it does work, so review your query
try auto_increment=324324;
Savvy, "didnt work" is not a very smart report *sigh*

is this a good mysql intro? http://php.about.com/od/phpwithmysql/ss/mysql_php.htm
or should i use the one at w3schools site
hi btw

tutorial

http://www.mysql.com/doc/en/Tutorial.html SQL Tutorial: http://www.sqlcourse.com and tut.php-q.net/mysql.html

archivist is that useful for php

this is a mysql chan dunno about php tutorials

i want to use mysql hosting in php

php manual

For a good MySQL tutorial get The Book
book

http://www.kitebird.com/mysql-book

salle that book needs a new edition

will 2nd edition do?

Tell Paul

because I found it on emule

salle I looked at it and the stored procedures and triggers section is bit short!
any better than O reilly "mysql stored procedure programming" for that

unStoppable buy a book you cheap fuck

I can't afford it. I'm a waiter and a starving artist

bs

a book is 20 dollars
I'm sure you can save 20 dollars

if he wants he can, like everyone else …

Teach yourself php, mysql and apache all in one — Sams Teach yourself Mysql in 21days — How to do everything with php and mysql McGraw-Hill — [Apress] Beginning PHP and MySQL — Beginning PHP, Apache, MySQL Web Development — php host & MyQQL
for dummies, 2nd edition — Professional Lamp - Linux, Apache, MySql & PHP5 Web Development (2nd edition) — Willey - MySql Bible (2002)
any good ones there?
or should I just get 2nd edition of Paul
20 dollars less means I don't have anything to eat for a week

http://flazx.com
It blocks out the download links for some countries, hope you get lucky .

Dbyt3r thank you.

If you aren't lucky enough, get a proxy running from somewhere els
e
no prob

Hello World ?

you can really find everything essential on the internet, if money is such an issue.

he's bullshitting. he wouldn't have internet or computer if he couldn't afford $20
uhe's bullshitting. he wouldn't have internet or computer if he couldn't afford $20/u

used*

Any offers on my upper case words question?

!man string f

see http://dev.mysql.com/doc/refman/5.0/en/string-functions.html

nope

ok no worries

I think somebody did a ucwords user defined function

its a handy one for sure when you are importing data

search the google god

or use a script language function

timr, there is a function for that, rtfm it
ffw
why cant people rtfm before asking? *sigh*

sorry tibyke I did have a look and only found UCASE

timr, next time look harder

no need to be an ass about it - I have done plenty rtfm in my time

tibyke, not in mysql

hm?

no there isn't so cheers for giving me hassle

i have a table with and id as a key, some data (e.g., date, size, description, etc.), and a 'counter' (integer) as fields. i would like to select by description but also return all trades with the same date and counter. how might i create this query? this what i have so far: select id, date,
size, description from table where description like '%something%' OR (help needed here!); i woul

like '%something%' wont use an index

how should that be changed? the table is indexed by description

dont use like '%blah%' use like 'blah%' or a fulltext index

thank you. i will be sure to change my queries. do you have any suggestions for the other part of the query that would allow me to pull data from rows with the same counter?

Im a mysql noob, but wont ORDER BY work?

i don't think so. i think a subquery might be necessary in the where clause, but i'm not sure how to word it

use 2 queries?

noooo

there you go

could you shed some light on the matter? it would be greatly appreciated

not or but and

http://pastebin.archlinux.org/11147
inner join fails…I don't get it

'…where fielda like 'whatever%' and fieldb = 9 and fieldc = 'yadda'…'

the other rows would have different fielda's, but the same fieldb's

so you want either condition a, or condition b?

amon_, why group by when you have a where for a particular id

Cause I'm using count() as well
Missing in the paste

dont ever past half the story and expect sense
paste

i would like to have return conditiona as well as conditionb

'…where fielda like 'whatever%' or (fieldb = 9 and fieldc = 'yadda')…'
'…where fielda like 'whatever%' or fieldb = 9…'

http://pastebin.archlinux.org/11148

i don't think that would produce the intended results. this is my situation. each row has a counter and description. i am looking for a row that matches that description, but would like to return all rows that have *the same counter as* the row that matches the description

crap
http://pastebin.archlinux.org/11149

i.e., the 'counter' cannot hardcoded in

select * from table where fieldb = (select counter from table where fielda like 'whatever%') sort of thing

SELECT *, @m:=max(bluepill) as m FROM redpill WHERE bluepill = @m GROUP BY id — why no result?

thanks, that's what i was about to try. however, the subquery could potentially return multiple counters that match. is there any way around this?

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

so i should use select * from table where fieldb in (select counter from table where field a like 'whatever%') ?

not in, =. in is used like 'select * from table where field, in (a, b, c, d, e)'

how does one get the highest of each group?

but the subquery will return multiple counters. will not = produce an error?

try it and see

hi guys!

!tell kombi_ about groupwise max

kombi_ http://jan.kneschke.de/projects/mysql/groupwise-max/ "http://dev.mysql.com/doc/refman/5.0/en/example-maximum-column-group-row.html">http://dev.mysql.com/doc/refman/5.0/en/example-maximum-column-group-row.html

using = returned 'subquery returns more than 1 row'
and in seems to have the intended result

then that is what should be used.

thanks archivist!

thanks for your help!

I have trouble finding on Google wether it is possible or not to have a cell autoincremented everytime you insert a value that is already in the table… For example: one cell would contain "helicopter" and the other cell, counting occurrences of the value of each first cell, would be
incremented every time "helicopter" is inserted… see what I mean?

Kryczek, a trigger

well, I can always first get the value of 'occurrences' and update with 'occurences + 1', but I wanted to know if I could avoid having an extra select each time..

you need to run an update statement in a trigger

CREATE TRIGGER ins_sum BEFORE INSERT ON account

FOR EACH ROW SET @sum = @sum + NEW.amount;

thumbs: looks like what i'm looking for thanks!
INSERT INTO table SET word = 'helicopter', @occurrences = @occurrences + 1;

how do you match 'NULL'
select PSI from SiteInfo where District='NULL';

INSERT and SET don't go together

http://pastebin.archlinux.org/11149 can't get my inner join to work

WHERE Distrcit IS NULL

ah

oh wrong error message, http://pastebin.archlinux.org/11150 there we go

update SiteInfo SET District IS NULL WHERE District='NULL';
i actually have some values = "NULL"
where i typed them in

apparently not but you're saying there is no better way to do it than setting up a trigger?

i want to change it to NULL

IS is a comparison operator
not an assignment operator
a trigger is the most efficient way

I think I just found better INSERT INTO table ('word') VALUES ('helicopter') ON DUPLICATE KEY UPDATE occurrences=occurrences+1

"NULL" and NULL are completely different things
"NULL" is string value, 5 characters long while NULL is absent, unknown, no value

salle, which is what one of our developers doesn't understand
and just uses 0, or "empty" or "NULL" instead

"NULL debate" is as old as relational model is

there is no debate…
there's NULL and NON-NULL

Since Codd died in 2003 one can argue the original debate is dead too …

NULL and 0 can't be used interchangeably as they can in C??

Of course NOT

ahah

Try that one: SELECT NULL = NULL, NULL != NULL, NULL = 0, NULL != 0;

unStoppable, think about NULL pointers instead of NULL values

NULL in sql hosting means "unknown"
Hence the three-way boolean logic. Each boolean expression in SQL can evaluate to True, False, NULL

ok, good to know

Unlike most of programming languages which only define True, False

NULL and 0 can't be used interchangeably in C

salle, I was thinking of adding 3-state booleans to SQLite, just to be a bastard

jacks- they can in a pointer context, 0 is a valid null pointer constant in C. but we're getting offtopic

My beloved example is if you have to store daily temperature. (Say air temp or hospitalized patient temp). At Wednesday the guy who measures the temperature is drunk and fails to measure it. What do you put in there so it doesn's skew weekly average?
Ask your programmer

salle, you don't insert a row ofcourse…

What if you have to?

or just fill them with nulls..

What if you measure several parameters?
Exactly
But your developer writes 0 or 'NULL' instead and gets into trouble to calculate averages

ofcourse at this company, you'd probably insert 999666

I know …

then have a cronjob which spawns e-mails to the developer to fix them

… and then write workarounds. While NULLs gives you simple SELECT AVG(t) … and it is always correct no matter how many NULLs there are

nah it's not my problem
I don't have the authority to just go ahead and implement the workarounds
I have to bring them up at meetings and have everybody agree to an action plan to migrate away from using stupid values instead of NULL

ok im having the problem i had before live

Hi there, I was wondering if MySQL kept any sort of command log or anything like that. Need to see commands executed by a user within the last hour. Is this possible?

mysql doesn't connect through php but the php mysql web hosting client does
help help
quickly
anyone?

no, but you can enable one

The_SNiFF, so what? let me get my crystal ball
joebrg, ~/.mysql_history

tibyke huh?

is that only via the mysql command line tool?

The_SNiFF, what? "doesnt connect" is a crap, not an error/status report

hello can i dump all databases at once or doi i have to dump each db ?

joebrg, sure.
blubs, man mysqldump

ok

ok. Is there a way of checking commands executed via PHPMyAdmin for example?

DB Error: connect failed
thats as far as i can see

joebrg, enable logging of mysqld

anyway to make the mysql client connect throught the socket file?

The_SNiFF, remote connection thru tcp/ip?

local host tcp/ip vs unix socket file. first works latter doesnt

whats your connection line?

hmm well i see mysql client can connect through the socket file
dunno its a db object and it used to works
dunno its a db object and it used to work

so connect thru socket

well localhost works aswell

http://rafb.net/p/F9vtCU55.html

smth wrong with php or smth

Does having the log feature enabled effect performance much do you know?

joebrg, depends. some i/o overhead for sure

k, I'll do some testing. Thanks

yw

som1 ?

man thats weird

ocZio, what?

check the link I pasted
everything is there

you just toss a pastebin url, and expect instant reply/help?

ofc not…

SELECT tl__id, tl__payment, tl__money, tl__timespent FROM timelogged ORDER BY tl__id DESC LIMIT 1
I get = You have an error in your sql syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'LIMIT 0, 1' at line 1

why does it work but not with apache??? and restarting apache doesn't help

anyone?

rick111, in your query, you have limit 1, in your error, you have 0,1, rethink that.
The_SNiFF, stop whining, that wont help you. and start thinking.

im thinking real hard

The_SNiFF, define "with apache"

using a php script on apache
i saw smth regarding RH about SELinux but im using debian
IAM THINKING:~

i already asked your bloody connection stringe minutes ago
string*
ocZio, you cant have subselect in insert this way, i guess

tibyke I ANSWERED it's a db object which worked few minutes ago

The_SNiFF, well, then contact your 3rd party software wendor

i can't solve my issue, can you help?

rick111, 100% sure you get it for that query?

no
90%

where is a good place to look for a mysql/php consultant to help me improve the performance of my site (currently I think the bottle neck is mysql)?

yes 100%
just tested something
if I remove, LIMIT 1 it works
but I need LIMIT 1

rick111, something must be fishy there, it should work
maybe a typo?

spooky

hello, can anybody tell me if mysqld automatically kills connections that have been idle for a certain period of time?

dhruvm, it does

Hi! I have a problem connecting to mysql from outside. I can do a telnet localhost 3306, but not from outside. iptables have no entry and the default policy is accept.

dhruvm, show variables like "%timeout%";

!tell replman about external

replman remove bind-address= and skip-networking from my.cnf and grant permission to the external 'user'@'host' and remove any firewall rules blocking port 3306

replman, skip-skip networking & bind-address, google them up

ah, ok
I can remember something like that
Thanks alot!

yw

can i change these variables?

dhruvm, sure you can, vi my.cnf

i unfortunately have to deal with mysqld installed on a windows 2003 server. i'm looking in the program files folder at my.ini, but i don't see a section for timeout variables :-/

[mysqld]
foo=bar

amonroy, there are some in here and you can use mysql consulting

mysql.com/doc is still _free_!!

are you one of them?

hehe

no I have full time job twist my arm, firewire is though
and Leithal is a Mysql person

Hey
How do you copy over the auth details when migrating a server?
mysqldump doesn't seem to dump very much of the information_schema db, even running as root

infromation schema is avirtual db, its the mysql db that matters

Well, when I did a mysqldump –all-databases -A (which I really didn't want to do), it still wasn't there

wingot, are you in write-only mode?

What mode?
:P

information_schema can not be written to, it is a read_only representation of the metadata

Ok
How do I migrate auth details
Google seems to be responding with lots of junk

move the mysql database

as I said the mysql db

which your all-databases did

Google doesn't make millions of dollars by responding with junk!

hehe

Ok. Well, I still can't log in as the users?
Sure they do. That's how they make their money
All the junk on the right and top, "featured links"
So any idea why the logins aren't working then?
Actually, the information_schema.user_privileges is still the old info
That's why

wingot, add drop database

where does mysql saves it's error?

The_SNiFF, where you tell it to

Ahh, right, drop database, as it already exists

wingot, just common sense

the config says syslog

and flush privileges after the table is in

how do i read syslog/ X:?

common sense that I was overlooking, yeah

The_SNiFF, you can set alternate logfiles, binary packages usually log into /var/log/mysql

i need to read the logs

so enable logging first, mysql.com/doc will be of great help to you

but the config files says the default is syslog!
and i need to read the error logs from past
!!

show the relevant part of the config file in pastebin

# Error logging goes to syslog. This is a Debian improvement

so check /var/log

Yep, just used the "drop database/table if exists" syntax, and can now say for certain that information_schema.user_privileges is not updated from mysqldump –all-databases -A in my case at least

ok i see nothing
this means my problems isn't the mysql server

wingot and flush privileges after the mysql db is in?

wingot, restart mysql
mysqld

Well, what I said was right. information_schema.user_privileges wasn't updated by the sql from mysqldump. it was updated from the flush privileges
Thanks

hehe

Thanks as well tibyke for the hit over the head with the common sense stick

=

hmm could add a clue stick to the bot

how do I make ndbengine the default storage engine on MySQL Cluster? when I add 'default-storage-engine=NDBENGINE' to my.cnf I get:
[ERROR] Unknown/unsupported table type: NDBENGINE
and MySQL won't start
gah. default-storage-engine=NDB. google ftw.

anyideas on why this might not be updating my database? http://pastebin.com/m7aeb6199

kippi, echo $sql, and you will see the problem
crystal ball is broken
btw end of line is missing

that just prints the mysql, dosn't give any errors

what? but copy the actual query, not that crappy php webhosting stuff
*sigh*

yeah just done that
it dosn't like key = 1

so correct that

can you give me any points on where i am going wrong? i can't seem to work it out

Guess at the error in a line of sql that will not be shown to you. Bonus points for also guessing the correct piece of sql to fix the error message which will also be hidden.
It's gunna be a wonderful day

I have a table with a million records. I am doing DELETE FROM table WHERE sn=$var on 16,000 records. THis is taking a long time. I have an index on sn. What else can i do to speed things up?

what percentage of the table are you removing?

UPDATE time SET monday1='14', monday2='12', monday3='13', monday4='11', tuesday1='', tuesday2='', tuesday3='', tuesday4='', wednesday1='', wednesday2='', wednesday3='', wednesday4='', thursday1='', thursday2='', thursday3='', thursday4='', friday1='', friday2='', friday3='', friday4='',
saturday1='', saturday2='', saturday3='', saturday4='', sunday1='', sunday2='', sunday3='', sunday4='' WHERE key='1' it dosn't like the key part

"doesn't like"?

actually i was wrong. there are 14 million records. im trying to delete around 16,000

its coming back with a syntax error

Hello and thanks for taking the time to help ignorant people with questions. I need to know the mathop for returning the value of a raised to the power of b. is this possible and if so determine the integer remainder of the division? Also is this ansi sql host
compat?

Sounds like that's the appropriate way to do it then
!man keywords

see http://dev.mysql.com/doc/refman/5.0/en/reserved-words.html

how to uninstall mysql ? its already not working its installed by cd .. i did't install by rpm.?

I'm guessing at the error message as well as the solution. Read that link.

am i in the correct channel?
im using SQL5

if it pertains to MySQL, yes

well yes it does
but I thought I was being ignored
im not used to coming here to ask questions

nah, just hang out in here for a bit
lots of people idle

ok thanks

like me

my husband just left me for some whore - he and I were working on a dream now gone if I cant do this.

EXP() in my SQL, you'd have to check the ANSI SQL standard to see if it's defined there.

it angers me. i feel like throwing this thing out the window and then die

hrm, how'd that space get in there.

where is the standard?

!man mathe

see http://dev.mysql.com/doc/refman/5.0/en/mathematical-functions.html

!man pow(

see http://dev.mysql.com/doc/refman/5.0/en/mathematical-functions.html

its not going to work
its not ansi compat

www.freepenguin.it/index-en.html

key is a reserved word ffs
bbl

!man keywords

see http://dev.mysql.com/doc/refman/5.0/en/reserved-words.html

I just installed mysql hosting and mysql-server using yum and when I try to start mysqld, it says that a timeout error occurred trying to start mysqld. Any ideas?

do a ps aux | grep mysql

0 grep

sodani_, chek my.cnf

0 grep
what am I checking for?

any typo errors

doesn't seem like there are any typos there. http://www.pastebin.ca/633243

anyone help me with an auto_increment question? Trying to move the current master to master-master.We have a couple hundred tables with auto increment. Is there any way to say from now on, all autoincrement values should be odd. Thus letting me tell the new master to make all auto_increments
even?

file /var/lib/mysql

these auto_increment_offset and auto_increment_increment settings are confusing me

directory
I should mention that I installed mysql 5 using yum with centosplus enabled. Not really sure if that makes any difference

/usr/bin/mysqld_safe –datadir=/var/lib/mysql –pid-file=/var/lib/mysql/mysql.pid

Starting mysqld daemon with databases from /var/lib/mysql
STOPPING server hosting from pid file /var/lib/mysql/mysql.pid
0 mysqld

sodani_, check binary with strace
there may be sth missing

tail /var/log/mysqld.log
Maybe your missing the static client lib

i'm trying to backup a mysqldb locally on my computer so i can configure it locally first. my hosting company has an option for "backup" basically a tar.gz filled with frm and other files. i put that in my /var/lib/mysql/blah.

1 mysqld

5 mysqld
6 [ERROR] Error message file '/usr/share/mysql/english/errmsg.sys' had only 304 error
but it should contain at least 472 error messages.
Check that the above file is the right version for this program!
6 [ERROR]

6 mysqld

now mysqld doesn't start, gets a lot of errors like cant find file ./blah/products
does that for all the frm files

I might be since I don't recall installing it
do you know what I should search for in yum?

thqks and by

select * from products;

Can't find file: './foo/products.frm' (errno: 13)
thats the error i'm getting. sorry wasn't more detailed before

Is it possible to create a table with the columns a, b and c (all integer), where c is the sum of a and b ?

!perror 13

Permission denied

File permissions
Not really. You can use a trigger to update C
Or you can use a view

Not interested in that. Thanks :9
*

SOB!….
thank you

You called me a SOB after I helped you???

sigh… that was a nice 30 min waisted
no sob was at the problem
thank you was for you

uh huh yeah
That's what they all say

lol
sigh…man that sucked :/
hmm while i'm here, any other command other then "show tables" to show tables?
like is there some sort of select * from master_table; or something like that?

You can peek in information_schema

Hey ,How do i change the password on mysql from blank to somthing

!man securing

see http://dev.mysql.com/doc/refman/5.0/en/default-privileges.html

hi
please how to select 1 or 0 if I have a special string in a field ? For example : select test is equal "something" from …

!man if(

see http://dev.mysql.com/doc/refman/5.0/en/control-flow-functions.html

mmm

ok last question, i promis. what is the easiest way to look at table column names?

hi all

desc tablename

i accidentaly deleted phpmyadmin symlink from /var/www
how can i restore it ?

That's not really a mysql question

ok , srys :/

thanks
very helpfull channel thanks a lot everyone

to check a list of numbers within a database you'd use SELECT * FROM dbtable WHERE id IN (1,2,3,4,5) right?

creating a symlink in var/www is more of a #linux question, or go checkout the man for ln

That's one way to do it. Not very effective if you've got a large number of ids to search for though

I've got a row called `id` (how original) that is an int(11) unsigned NOT NULL.. what's the command to add AUTO_INCREMENT to it?

!m sensei alter table

sensei see http://dev.mysql.com/doc/refman/5.0/en/alter-table.html

Alter table gleh ?? id int(11) UNSIGNED NOT NULL AUTO_INCREMENT ?
ALTER TABLE gleh CHANGE id … ?

inviso, whats another way?

if it's a range, you could use between. If you've got a large number of disjoint IDs, you could load them into a temp table and then join to that table.

inviso, ah thanks

if you look through the docs, there is an example of exactly what you need to do.

I am trying to capture the time and store it into a DB. I am usign INSERT INTO table (value) VALUES(CURTIME()) but when I select the records from the table it only appear 0000-00-00 00:00:00
the field is DATETIME()

Yeah, I got it

how does one use innodb_table_monitor?

Hm, when I reapplied the AUTO_INCREMENT, it changed all the ID's, why would it do that ?

CURTIME is not a datetime, it's simply a time

It for instance changed the MAX(id)

what if the function so I can capture the actual data and time

!m X3rus date time functions

Sorry - I have no idea what function you're talking about! but try http://dev.mysql.com/date time functions

!man date functions

see http://dev.mysql.com/doc/refman/5.0/en/y2k-issues.html

/sigh
http://dev.mysql.com/doc/refman/5.0/en/date-and-time-functions.html

4[inviso4] thanks

The docs imply that it resequences although I wouldn't have expected it to do that on an existing column

Me neither :/
What I'm trying to do is to move a table from innodb to myisam, but when I do: insert into table (select a,b,c from theotherdatabase.table); it complains over duplicate keys
Even though there are no duplicates what so ever in the source db

That sounds like an index mixup

Okay, but why would it try to insert values that aren't in the source table ?

I mean you did everything correctly, and hit the wrong column for unique.
on the new table

I did a show create table and only replaced InnoDB with MyISAM and went on with it, then did the insert into.. select(..
Not a lot of room for mixups :/

hmm you're right.
Try insert into table select * from other.table

I get the same :/
If I run it several times in a row, it complains about the next 20
Duplicate entry '2562' for key 'PRIMARY' ERROR 1582 (23000): Duplicate entry '2582' for key 'PRIMARY' etc..

If I run it several times in a row HUH?

truncate table; insert into table select * from other.table;

Hm, sickness.. if I ignore it and just keep going, it works!
Same amount of records, same values for id

is it possible to set the default host to connect to when using the mysqlcommand in the [client] section of my.cnf?

Ignoring errors is surely going to cause more problems later…

Indeed, but it's an interesting observation

the username as well

I've tried dropping the table and recreate it, so truncating it would make no difference, or ?

if you're running it multiple times, you clearly have no concept of what insert is doing especially with respect to unique keys

no need to be rude.

As I said, it was only an interesting observation, I don't consider this 'issue' resolved yet

Error: trying to open a table, but could not open the tablespace file '/tmp/#sql283c_1fecb_234b.ibd'!

sounds like file permissions

thats on startup after a crash. no such file exists;
keeps talking about that in my err log

ah, interesting

Maybe if you have a deeper insight have a useful suggestion?

perms/space in temp for the repair

the thing that concerns me is that you seem to be blindly poking at the database until the results end up matching what you want. You may be creating problems that you're not even aware of even though.
Insert is going to create a new row so if you run your insert multiple times, you'd end up with duplicate rows
auto_increment puts a unique key on the column which blocks duplicates
so it's only lucky happenstance that the database blocked you from doing something bad

i cant figure out how to clean the innodb tablespace of that table. drop table says no such thing, discard tablespace says the same.

Yeah, as I said it was more my curious mind that wondered what'd happen (learning by experimentation). I've sort of run out of rational ideas, which is why I cam here
Basically I've got two databases on the same mysqld process, I just want to copy an InnoDB database from one to a MyISAM database on the other

And you were indeed running insert into new_table select * from old table; multiple times?

No, I was running INSERT INTO new_table (SELECT a,b,c FROM olddb.table); and it complaiend about duplicate indexes

Only once?
ie, new table was completely empty before running it?

The first time, yes. Then I tried to have it InnoDB and then alter table Engine=MyISAM, then I tried making it MyISAM first and then add the AUTO_INCREMENT afterwards, then I by mistake pressed the up key, and noticed that the duplicate key error was 'incremented' by 20, so I did it a few more
times.
Believe me, that's not my second best, or even an attempt.

So, when the new table was empty, it gave that error?

Yes, I've both deleted from and dropped it completely between attempts.. I've been at this for quite a while at this point.
Yeah

Hey, I got this "Timeout error occurred trying to start MySQL Daemon" when trying to restart. Does anyone have any idea's?

It does have the auto_increment_increment set to 20 if that helps

Anyone know where I can find docs on getting 5.1.20 to listen for public queries?

pastebin the following: show create table new_table; show create table old_table;

That INSERT is not proper.
INSERT INTO new_table (f1,f2,f3) SELECT a,b,c FROM olddb.table;
Don't place the select inside ()'s.
Essentially, your insert failed, which is why you have no records in the new table.

http://pastebin.ca/633348

Pay attention.

I do have records in the table, and the method I've used has worked for 18 tables before it, but I'm happy to give your one a go

Revisit the docs. It's wrong.

Will try, thanks

yeah, I didn't notice the syntax earlier. Definitely a good idea to get that fixed first

hey guys, I've got a quick question. is there a way to get mysql to execute a stored proc when its modified?

Xgc: ERROR 1582 (23000): Duplicate entry '2' for key 'PRIMARY'
Same again
(with that syntax)

paste the full sql statement you used

oh triggers, neat.
I guess my other question would be… is there any way to use a global variable for a database & stored procedures?
e.g. I want to remember a certain value in memory without looking it up
and I want to use that by other stored procedures

http://pastebin.ca/633375

Timeout error occurred trying to start MySQL Daemon.
Anybody know why?

inviso, this server has been taken out of a master/master DB setup which before that was a single DB, so the id's in the source DB is not 'linear' if that's got to do with it?

that should be fine as long as they are distinct

Yeah, they are. No duplicates in the source

I know I'm missing something obvious.. grr

We're three people who's been scratching our heads over this for a while now

what version of mysql?

5.1.16

Moving from InnoDB to MyISAM, correct?

First time I run it, it says that 2 is the colliding one, which is the offset, and the next is 22, which is the offset + increment_increment
Yes, that's correct

Also, make sure you have an explicit insert field list. Don't leave that out. That could be one reason your primary keys aren't as expected.

check the error log
yeah, he's got that matched up

I'm trying to match a series of queries (of the form "query1;query2;query3"), where query3 may or may not end with a semicolon. matching on "[^;]+;" only returns query1 and query2. any suggestions?

*nod* It didn't appaer to be from the first post.

True, I've dropped the table and used your suggestion since

sorted, thanks.

Can you check the source table for home_user_id of 0 or NULL?

Yeah it exists on the source

which does?
and how many occurances

The source has one id which is 0

winner
So….

if 1 of the slave needs to be turned down for repair.. what should i do when i to put it back.. would it automatically sync on the master?

I like this

When MyISAM gets a row with (0, val, val, val, val) where the 0 is the column marked as auto-increment…
it regards that as a row it needs to generate a value for
It generates the next in line which would be 2

Ah, good point!!

immediately following, you insert another row (2, val, val, val, val)
ta da

So, what's the workaround ?

set the ID on that row to something other than 0
or insert it last
and deal with any ref integ problems that causes
Should only be that single row that doesn't match the old tables

how can I get MySQL to listen for public connection attempts? I have MySQL 5.1 running on my home box and I'm trying to connect to it with the MySQL Query Browser, but I think it's only listening on localhost right now.

I know, I know, I know!!!

!tell xbradx about external

xbradx remove bind-address= and skip-networking from my.cnf and grant permission to the external 'user'@'host' and remove any firewall rules blocking port 3306

you have to make a user and grant it the ability to connect from % or hte specified IP adress
yes _

cool, lemme check it out

mysql administrator is saying that /etc/my.cnf is not writable do I have to open this program as root?

Should socket stay as /tmp/mysql.sock ?

that's what mine's set as and it works for me

cool

Thanks a great bunch! I'd spent hours more on this!

no problem, glad we got it figured out

containsresult.txt ?

there probably is!

select * from contains into outfile "/tmp/contains"

Oh my

When would it be important to make a new database, is there size limitations that matter, or is seperate databases just a logical division of data?

Logical division

logic(nonsense)/tomfoolery

thanks dkr it worked

Thanks

If I roll back my binlogs, will it affect all databases on my server or only the databases I select?

roll back binlogs?

any input on why this is returning file_server|file_reference and not var_ip? http://pastehere.com/?xgngiw

is anyone runing the mysql clustering on centos?

field names are all correct … and the queries work independently
its just some misunderstanding with stored procs I suppose?

how to mysql server start

mysqld

i got a system where already mysql-4.1.22 already installed .. i just insatlled mysql-5.0.45 with client and installed with "rpm -i mysql"
but getting error "Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2)"?

Is it better to truncate, or 'delete from'?

that error only occurs if your server isn't up
I use delete!

hi
I have a problem with mysqldump

I dumped a databse from a mysql4.0 system and then imported it into a UTF8 5.0 and I've got character problems in content, things like smart quotes are now "¬"" — how can I fix this?

"mysqldump: Got error: 1103: Incorrect table name 'animal_bannerclient ' when using LOCK TABLES"
the table exist, have data inside, I can select in

mysql administator says that I do not have permissions to save changes to /etc/mysql/my.cnf
I am trying to enable log files.

Does the whitespace between the name and the ' got anything to do with it?
Some unprintable character data ?
(likely if you're copy-pasting from something like a word processor etc)

hi
what is the command i type at mysql command line to export a field in a table

Export where?

I don't know, I just want to dump this table
with mysqldump, as it is expected to do

With mysqldump you dump the entire table. If you want single column you can do it with SELECT col FROM tbl INTO OUTFILE '/full/path/to/the/file';
oops
sorry .. mixing two posts

I want to dump the full base
just that

Check what exactly the name of the table is
You can create table with space in the name and then it causes all sort of troubles

omfg who created that shit
thank you
didn't see the ' '

ping

does mysql write to disk if youre only doing selects?
like does it periodically do checkpoints or something

Easy to miss it Better RENAME it

it's not my base, client crap
but in fact there was an animal_bannerclient\ .frm and a animal_bannerclient.frm

Is there a reason why COUNT(*) ignores GROUP BY (or another way to get the same result) ?

and the one with space is not used, just here
makeing phpmyadmin to not list the tables, and mysqldump to fail

What do you mean?

REANAME TABLE `animal_bannerclient ` TO `animal_bannerclient_nospace`;

hi…can somebody tell me whats wrong with this SP…http://dpaste.com/15311/

there are no .MYI/.MYD aith that name

I have a COUNT(*) query which contains a GROUP BY statement. The actual SELECT * returns one row, the COUNT(*) says 2.

ALTER TABLE `ep_de_revision` DROP INDEX `rev_id`;
Incorrect table definition; there can be only one auto column and it must be defined as a key

As the GROUP BY is grouping the two rows together

so I think it's a pebkac of one of my techs

I need to drop that unique key… why can't I drop it?

Better say you have GROUP BY query which contains COUNT(*)

Hm? It's SELECT COUNT(*) FROM …. GROUP BY..

fosco .frm only means an innodb table

Post both queries on some pastebin along with result

rgr

COUNT(*) is aggregation function. Aggregation functions are most useful with queries with GROUP BY clause

prompt and got procedure created.

hmm…i was using a gui…let me try on the console
maybe some problem in gui

Something fishy here

add_new_clinic(1,2,3,4,5,6,7,9);
Of coruse I get errors

I have two identical tables, ep_de_revision and ep_en_revision… and I can drop the unique key 'rev_id' from ep_en_revision, but not from ep_de_revision

WHat is the error message?

salle, Let me pastebot the two table outputs

if I have a stored procedure that has a id CHAR(32) param, can I take id(0) and id(1) to make a single byte? those two bytes represent hex portions of one byte
e.g. FF
I want 255

http://pastebin.com/d27a75090 and http://pastebin.com/m6a7c63ae
Personal information replaced by ..

sweet, there is.

salle, http://rafb.net/p/eghwqY19.html

can you turn read-only mode on and off while a server is running?

You are doing GROUP BY on "hidden" column. One which does not appear in SELECT part

what are some common but not obvious causes for a stack trace to show up

nm

* doesn't cover it?

backtrace*

salle, I don't understand why I can't delete that key

argh … yes there is * !

Oh sorry looking the wrong one
So SELECT COUNT(*), customerID would fix it?

how can I tell if my database is set up as ISAM or INNODB?

But then any other column is non-deterministic

schema here: http://rafb.net/p/momObf47.html

Ok that doesn't fix it. Hints?

Isn't the error message clear enough?
Incorrect table definition; there can be only one auto column and it must be defined as a key

salle, No, because the tables are identical
It works with one, does not work with the other
They are _identical_

Look. The practical rule for GRUP BY query is to SELECT *only* the column you are grouping on and aggregation functions. Nothing else as it is non-deterministic
SELECT customerID, COUNT(*) FROM … GROUP BY customerID;

I tried that already, doesn't work

Its not auto-anything… why is mysql confused?

On http://pastebin.com/m6a7c63ae you don't have customerID and it is very difficult to interpet the result. This is even prohibited by the standard
"doesn't work" means nothing you know

Still says two, sorry

whats the sql command in mysql to generate the create table statement for a table ?

`rev_id` int(8) unsigned NOT NULL AUTO_INCREMENT,

salle, Exactly, _just like the one that works_

AUTO_INCREMENT must be indexed by either Primary Key or Unique key

Precisely
Which it is… its unique
I'm trying to delete the unique key
The same exact syntax and schema as another table with the same exact key and parameters
See my first rafb paste above

salle, works
thanks

| rev_id | int(8) unsigned | NO | PRI | NULL | auto_increment |
vs.
| rev_id | int(8) unsigned | NO | PRI | NULL | auto_increment |

You have auto_inc column defined in two indexes.

Can I only have one auto_increment in each db?

I wonder how did you manage to do it.

None of the tools I have shows me that
mysql cli, phpmyadmin, nothing
There is only one auto_inc here, not two
http://rafb.net/p/EWsUiO46.html
Look again

ALTER it to get rid of AUTO_INCREMENT flag, then drop the index and then add AUTO_INCREMENT again

hrm, ok, so some internal corruption then? Let me try that

If you ask me you should't be able to create such confusing definition. AUTO_INCREMENT column must belong to only one unique constraint otherwise it is not clear which one will dictate the behaviour

I didn't create this schema, its stock mediawiki

Who says two on what?

hey folks… i just got a new Red Hat 5 server and it came with mysql 5.0. When I try to connect, it gives me this error.. any idea why?? #2013 - Lost connection to MySQL server at 'reading initial communication packet', system error: 111

I can't alter it to remove it
arg

SELECT COUNT(*),customerID .. the COUNT(*) still says 2 instead of 1.

so?

anyone ?

That mans there are two rows matching the WHERE clause. What's wrogn with that?

With the GROUP BY… the corresponding SELECT * query only returns one row, I need the COUNT(*) to show me the number of rows returned after the GROUPing..

salle, Nope, it won't let me fix it

Is there anyway to get it to match the WHERE and GROUP BY

I can remove the auto_inc, delete the unique key, but then it won't let me put the auto_inc back

COUNT() doesn't match. It counts.

Too many columns ?

Not related to the issue at hand -_-

ALTER TABLE `ep_de_revision` CHANGE `rev_id` `rev_id` INT( 8 ) UNSIGNED AUTO_INCREMENT;

Incorrect table definition; there can be only one auto column and it must be defined as a key

But since you bring it up it should count the number of records returned by the entire query.

this returns 2 rows. Hence with GROUO BY you get COUNT(*) rwturning 2

is it possible to export a field with phpmyadmin?

Nope.
Where did you get that wrong impression?
Get rid of AUTO_INCREMENT for a second

That was opinion, not impressio

I did
its gone

nobody knows what might be wrong with my copy of mysql?

So is there anyway to get the number of records after the GROUP BY without reading the entire result of a SELECT * first?

is it possible to export a field with phpmyadmin?

http://rafb.net/p/9BavIN50.html

ALTER TABLE `ep_de_revision` MODIFY `rev_id` INT( 8 ) UNSIGNED; ALTER TABLE `ep_de_revision` DROP INDEX rev_id; ALTER TABLE `ep_de_revision` MODIFY `rev_id` INT( 8 ) UNSIGNED AUTO_INCREMENT;

Repating yourself won't get you anywhere, what does 'exporting' mean in this sense anyways ?

Hello

well i'd like to export the SQL into a file

dead on the second command
ALTER TABLE `ep_de_revision` DROP INDEX rev_id;
Can't DROP 'rev_id'; check that column/key exists

Very easy one SELECT COUNT(DISTINCT customerID) FROM … WHERE …

But it does exist:
| rev_id | int(8) unsigned | NO | PRI | 0 | |

I have feeling you don't understand the concept of GROUP BY at all

I surely do

It worked for me after removing AUTO_INCREMENT

The column exists, and it claims it doesn't
I'm going to drop the whole table and rebuild it

I have a table with 3 columns. The first column is "id". The others contain data about that "id"., How can I return the data from a row, by calling it's id?
Would I use SELECT - FROM - WHERE ?

And that worked perfectly, thank you very much.

Can you explain with one sentence what this does: SELECT x, COUNT(*) FROM t GROUP BY x; ?

sensei are you there?

All questions you asked so far make me think you got it wrong.

hello sensei

I think I have COUNT(*) wrong, not GROUP BY

salle, Is it normal to have a primary key on two fields?

Because that query did not return what I expected in the COUNT(*) field

Absolutely!

salle, Ok, I recreated the table using the ep_en schema, so they are absolutely identical…

Yeah, your reply didn't really clarify it for me

I need a recomendation for a mysql report tool

Did anyone understand my question?

Please define what SELECT x, COUNT(*) FROM t GROUP BY x; does. I am trying to help you.

cooner you wrote most of what you need

first steps in SQL

anyone running the cluster on centos?

Visit http://sqlcourse.com
it's free

how can I sort by a date formated like "May 15, 2007"?

Well I ran it and it didn't return what I expected as I said. I figured it'd return unique customerIDs and 70,000 in the COUNT(*) field

!m MattKelly DATE_FORMAT()

MattKelly see http://dev.mysql.com/doc/refman/5.0/en/date-and-time-functions.html

I need a recomendation for a mysql report tool (free software)

Just tell me your definition.

But it actually returned 1 in the COUNT(*) field, which was unexpected which is why I say COUNT(*) I don't get

And I will point you where you got it wrong.

I thought I just did. It will select all of the unique x and the number of rows from t. But the second part is wrong

Ok. Here it goes. By SELECT x, COUNT(*) FROM t GROUP BY x; you tell the server: Give me each distinct x and count how many times each distinct x is found.

mysql_connect() [function.mysql-connect]: Lost connection to MySQL server at 'reading initial communication packet', system error: 111 in /www/sticko/test.php on line 2
please help?
what could be causing this?

Like I said, I don't understand COUNT ( * ) like I thought I did.
I do get GROUP BY though.

Nope! What you don't understand is the aggregation performed by GROUP BY

salle, How do I drop the primary key for one field, leaving the other?
drop key rev_page; throws an error

You don't! Hey! Don't you know what composit primary key is?
composite

salle, sigh, ok… somewhere along the line, this all got fscked up

PK (x,y) enforces uniqueness of x,y as pair.
PK(first_name, last_name) allows (John, Johnson), (John, Smith), (Tom, Johnson) as long as the *combination* of the names is unique

O..k..

Here is another question: SELECT x, MAX(y) FROM t GROUP BY x;
Do you get what it does?

Guess I didn't focus enough on the DB side of things when teaching myself this stuff. Once I've got it I can perform magic with coding.. but yeah need to do more learning on SQL/MySQL

Definitely And then you can do lot of things. Here is more realistic example

Maybe if trillian wasn't converting stuff into symbols, what was the MAX?

Imagine table: (department, employee, salary)
SELECT department, COUNT(*), MIN(salary), AVG(salary), MAX(salary) FROM tbl GROUP BY department;
This will give you the number of employeesm min, average and max salaries *per department*!

Wait, COUNT(*) is counting what now?

Compare to the same without GROUP BY: SELECT COUNT(*), MIN(salary), AVG(salary), MAX(salary) FROM tbl; which will give you totals. I am assuming there is PRIMARY KEY(employee)

i'm adding an approved column to my expense table. I was going to make it boolean but then there are really three states, approved, not yet approved and declined. Shall I simply represent not yet approved by a null value or use a different data type?

hello people
i have a small question regarding stored procedures

Hey, I'm running 5.0.41-community-nt — any reason why you can't send an int(1) null default(0) column an empty '' string? It's complaining "#1366 - Incorrect integer value: '' for column 'my_column' at row 1"

is there a way to create and edit a stored procedure easily

thomash__, or even disputed

and more important; debug it?

Create a test table and play with it And visit http://sqlcourse.com too. GROUP BY is covered in second part at http://sqlcourse2.com/

I used to be able to do this on other versions of mysql.. what's the deal?

is there a reason mysql isnt letting me create a password_id field in the users table where there is already a password field ?
this seems very strange, is there a way to bypass this ?

Thanks

SQL mode

I just found something through google about that — how do I change that so that it allows such things?

Do you get Error or Warning?

an error.. I assume I'm mistakenly running it in strict mode?

what do you mean?

it seems to be very unforgiving — and I'm looking for a little slack

thomash__, another posible status

That's why it is strict

so how do I get it to relax a little bit?

so two columns?

how do I get the current timezone… ?

globally, that is — I don't want to have to rebuild this app.

ie EST or EDT

my.cnf

depending on daylight savings etc

thomash__, no just be aware that a status like that could have more states

right now I have it set to "sql-mode="STRICT_TRANS_TABLES,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION"
what should that look like?

I can't think of any more than approved, not approved and pending
in my case

and somebody disputes the expense…
does it go back to pending or …

Hm. With these flags you should get Warning not Error
forget about last one
I forgot that GLOBAL mode does not override the SESSION one

I think I got it, one sec.

Hi, i'm running mysql 5 on debian linux on a machine that has 4 Gb of ram. Everything works fine when innodb_buffer_pool_size= 2 Gb. Everything breaks when innodb_buffer_pool_size = 3 Gb. Does anyone know why?

Hey — I've got a sort of a situation here. I want to select rows based on a couple things, some which conflict. Are there OR statements in sql?

Hello, how can i easily develop,edit and debug stored procedures?

Of course there are. Please consider going through SQL basics first. http://sqlcourse.com can help you a lot

Yes there is

Let me guess. 32-bit CPU?

2007-05-10

tias

Try it and see, its quicker to type it on your system and try it than wait for one of us to tell you its ok

2007-05-10)

yes. but isn't the limit on a 32-bit CPU 4Gb?
^salle:

there is no 17th month, btw

The limit for the entire process. mysqld in this case

ah. do you know how I would go about overriding that?

And 4G is the theoretical limit. Usually it's somewhere between 2 and 2.5
Very easy! Switch to 64-bit CPU

please help me i want general advice :/

thanks for your help.

How can i easily develop,edit and debug stored procedures?

The fact that 32-bit architecture limits the address space to 4G can't be avoided.

2007-05-10?

Note the ()s

anyone know if there is a way to say, select * from tablename where any of the columns contain the number '95'
or the string 95

a AND b AND (c OR d) != a AND b AND c OR d
Redesign that table if you need something like that

I didn't design the table, i'm looking through someoe elses program to figure out how it handles data

joins

http://hashmysql.org/index.php?title=Introduction_to_Joins - For more indepth info: "http://dev.mysql.com/tech-resources/articles/mysql-db-design-ch5.pdf">http://dev.mysql.com/tech-resources/articles/mysql-db-design-ch5.pdf

hi whats the easiest way to trash everything in a mysql5 database without losing existing privileges?

I was wondering what people did to avoid the "double headed" problem of having two masters when using ucarp and mysql replication between a master and slave
i.e. start with one host acting as master and one as slave
with identical databases on each server

hello folks! what method would you recommend for a quick change of 5000 values in a table? i need to replace all records in one column with 30

oh hell, i just drop the database

heh

it is only one privilege after all ;\

You want to update every row, or ones matching a where or?

update … where…

!m robboplus update

robboplus see http://dev.mysql.com/doc/refman/5.0/en/update.html

Lathiat hm, basically every row in that column would do
hm

UPDATE table SET col1=30; ?
that would change col1 to '30' on every row of 'table'

so just force it regardless the current value

correct

archivist thank you for that one too

ok, back to the double-headed problem… The network connection goes down between server A and B, but the public interfaces for both are still answering requests

Lathiat and how about making a little "if then"?
if 0 then change to 30, if not - leave alone

well you can put a WHERE on an UPDATE just like a SELECT

now both believe they have the primary address , and both are acting as masters

read the manual page for UPDATE
example; UPDATE table SET col1=30 WHERE col1=0;

heading there for some more examples - thank you very much!

I'm not sure it's avoidable without some sort of watcher, or third machine on yet another network
bI'm not sure it's avoidable without some sort of watcher, or third machine on yet another network/b

hm right that one is very good and i will use it, Lathiat - thanks again - testing

greetings. I'm trying to install mysql on my linux box and I have it all up and running. However, type mysql -u root returns the following error:
Access denied for user 'root'@'localhost' (using password: NO)
any idea how I can change the password and get this up and running?

ChrisF-: use the local UNIX socket to get in, then change the password

^chewie: Not sure what that means

ChrisF-: mysql -S /path/to/sock -u root

ah

ChrisF-: i.e. /var/run/mysqld/mysql.sock

I'm very new to this… tahnks for the help
let me try
still getting that same error

how do i make an .sql script with such line inside so it works? UPDATE table SET col1=30 WHERE col1=0;
i mean what else should i add to tell mysql what to do?
do i just use regular commands same as in mysql shell?
so e.g. use dbname; ?

ChrisF-: the server instance is running/
ChrisF-: ?

I think it was
I just removed it
going to reinstall the package

good luck
so, is anyone using mysql replication for fault recovery in a semi-automated manner?

ok, still doing it after reinstalling
how do I stop the process?

ChrisF-: which distro are you using?

ubuntu

Ubuntu is an ancient African word meaning 'I can't configure Debian'

rofl
/etc/init.d/mysql stop
ChrisF-: then go read the docs in /usr/share/doc/mysql-server
ChrisF-: particularily those suffixed with .Debian

got it stopped…
crazy stuff
now mysql -u root is giving me an all new error

lol @the_wench

how do I select all rows with that have a certain column set to null?
ah found it (is null instead of = null)

"SELECT (a == 1 ? b : c) FROM foo"?
where a,b,c are columns in foo.

anyone know what this means? i'm confused on hows its referencing EntityID… http://rafb.net/p/QLhlNi34.html
at the bottom

what in the world could cause the error:

Is there a limit to the length of a query string given to mysql (through php)?

/etc/init.d/mysql stop
wait
not that
Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)

that perhaps the file doesn't exist there?
look at your /etc/mysql/my.cnf for the real location

but I'm getting that when type mysql -u root

or start your server

heh

ah, that's the problem… server not started
hmm /etc/init.d/mysql doesn't start it.. still getting that error… and I just installed it

perhaps
sorry, don't know, really
ok, gotta go

ChrisF-: OS?

ubuntu

Ubuntu is an ancient African word meaning 'I can't configure Debian'

ChrisF-: Did you install the mysql server package?
mysql-server or something similar to that
apt-cache search mysql | grep server

yes
so weird

yes what?

I installed the mysql server page using apt

ok, so /etc/init.d/mysql exists?
Yes

Uh, I just (re)moved ibdata1 and ib_logfile*, and now I can't (less surprisingly) drop a DB which uses InnoDB.. is there a way to sort of force it to drop ?

remove .frm files of (former) InnoDB tables

it exists.
running it returns this
Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)

Ah, cheers

I hope you didn't remove them while mysqld is running

Nops

ChrisF-: hmm, that's probably from the debian-maint user… Have you checked the error log yet?

With internationalization in Drupal 5 do I have to manage two sets of menu structures … why doesn't en/node/1 and es/node/1 point to English and Spanish versions of the same article?

drop tables like 'ep_sv%';

Whoops
Wrong ears

moin

moin moin

What is the equivalent syntax to drop a series of tables matching a regex?

haven't
I'm too new to this
gotta run though
bbl

setuid, bash

thanks

o_O

No such syntax

tibyke, I was going to suggest that next

for i in (…show tables….); do ….drop $i…;done

rm /var/db/mysql/mydb/ep_sv*

thats not smart
rather stupid

No?
Why?

not at all

for I in `echo "use db; show tables" | mysql`; do echo "drop table $I"| mysql ; done
ufor I in `echo "use db; show tables" | mysql`; do echo "drop table $I"| mysql ; done/u

sensei, you can use mysql in batch mode, no need to pipe to mysql

What tibyke said

innodb_file_per_table = 1
So that shouldn't affect the ibdata files

It does

They're never even touched

Some of the internal data always goes to the main ibdata files

Yeah they are, there's internal stuff in them

when i include something with php based on for example an action, and do exit(''); after the include statement, everything is okay except for that the newly loaded page is not shown in the location bar, and $_SERVER['self'] refers to the previous page. What cou
ld i do about this?

They haven't been touched in weeks

hi people.. somebody has used mysql code factory=
?

InnoDB internal dictionary for example which holds the info about all the tables

They won't grow in size if that's what you're referring to, as it allocates 11M and then uses that space

sylvanthis, register your nick, check the help at freenode.net
no php here, please

It can grow over that, but that'd be rather "extreme"

who use mysql code factory?

sensei, Mine is 4.6G

I'm a bit intrigued, how would you do that in mysql batch mode?

Not sure why, but I wish I could reclaim that space

Then it holds old data as well as the internal one, it can't ever be truncated

sensei, mysql -uuser -ppass dbname -e "whatever"

I'll dump all the dbs and reimport

Ah, nice! Can't believe I've gone without noticing that one

sensei, mysql –help is a nice thing

tibyke, ok thnx xD

sylvanthis

Yeah I've read that at some point, just didn't reflect on it at the time

hi, can I group a select by a timestamp, so that I group it by a day?