it. The
semantics for non-transaction access should be identical.
Are you explicitly telling Python not to use
Ofer Inbar wrote:
Kenneth Loafman [EMAIL PROTECTED] wrote:
Sounds like InnoDB is still borked though. You should not have to use a
commit unless you have started a transaction, as I understand it. The
semantics for non-transaction access should be identical.
Are you explicitly telling
On Tue, May 15, 2007 at 04:13:33PM -0500, Kenneth Loafman wrote:
Can't tell. The docs are somewhat lacking in detail, however, if I do a
db.autocommit(True) it works as it should.
Will have to dig into the API code and see if that is where the semantic
discontinuity lies.
The
Jon Ribbens wrote:
On Tue, May 15, 2007 at 04:13:33PM -0500, Kenneth Loafman wrote:
Can't tell. The docs are somewhat lacking in detail, however, if I do a
db.autocommit(True) it works as it should.
Will have to dig into the API code and see if that is where the semantic
discontinuity lies.
On Tue, May 15, 2007 at 06:39:21PM -0500, Kenneth Loafman wrote:
Interesting... guess the intent was a disconnect that would break code
trying to work on MySQL, regardless of engine selected. That decision
makes it two products, MySQL/MyISAM and MySQL/InnoDB with different
semantics. Yes,
Baron-
Thank you for the InnoDB Lock Monitor pointer. I now have a greate deal
of informaiton to digest. I will try innotop when I have a chance.
:)
-Paul
Hi Paul,
Power, Paul C. wrote:
I have an INSERT waiting for a table lock, and i do not
understand why.
---TRANSACTION 0
Hi Paul,
Power, Paul C. wrote:
I have an INSERT waiting for a table lock, and i do not understand why.
---TRANSACTION 0 308691, ACTIVE 5 sec, process no 8876, OS thread id
1296547864 inserting
mysql tables in use 1, locked 1
LOCK WAIT 1 lock struct(s), heap size 320
MySQL thread id 79126,
Hi Dan,
Thanks for this, fixed the problem perfectly when we applied it.
Andrew
On Thu, 2007-04-12 at 10:45 -0500, Dan Nelson wrote:
In the last episode (Apr 12), Andrew Simpson said:
One server had a problem while creating a backup last week. The routine
normally takes about 30 seconds,
In the last episode (Apr 12), Andrew Simpson said:
One server had a problem while creating a backup last week. The routine
normally takes about 30 seconds, but in this case went on for over 30
minutes. During this, the application was responding correctly to other
users. After a reboot,
On 3/11/07, Jean-Sebastien Pilon [EMAIL PROTECTED] wrote:
I am running a mysql database server and we experienced a power failure.
The mysql server does not want to restart because innodb is corrupted.
Version info:
Mysql version 4.1.11-Debian_4sarge7-log
Debian sarge
Reiserfs filesystem
What
Hi,
Michael Fernández M. wrote:
2 CPU Pentium III 700 Mhz Aprox.
4 GB RAM.
Redhat 7.2
Mysql version: 4.0.14-standard-log
Kernel: Kernel 2.4.18-17.7 (highmem)
It is possible that mysqld could use up to key_buffer_size +
(read_buffer_size + sort_buffer_size)*max_connections = 3666809 K bytes
of
For what you described... you will not get a fixed size...
If you have set file_per_table flag in my.cnf you might want to know
that the .ibd files in the database directory are by default
auto-extending... so those files WILL grow... along with your data...
The shared tablespaces that you
For what you described... you will not get a fixed size...
If you have set file_per_table flag in my.cnf you might want to know
that the .ibd files in the database directory are by default
auto-extending... so those files WILL grow... along with your data...
The shared tablespaces that
-Original Message-
From: Gary W. Smith [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 14, 2007 8:01 PM
To: mysql@lists.mysql.com
Subject: InnoDB fixed file size, how much is left?
I'm working on migrating an bunch of MyISAM tables over to InnoDB.
For
development we want to
Marten Lehmann wrote:
How can I check which tables are using innodb with sql? How can walk through
the tables with show databases and show tables. Thanks.
This somewhat depends on how the tables were declared. If you used
ENGINE=InnoDb; in the CREATE TABLE sequence, you'd be able to loop
: Tuesday, February 6, 2007 1:24:46 PM (GMT-0500) US/Eastern
Subject: Re: innodb madness
Marten Lehmann wrote:
How can I check which tables are using innodb with sql? How can walk through
the tables with show databases and show tables. Thanks.
This somewhat depends on how the tables were declared
Hi,
According to the manuel, Falcon is not yet optimized for performances,
so benchmarking it would not be fair.
And I do not recommand using the binary alpha release in production, you
could corrupt badly your database (some bugs has only been fixed a few
days ago concerning this
Size is an issue with InnoDB and deleting records does not reduce the size of the file. In my experience, the performance drop off
is considerable once the table reaches a certain size. And it's not a slight drop off over time.
If your table is going to get very large, I would reccommend using
On Friday 26 January 2007 06:17, Olaf Stein wrote:
From a feature perspective, I do not need full text indices,
This is about the only reason I've seen MyISAM promoted as table engine of
choice.
I know this is a very general question but it seems not to make any sense
not to use innodb
Another thing to consider is:
heh, silly mail client :). Another thing to consider is this:
http://dev.mysql.com/doc/falcon/en/index.html
Though it's Not recommended for production use, I've heard people still use
it in production environments.
--
Chris White
PHP Programmer
Interfuel
--
Hi Olaf,
I know the innodb vs myisam issue comes up quite frequently. I went
through
old threads and could not find an answer to my questions.
Generally, is there any reason/scenario not to use innodb?
From a feature perspective, I do not need full text indices, foreign keys
are usefull
At 03:54 PM 1/25/2007, you wrote:
Another thing to consider is:
heh, silly mail client :). Another thing to consider is this:
http://dev.mysql.com/doc/falcon/en/index.html
Though it's Not recommended for production use, I've heard people still use
it in production environments.
--
Chris,
On 2007-01-04 Octavian Rasnita wrote:
I have seen that by default some tables are created as InnoDB and some as
MyISAM.
I guess the table type is not chosen randomly. How is it chosen the table
engine used?
You can set a global and IIRC a database specific default for the database
type.
InnoDB supports foreign keys, MyISAM does not.
MyISAM supports full text indices, InnoDB does not.
This is unfortunate. It has kept me using MyISAM where I'd rather use
InnoDB, although fortunately none of my applications are really hampered by
it.
The only work-around I can think of is to
And is InnoDB recommended now?
It depends.. :)
Depends on... what?
I mean, if I don't need transactions, is there another reason for using
InnoDB?
If it is necessary I can build the client program without foreign keys
support also.
Thanks.
Octavian
--
MySQL General Mailing List
For
Octavian,
1) You can use MyISAM for example when you use static information in a
webpage. For example, only for store information of customers, something
like that..
2) Innodb is a engine that support ACID, you can use for transactions. For
example, load information of sales from PDA (
At 08:38 AM 1/4/2007, you wrote:
Hi,
I have seen that by default some tables are created as InnoDB and some as
MyISAM.
I guess the table type is not chosen randomly. How is it chosen the table
engine used?
And is InnoDB recommended now?
If you need transactions or RI.
Does it support
level locking, and foreign keys for MySQL
InnoDB Hot Backup - a hot backup tool for InnoDB which also backs up
MyISAM tables
http://www.innodb.com/order.php
.
From: Ratheesh K J Date: December 11 2006 10:23am
Subject: Re: Innodb log sequence error - urgent
Get Plain Text
Thanks,
I have
- Original Message -
From: Jan Kirchhoff [EMAIL PROTECTED]
To: Ratheesh K J [EMAIL PROTECTED]
Cc: mysql@lists.mysql.com
Sent: Monday, December 11, 2006 1:25 PM
Subject: Re: Innodb log sequence error - urgent
Ratheesh K J schrieb:
Hello all,
yesterday we seperated our app server
Ratheesh K J schrieb:
Hello all,
yesterday we seperated our app server and db server. We moved our 70GB of data from our app server to a new DB server. We installed MySQL 4.1.11 on the DB server.
Now the following happened. On the DB server the ibdata1 and all the databases
are the old ones
Daniel, innodb data files keep growing as needed, it even be of 20Gb
or more. If you want several data files (mainly because performance)
you need to add them in the my.cnf following the instructions in:
14.2.7. Adding and Removing InnoDB Data and Log Files
Carlos Proal wrote:
Daniel, innodb data files keep growing as needed, it even be of 20Gb
or more.
I was under the impression that you should avoid files 2GB on 32 bit
systems, which have to do some dodgy stuff to support files bigger than
2GB. Does this advice apply?
If you want several
At 10:57 AM +1100 12/4/06, Daniel Kasak wrote:
Carlos Proal wrote:
Daniel, innodb data files keep growing as needed, it even be of 20Gb
or more.
I was under the impression that you should avoid files 2GB on 32
bit systems, which have to do some dodgy stuff to support files
bigger than
Paul DuBois wrote:
At 10:57 AM +1100 12/4/06, Daniel Kasak wrote:
Carlos Proal wrote:
Daniel, innodb data files keep growing as needed, it even be of 20Gb
or more.
I was under the impression that you should avoid files 2GB on 32
bit systems, which have to do some dodgy stuff to support
At 11:34 AM +1100 12/4/06, Daniel Kasak wrote:
Paul DuBois wrote:
At 10:57 AM +1100 12/4/06, Daniel Kasak wrote:
Carlos Proal wrote:
Daniel, innodb data files keep growing as needed, it even be of 20Gb
or more.
I was under the impression that you should avoid files 2GB on 32
bit systems,
Mike Kruckenberg wrote:
mysql SET @staff_id = LAST_INSERT_ID();
Query OK, 0 rows affected (0.01 sec)
I don't know if this behaviour has changed in later versions of mysql,
but using session variables, although lovely, was the quickest way to
break replication (at least up to and including
Andre Matos wrote:
The idea is to have a audit trail to record the changes made. So, I want to
insert a new record in the staff table and right after this, insert a
record in the changes table.
SET AUTOCOMMIT=0;
START TRANSACTION;
INSERT INTO staff (`Name`) VALUES ('ABC');
INSERT INTO changes
Andre Matos wrote:
SET AUTOCOMMIT=0;
START TRANSACTION;
INSERT INTO staff (`Name`) VALUES ('ABC');
INSERT INTO changes (`Key`, `Table`, `Value`) VALUES (LAST_INSERT_ID(),
'staff', 'ABC');
COMMIT;
SET AUTOCOMMIT=1;
This works fine in my test environment, however what about many users doing
at
Thanks Mike.
I understand the possible gaps that I might have if I use the ROLLBACK.
This is acceptable in my case.
What I really want to avoid is what I am doing now: open one transaction to
insert, or update, or delete certain information and close with the commit.
Then, I get the
Andre Matos wrote:
Thanks Mike.
I understand the possible gaps that I might have if I use the ROLLBACK.
This is acceptable in my case.
What I really want to avoid is what I am doing now: open one transaction to
insert, or update, or delete certain information and close with the commit.
Then,
Thanks for all your help Mike.
Problem solved. I divided to process in two parts: one write the
insert/update/delete and then write the changes in the audit trail. All this
inside one transaction. If the first part fails, ROLLBACK. If the second
part fails, ROLLBACK, otherwise, if both were done
We ship out mySQL on our appliances in enterprise level scenarios. We
often
like to start the AUTO_INCREMENT for several tables at 10,000 -- this way
we
can reserve the lower 'block' of IDs for our own internal and 'default'
use
so all customers have the same basic database schema. It also
I came up with a work around when we encountered this. I don't remember
exactly (and I don't have access to that code anymore), but I think we
manually put a piece of code in our SQL setup scripts, before any of our
insert statements. This 'mysql command' would set the next available ID
to
201 - 300 of 1574 matches
Mail list logo
