It's much better in some respects, because it actually works properly, i.e. you can do arbitrary sql host commands and they should worrrrrrrrrrrrrk

hehe

there is no perfect solution… Others store data on a SAN / NAS and just fire up the failover mysqld

I suppose you could do your own master-master by writing each thing to each db, and either for reading

omg

a thing to note with replication is that data corruption is not as dangerous as with shared storage solutions.

I would personally go with readonly replication with manual failover, depending on what your requirements are

so if your ibdata file breaks you might still have a reasonably good one on the slave.

mysql's replication is only reliable one direction, but it is relatively low impact on performance

hm

and there are no guarantees in replication
(google offers some patches)

There are a number of problems with replication, particlarly the difficulty of checking the replicas are in sync
you can issue nondeterministic queries which will have different results on the slave, this is a problem

with master-slave replication i went fine most of the time

In practice it rarely happens
but you must have a way of monitoring it, otherwise you may not notice until you need to failover
5.1 row-based replication partially fixes it
i.e. makes it much *harder* to write SQL which replicates badly

until now this were only master-slave setups, this is the first time i want a master-master one

There are still some things you can do which break replication completely, like if a slave shuts down in the middle of a session which needs a temporary table

i don't want to think such far *g*
perhaps i should describe the scenario of usage

You need to write guidelines for your developers and check existing code is replication-safe

i have a table with a datetime field. I want to extract all of the rows with a date of yesterday … what will the select query look like ?

NOW -1 DAY

0'), INTERVAL 1

if I do now -1 Day … that also finds elements in the last 24 hours … doesn't it ?

hi i'm using mysql on win and instructions says to do "type openfire_mysql.sql | mysql [databaseName];" however it doesn't pipe this we get an error

NOW(), INTERVAL -1 DAY

DATE_SUB(now(), INTERVAL 1 day)

the cluster is used for databasebackend to a configuration frontend for domain setups

The process tried to write to a nonexistent pipe.

for past 24 hours

pong

anyone?

Jabber server down? Or are they going to can me?

so possibly there can be many changes at the same time on the same table

um…. no one told you eh?
rough way to find out

Apparently…

try php mysql web hosting dbname the_file.sql

But I do not want the past 24 hours ..
For example, right now, I would want to return everything entry that has a date of 8/20/2007 …
nothing that had 8/21 or 8/19

ah finally, mysql 5.1

do you know how to do that

compiling on ec2 takes ages

progolferyo, what's the column type? datetime or date?

Well, you have to consider what your priorities are. I'd go with master/slave - you can potentially read data from the slaves if you really want, or just use them for backup and failover

1) transaction safe, 2) fast

how can I reset a row so all columns have their default values?

nils_ solution with master-master only with one active for write access has the nice addon the failed master will be automatically resynct

Not necessarily

s/will/might

this is probably a relatively basic question, but for a boolean field what would I enter the data as?

At the very least, you should have some operational monitoring to check that the replication is up and running
And almost certainly to check the data integrity too, maybe not so often
If a slave is stopped e.g. to backup, you need to ensure that it starts again correctly

i think to check up and running the "seconds behind master" value is fine

and monitor them so they don't get too far behind

i'm currently using nagios with mysql-replication check

Replication does not self-check, it can continue even when master and slave are wildly out of sync
the only case it will stop is if it gets an error

okay, if it gets too far behind my monitoring checks this

how do I import scheme into mysql db on win

what i cannot check is possibly if the data is really the same

datetime

hello, does anyone know how to install mysql 5 with version 4 compatibility? I install a packaged that recomipled version 5 of mysql and some of my other app's cannot find libmysqlclient.so.14 anymore becuase it was removed
under gentoo

revdep-rebuild evilstrike

if the column type is datetime, what should the query look like?

interessting

Checking the data for being the same is difficult. I spent a long time constructing a script to check it. there are some others available too

in package: gentoolkit

Anyone know of a static query I can use that has a from clause
Like select 1, but with a from clause

It's even harder for tables which don't have a primary key

yeah im installing it at the moment

you still with me ? ? ?

DATE_SUB( DATE(field), INTERVAL 1 day) AND time DATE_ADD( DATE(field), INTERVAL 1 day)

Select * from (select 1 as one) d1

they all have a such

ahhh
awesome… thanks trix

you're welcome

i appreciate it

the inner query has no from, so it will fail in this case

that should rebuild the dynamic links to the new mysql.so file

what are you trying to do? I can almost assure you that there is a different approach

yeah it seems to actually do what I need thats cool, i need it rebuilding the libmysqlclient that is nice, thx for the info

would be nice if you could drop me your script

how do I import scheme into mysql hosting db on win

'field' should be replaced with 'time' ?

Check if a connection is open, so I need some query to execute on most DBs which isn't dependant on schema

yes, sorry.

hmm … but dont you need some kind of NOW() or CURDATE() in the query

select a timestamp

it has not from clause

so

SELECT NOW()…

progolf, whoops. yeah. replace 'time' with DATE(time) and then replace DATE(field) with NOW()

hsqldb give "invalid syntax"

this query returns every entry in my db //.

Well, I can't do that very easily, it was developed commercially for my job
One easy but slow option is to select the entire contents of the tables on the master and slave (lock the master and wait for slave to catch up), and compare them, ordering by primary key
Unfortunately this approach is very inefficient, especially in terms of bandwidth if your slaves are remote

i didn't say NOW()
you did

ok this now returns all records from today
but I want yesterday only

i understand no problem, i think you and nils_ helped me to get an overview over the possibilities - i will have to choose. but i think master-master is quite the same if second master doesnt get written as master-slave

what did you mean?

progolf, err.. WHERE DATE(time)=DATE(DATE_SUB( NOW(), INTERVAL 1 day))

sorry im back … any idea how to modify that for just yesterday

give that one a try.

WHERE DATE(time)=DATE(DATE_SUB( NOW(), INTERVAL 1 day)) ?
ewww

that is ugly

chad, welcome to the conversation.

NOW() - INTERVAL 1 DAY is a bit more readable
why the double conversion to DATE() ?
just use a BETWEEN and you should be fine

did that work, progolf?

yah that worked actually

putting a function on a field kills the use indexes

great

getting some major criticism … haha

yep and should especially be avoided when it isn't needed

yeah, they didn't read your scenario
ya cant argue with progress = )

haha, thanks trix

you're welcome

hey trix, you ready for another tricky query ? i want to select all the rows from last sunday at 7pm until now

that's the same query

i'll let the optimizers take that one goin to grab some food, back in a bit

i just replace the interval 1 day with ? ..

take a piece of paper and try to find it out yourself
WEEKDAY() is the way to get the Sunday
TO_DAYS() might help to get the delta

Hey guys. I've got a backup sql file of a fairly big database. Can I import selected tables from it from the command line? I know how to do the whole dump, but I only need a few tables.

weigon … thanks for the rude yet inciteful suggestion

rude?

hehe

i just dont get some people

it was a very gentle push to think yourself

ok …

a piece of paper and a pen help a lot in such cases

yah i know … it was a bit sarcastic … you cant deny that ..

how can I combine weekday with the 7pm time though ?
weekday()

can anyone here give me some help with imagemagick?

in which way is this mysql related?

lol
i'm using a mysql db to reference the images uploaded by imagemagick

do you know how to get the date of last sunday?

no i am unclear about that
if I know that i can figure this out
can you point me in the right direction for that piece of data
last sunday at 7pm actually ..

one item at a time. How could you go about getting the date of last sunday?

hmm …

you know the date today, and you know dayofweek() will give you the day of the week as an integer. It's simple subtraction from that point right?

yes

so getting the date is easy, and datetime format is yyyy-mm-dd hh:mm:ss so you know the end will be 19:00:00.
and you have the date

ok … let me assemble it

BETWEEN (the above product) AND NOW();

Hi! I've got a backup containing multiple .MYI and .MYD files. This is data from server 1. How may I "import" those files to server hosting 2's files?

do you have the .frm?

yes

cp

copy them into a databases folder and issue a flush tables query

chown

wouldn't that delete my old databases?

if you overwrite an existing table yes

mv

yep, mv. Or copy when the server is down.

The tables that server two has can't still be there if I replace the existing files with those new files, can they?

to assemble the date, should I use makedate ?

the new files doesn't contain any info about the old tables - just a new set

no

or which function to create the date?

I have a table with one way connections, a to_user and a from_user. I want to select any users which have BOTH a from_user AND a to_user connection, i.e. the connection is mutual… How do I do this?

Hello. http://pastebin.com/m3e805515 How can give parent id as start point to get data ?

because i already have the time … so i need to combine the time with the date

if you coalesce(field1,field2) AS c will it return NULL for c?
or not set c?
(c would not be in the result set)

select coalesce(NULL, NULL) as c;
if field1, field2 are both NULL then c is NULL

I might not really get it. But I must say that it feels like all old tables will be gone when I do replace them with server 2's data.

so what if you coalesce(t1.field, t2.field) would field be NULL if both are NULL ?

yup

(so AS is not necessary to return a NULL)
ok

if you overwrite anything then yes it will be gone. You can rename the directories if you wish, though.

i looked on the manual, but found no way to GRANT a priv for a user on all databases starting with something such as "monkey_"
i.e. monkey_1, monkey_baboon, etc…

there is no such thing as far as I know.

oh, so phpmyadmin is magical then… thanks anyway

% is the wildcard you are looking for

i tried GRANT stuff ON monkey%.*, or 'monkey%'.*, but it didn't like it

why do you havea percent sign?

nope… there is a wildcard, I just forgot

0 AND NOW() … the results are

no quotes ?

yah quotes …

0' AND

well i have the logic inbetween … i dont have the date formulas …

'monkey_%'.*

if I add quotes outside of the logic … then everything is returned to me
time BETWEEN '(INTERVAL 19 HOUR + DATE(DATE_SUB( NOW(), INTERVAL (WEEKDAY(NOW())+1) day)))' AND NOW()

Whoa

lol
it actually prints out the correct date … but just the results are wrong

it complains about bad syntax that way. also bad syn for 'monkey_%.*'

Show the grant?
Or was it `monkey_%`…

5 [ERROR] /usr/local/libexec/mysqld: Can't find file: './mysql/host.frm' (errno:
How can I tell mysqld_safe where the file is?

!perror 13

Permission denied

you need to chown your files
(in the datadir)

even if I have: WHERE time BETWEEN '2007-08-19 19-00-00' AND NOW()
it prints out the wrong results …

Thanks!

it leaves out all the results with the date 2007-08-19

19-00-00?
19:00:00

hmm … i got it
mispelling or something

hi all! By default, MySQL creates tables of type MyISAM. Can I tell it to, by default, create tables of type NDBCLUSTER?

you sure can

What about SELECT queries during a transaction?

SettlerX what about them?
Bentley –default-storage-engine=blah

can i execute SELECT queries when autocommit = off and get their results instantly?

ebergen, I could kiss you!

no idea what you expect, but I would say: yes
yes

Bentley please don't

hi shrews!

Hey Baron. Heard you met with Tom recently.

ebergen, i won't - thanks!

for example i start a transaction. However, i want to get some values from database (before i commit transaction). Will it work?

Yes, he met me for supper with his son Thomas in DC

SettlerX yup

cool. super nice guy.

Why is this silly xchat client signing me in with an underscore after my name???? Bah. Yes, he's very nice. Are you in his team now?

yep

xchat does that to me too

good for you shrews! I have to go right now… was just signing in (on a new computer) to make sure I got things set up right, which I don't apparently. Hello all, goodbye all, see you soon in NYC??

you should be using Colloquy! xchat on macs sucks

I'm not on a mac

I'm struggling to make a query to find the active topics for a forum. I have a table with the posts which I figure I have to sort by the time they were posted, but how can I manage to join this with the table with the topics and have the sorting preserved throught a GROUP BY? .. or is there
some more logical way of doing it I've missed?

but I do use xchat-aqua on my mac

i assumed you were on your laptop

na
when os x comes out with multiple desktops and keyboard shortcuts for them that work I'll probably start docking my mac

maybe with Leopard

leopard is getting spaces
I hope they don't suck

!tell userman about groupwise max

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

you missed a good time last week. jeremy taught me how to use chopsticks (sort of)

haha

i prefer to use the skewer
single pointy stick beats two dull ones

typically the only food you eat with a weapon is steak

litheum failed in his attempt to teach me

that is lies
you were too stubborn to pay attention

stubborn == drunk ?

haha

and too caught up in thinking that you were being inventive by mauling your delicately prepared food with a spear!

hogwash!
American ingenuity, man

Shrews where are you from?

wait for it….
…. wait for it……

still waiting

NC

they're a bit slow where he's from

haha

damn west coasters

hey if I can use chopsticks anyone can
I'm from idaho

we like our fish fried on this side

hi

wrapped up in tinfoil with bacon or smoked where I come from

i need help
i have installed mysql 5 with 23mb installer

idaho, huh?

yup

nice
which part?

south east

ah

does idaho have "parts". all the same

yeah, it sure does

parts
it's all one state but there are parts

seen one potato, ya' seen 'em all

we're way too lazy to split it up into separate states

i bet most states have significantly different regions

yeah

time for dinner. goodnight

bye

see ya

idaho is either desert or trees
or mountains
great fishing though

hi sfannan

Can I update fields in 2 tables in 1 query? Probably yes but i want to make sure.

yes
update supports the same join syntax as select

sfannan, did you get that SQL solved?

do you know how is it in SQLite?

no
this isn't #sqlite

is there something like, configuring the maximum data I can send through my connection to the database?

not really
why would you want to do that?

actually I dont want it, but when I store a picture in a blob column, if it is big, I see half of it
so, somewhere it stops accepting data I think

max_packet but dont do that
images

http://hashmysql.org/index.php?title=Storing_files_in_the_database

I think he's using a smaller blob column
And getting truncated

max_allowed_packet won't store half an image it will just error
I'm with seekwill on this

maybe it stores all, but only send half because of some connection based limitation?

not likely

It's REALLY hard to do something like that without knowing

if you can put it in you can get it out

which JOIN should i use in UPDATE query? Left join? Inner?

SettlerX go learn about the join types

so?
umm
the_wench, ok, maybe not storing files will save my problem

How can i save field value to php mysql web hosting variable?
Select field From ….
@var = field ?

:=
= is for comparison…

MauL^, just store location of the file, but as the others said what blob type are you using

ok, :=

naughty boy

yeah well it's for blob streaming tests

trying the MyBS engine

yeah naughty me.

How is it with getting last inserted ID before committing transaction?

how do you reset the auto increment number?

read the ALTER TABLE man page
!man alter tab

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

hallo

hallo blue_n

what

lol

hehe

that was great

hi !
i can't access to my mysql function in my script
Function registration failed - duplicate name - mysql_connect in Unknown on line 0
Function registration failed - duplicate name - mysql_pconnect in Unknown on line 0
etc……same error for all the function in my apache log
mysql: Unable to register functions, unable to load in Unknown on line 0
Somebody can help me ?
i use apache2 on debian testing with php5

you seem to load the module twice…

nickm__, ok
thanks

hi..i need to convert a MSSQL database host file to mysql..anyone know if there is a tool for conversion ?

yes

besides using SQL server?

yesi don´t have a live sql hosting sever

if you have one, you can use the MySQL Migration Toolkit

yes but unfortunatley i don´t have access to mssql server
and i don´t think there is a port for linux

then get them to dump the db in a textual form
hmm is it worth someone writing a MS text schema to .sql converter

how can i "SELECT * FROM user_info WHERE another_table.active = "enabled""? Do I need to learn JOIN in order to query two tables?

bkeating, you need to do something like FROM user_info JOIN another_table ON user_info.userid=another_table.userid.
I assume the second table is something that contains activity status for each user.
If it's more like another_table has only one row, then do a separate query, or a subquery.

yes. and I think i solved it now, thanks to your help

SELECT first_name, last_name FROM user_subscriptions JOIN zua_user_flags ON user_subscriptions.user_id = zua_user_flags.iduser WHERE zua_user_flags.value = 'on';

That looks right.

seems to work right

that's a nice clean inner join.

"sum(length(sequenz)-length(replace(sequenz,"%s","")))/sum(length(sequenz))". It works, but only with limited accuracy: A=0.08. Is there a way to increase it so I can see if its 0.08111 or 0.78999 ??

join #linux-ha

the %s should be eg the letter 'A',

Is there a way to increment a key? I want to do something like UPDATE table schedule WHERE schedule_id = 10 SET schedule_id ( 1 + whatever it was before )… That way I could insert a new item at schedule_id 10. Do I have to do 1 UPDATE for each row I want to change in this way?

…ok multiplying by 100 improves it, but well….

update will work, but careful of references in other table
!m StevosICBM_ update

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

ah thanks

How can I specify this calculation to be returned with double precision?

multiply by 100? ?? * 100.0
ops, no
multiply by 100? ?? * 100e0

I tried multiplying, it helped a bit, but why can't it use double somehow?

AS DOUBLE)

100 is an integer, if integer * integer, it always return integer
100e0 is a double

but I multiplied it to the fraction

what fraction?

trying…
the fraction of occurence_of_letter_A in all rows of a field / total letters in there

int / int , not sure what this returns, is it int or double?

Hello friends
I need some help

the casting doesnt work
you mean cast(sum(length(sequenz)-length(replace(sequenz,"%s","")))/sum(length(sequenz)) as double) ?
it doesnt work.

hi wilfred_com, how's it going?

ask

"I have a question", Don't ask: "Is anyone around?" or "Can anyone help?". Just Ask The Question. Also, please read: http://workaround.org/moin/GettingHelpOnIrc

that returns an integer then cast

no, cast the source fields as DOUBLE, if they were INT

cast the first int cast( ) /

casting the end result will have no effect.

Ups, sorry

cast(sum(length(sequenz)-length(replace(sequenz,"%s",""))) as double)/sum(length(sequenz))

cast a single value

again, you're casting the end result.
cast the source fields.

Are there any form to restore de info from an .ibd file?

(1064, "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 'double)/sum(length(sequenz)),cast(sum(length(sequenz)-length(re")

alright seems like MyBS can't handle concurrent connections

I see (i think)

start simple.
don't try to write the whole query in one shot.

you mean sum(cast(length(sequenz) as double)-length(replace(sequenz,"%s","")))/sum(length(sequenz)), but it also gives the same error

no, I mean start simple.
select CAST(`field` AS DOUBLE)

is cast even supported in mysql4?

then add one iteration at a time.

update schedule SET schedule_id = schedule_id+1 WHERE schedule_id = 43;

but length returns integer

it is. Write your query from the simplest form first.

Duplicate entry '44' for key 1, and then doesn't do the update.

but the fields contain strings.
I have to cast length somehow…

you can't convert varchar to DOUBLE or DECIMAL

cast the result of length should be fine

I have no idea why but in some moment one table in my BD isn't listed, and looking in syslog say a messagge that mysql can't rename the table #sql**** to the original table

anyone know how to get around this duplicate entry stuff?

this also doesnt work: select cast(length(sequence) as double) from fasta;

what is the problem, error , or result not double

error

!man41 cast

see http://dev.mysql.com/doc/refman/4.1/en/cast-functions.html

… near 'double) from fasta' at line 1

used singed or unsigned, double not recognised

yeah, I guess..

I shutdown the server and rename the two files (.frm and .ibd) from de Operating System and start mysql again but mysql says that the table doesn't exists

now I have "cast(sum(length(sequenz)-length(replace(sequenz,"%s",""))) as unsigned)/cast(sum(length(sequenz)) as unsigned)" and it works without error, but the accuracy is the same.
that's integer anyway..

try int * 1e0 / int
or int / int * 1e0
maybe unsigned is considered as int

it the same
it's

what is the result like? any e or any decimal point?
cast to unsigned could still be considered as int

A: 0.08, C: 0.02, D: 0.05, E: 0.07, F: 0.04, G: 0.07, H: 0.03, I: 0.04, K: 0.05, L: 0.09, M: 0.02, N: 0.04, P: 0.05, Q: 0.04, R: 0.07, S: 0.08, T: 0.06, V: 0.08, W: 0.01, Y: 0.03

try the * 1e0

tried.

ok, that means it work

8.27, C: 2.29, D: 5.04, E: 6.96, F: 3.66, G: 6.56, H: 2.51, I: 4.05, K: 4.74, L: 9.47, M: 2.34, N: 3.67, P: 4.82, Q: 3.82, R: 6.93, S: 7.95, T: 5.76, V: 7.50, W: 1.10, Y: 2.54

ok, somwhow fixed to 2 decimals now
try 1e0 ?

yes
it doesnt help…

try select 1 * 1e0 / 3;

I could multiply by 10000 and then divide back when I doing the rest of the calculations in python, but its strange..
works

if that works, yours should work the same too

but 1e0*sum(length(sequenz)-length(replace(sequenz,"%s","")))/sum(length(sequenz))*1e0' doesnt work
maybe more of 'em
nope…
cool, "1e0*sum(1e0*length(sequenz)-1e0*length(replace(sequenz,"%s","")))/1e0*sum(1e0*length(sequenz))" works now!
maybe I can remove some now…

is there an easy way to do a sed operation on an entire table? like s//\n/g ?
without querying and updating every single record

!m tetsleep replace

tetsleep see http://dev.mysql.com/doc/refman/5.0/en/replace.html

huzaah there is!

Does anybody why mysql keeps increasing memory usage, even though I am not using it that often?

ok, "sum(length(sequenz)-length(replace(sequenz,"%s","")))/1e0*sum(length(sequenz))" is enough! hurray!

i think you want
!man string-functions

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

thats a different replace

thanks! I tought I tried that before… but…

ops replace always give this 2 meaning problem

Hey. Is there a channel where I can ask questions about mysql++, or is this the proper channel?

here is as good as any

sorry, it doesn't work, I just forgot a parantese..

what version?

I am trying to access a database with the use of specialised sql structures. My problem is the call to sql_create_#(…), I don't know what type to list for an enum field in the database I want to access

4.0.24_Debian-10sarge2

try 1.0 instead of 1e0

What is the best table sorting .js, most compatible and faster ?

this is sql channel, we use ORDER BY

tried. still the same

alwasy get 2 decimals?

yes

sry, I should have mentioned that I'm writing a c++ program to access a database with mysql++ library

kimseong i know how to do it in sql
but its not good to refresh the whole site

try show variables like 'div%';
then ask in web related channel

Empty set (0.01 sec) ?

is there a js channel? or whatever you use for your website
there is an option in 5.0 that specify the precision for div

so what can I do? I only have 4.0 on that server where I need to perform that calculation
and I really need more precision, or the error will be huge after what has to be done with these values

select 1 / 3, 1.0 / 3 , 1e0 / 3;
you see different result

yes, 3 different

1e0 gives you the longest decimals
so that is the best hosting for you

could it be because I am doing this though MySQLdb in python
?

cannot be

Does anybody why mysql keeps increasing memory usage, even though I am not using it that often?

argh, sorry! I was the problem, I'm just too tired… I was outputting it as %2.2f
now I SEE the effect of 1e0 as compared to 1.0 (or nothing at all), thanks for that advice with 1e0, it works!!!
cool the total is 0.999866900823
before it was 1.01
or something equaly bad
oh 1.02 even!

Store Google maps, fully zoomable and scrollable, in database text fields, without any api references. This will have great business applications. http://google-latlong.blogspot.com/2007/08/youtube-style-embeddable-maps_21.html

Hello friends any of you knows a method to restore the information from an .ibd file?

wilfred, innodb is so messy when trying to recover. I changed everything back to myisam again

I see
Is so bad
I'm many troubles with this

very bad indeed. a 15GB idb was damaged during a disk failure my backup was 2 weeks old…

But the case is my entire companie's mail is in that ibd file

the only problem with myisam is concurrent updates - that's why i moved to innodb initially.
you can try to start mysql with the recovery parameters. look in the documentation. didn't work for me back then.

All the day i'm tryng every in the help but noting
Try to create another table with the same structure and rename the .ibd file to that table but nothing
I'm desperate, my job depends of that

you work for someone who would fire you over a hard disk failure? Life is too short, get another job.

Uh, a hard disk failure shouldn't cause you to lose data. Not if you're using RAID and backups.
With no offense to the unfortunate individual asking for help.

innodb cannot rename the physical file

alter table `myTable` modify `column` date not null curdate(); - whats wrong with this?

Defaults can't be the results of functions, last I heard.

accurate.

is there a way to get around this?

timestamp

alter table `myTable` modify `column` timestamp not null; ?

!tell xange about timestamp

xange The DEFAULT clause specifies a default value for a column. With one exception, the default value must be a constant; it cannot be a function or an expression. This means, for example, that you cannot set the default for a date column to be the value of a function such as NOW() or
CURRENT_DATE. The exception is that you can specify CURRENT_TIMESTAMP as the default for a TIMESTAMP column. See Section 11.3.1.1, TIMESTAMP Pro

thanks

be aware that timestamp will be updated if you update the row.

so mysql on my server is lifo as far as data goes
but on my friends server it seems to be fifo
is there a config setting for this somewhere?
I believe I'm doing a select * from blah limit 1

there is no guaranteed order unless you use an ORDER BY statement

ok
is there anything internal I can use
or do I need to add a date field

once you start adding and deleting data from the database, there is no way to guarantee the order of storage. You need to use order by to control your result sets.

like created_on

you will need to do it yourself

ok thanks

what's that command called — I want to "dump" a whole database, all the data, in a format that I can use as a query to recreate it on another server

is there no longer a mysql-max package?

backup.sql

then 'mysql -uroot -p new_db backup.sql

ahh, great, thanks!

you should read the manual on mysqldump so you can see options, charsets etc.

ahh, got it. interesting … I had to reorder the inserts, because it didn't order them in compliance with my constraints can't expect *everything* !

hey ppls, if i had a table with a single column with 10 rows each with a digit 1 through to 10, and i wanted to do a SELECT on all of them but return an 11th row that wasnt stored in the table (to satisfy the client SELECTING from the table), what would be the best option?
basically, to append an additional row that doesnt exist

UNION

ah
thats what im after
i knew there was something, just coudlnt put my finger on it
woot got it
thx again

http://www.pennergame.de/ref.php?uid=3954

http://rafb.net/p/KIiqls22.html — given that the first query is the same as the second with an outer join added to it, shouldnt they both yeild the same result set?
the first gives me 0 rows whereas the second gives me 2 (which is correct in this case)

What's the mysql command line to dump all data in a .sql backup into a db on the server?

mysqldump

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

wow…fast response, thanks!

erm mysqldump dumps from the database INTO a file

mysql

if a table is locked for write, should a select still block for a long time ?

write will block select

I'm not sure how to use mysqldump…
threw the command line
I tried doing mysqldump –help
but nothing

dump is for export, for import use mysql

I'm wanting to take a .sql back up of some database that I backed up from a totally different location and throw it all into a db on the new server.
Not sure what steps to take

mysql -u user -p -h host the_mysqL-dump_file.sql

wow we're still on this topic

backup.sql ; get backup.sql to the other box; create the DB and user on the other box; mysql -uuser -p backup.sql

type mysql and hit tab a few times

better, mysqldump … | mysql …
mysqladmin debug and check the error log

is that really better or just shorter….?

short is good

lol

hmm. this app keeps queuing locks for the table, all of which I think are waiting on the update \ insert lock of another process

cause me I like to do things the hard way

mysqldump -uuser -p database | nc -lp 9000; … nc host 9000 | mysql -uuser -p database

argh

tetsleep, nice

shorter and no need space for the file

psilo, that's how i do all my backups.

memory

as will I, now

excellent.

you keep netcat listening all the time on the remote server?

lets say I want to put "ckill\php_db_backup.sql" and put it into mysql db "ren2007bb". What would the syntax be? I can't get it to work

it send over once created, just a small memory buffer, even to file also need the buffer

mysql -uuser -p ren2007bb ckill\php_db_backup.sql

cool
thanks tetsleep

if that works in windows
i dont even know how mysql works in windows

yea………. ……. …… anyway i was just being difficult
man i am feeling old.
the big 2 decades

it's all downhill from here, chad. I'm 24 :[

ah who highlighted me?
oh it was to chadmaynard
tab complete please… use it

uh ok

we just like to highlight you randomly cythrawll
it's how we keep busy

How to get the latest entry of something inserted in MySQL with SELECT?

SELECT … ORDER BY id DESC LIMIT 1

I think if mysql gets restarted it will fill in holes in an autoincrement id column
so that would break

ahh
MySQL Command Line Client, do i run mysql -uuser -p ren2007bb ckill\php_db_backup.sql . Windows 2003 server
mysql -uren2007 -p ren2007bb ckill\php_db_backup.sql
I can't get nothing to work

does windows have a redirector, news to me ( but I don't know much )

Thanks seekwill.

you can redirect stdout to a file in the command shell
i don't know about redirecting stdin
you can try a pipe instead
type foo.sql | mysql…..
or is it print, something like that, it's not cat.

On windows will output it to file and current output
will output to file only

you mean
'type' on windows

:P

weird name for it

mysql will never "fill in the holes" in auto_increment

the old version of isam engine will

which is still in release if you have version 3.23….

i just tested, even isam does no reuse

Is there a way to force it to? O.o

it will reuse them if they are no ids higher though

yes

why would you?

For exactly that reason (no higher ids)
What is the highest ID?
Or does that depend on the type you assigned the field?

depends on the datatype

:O… Firefox just crashed.. yay

you can insert values into the table manually to fill them in but If it's a PK i wouldn't reuse any values

Yeah, INT should be fine ^_^

maybe can use a before insert trigger to find an unsed number and force it
it will be slow though

seems worthless to me though. I can't think of a good reason

Bah, too much overhead
If its really that important, then write a script to manually shift them all down

again, if they are used inside your application structure that can be difficult
seekwill–

I don't know MySQL, but was given some pretty cumbersome MySQL code with which to work. I am trying to turn this: INSERT INTO foo(bar, baz, …) VALUES ('barval', 'bazval', …) into something which just has bar = 'barval' and baz = 'bazval'

So I took a little nap…

me too

Which is why you'd want to create a script to do it… especially if you use the IDs to relate to other tables… you'd have to make sure it works without a hitch

there are around 25 values, so lining up "the thirteenth key with the thirteenth value" is rather painful

that's the SQL syntax for inserting a new record

jp-: so no way around the difficulty correlating keys and values ?

the other option is to create a record and use it's key to do an update afterwards, might be more readable but it's another operation that has to be completed.

okay

INSERT INTO user SET foo = 'bar', foo2='bar2';

NEVER!

sorry there foo

Just kidding

lol

does that actually work ?

nope. I just like posting faulty sql

You don't need to line up the 13 rows… if you only need to insert say a username.. then just tell it so

I am creating a new record, I just was hoping for readable code

that insert might just work

Whats unreadable about it? :p

25 keys followed by 25 values

i've never seen but i'm not gonna say it won't work

Do you need to insert all 25 values?

!man insert

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

no one believe me.

I have no idea, just using the code the person doing the database gave me. I'm doing the PHP stuff

Just to let you all know!
INSERT INTO Member_data
SET Username = 'bar',
Pass_perm = 'bar2'
That works fine

that syntax he gave works Samus_Aran, according to the manual

Sorry about the way PhpMyAdmin displays it
But chadmaynard… it works

no shit?

in that case I can do that, one entry per line, and have it very readable. thank you, will give ti a try
*it

4 (Query took 0.0003 sec)
Yup yup =]

botsmack

The_wench slaps chadmaynard around the head

stop that

I have two machines, one front-end (fe1) and one database (db1). When I try to connect to db1 the authentication attempts stack (according to SHOW PROCESSLIST), making each connection attempt take longer than the last. Is there any way to avoid this?

show processlist? are you looking at the idle time?

how can I use lock tables?
for transactions

!m zed_ lock tables

zed_ see http://dev.mysql.com/doc/refman/5.0/en/lock-tables.html

:P ok

kimseong is all up on the_wench tonight

I have to say, MySQL is suprising me more as I need it more..

Yeah.
So, for example..
I have a php script which does nothing but connect to the MySQL database.

I was talking with another dev about how being able to dump with a where condition

hello, I am having trouble renaming a column. The command alter table group_member change community `group` bigint(10) not null;, which is giving me the error ERROR 1025 (HY000): Error on rename of './maharausability/#sql-3576_110' to './maharausability/group_member' (errno: 150)

And low and behold… MySQL delivers!

http://paste.dollyfish.net.nz/37e0a1

it means your php did not disconnect

^^^ is the definition of the table currently

!perror 150

Foreign key constraint is incorrectly formed

Hrm.

1025?
!perror 1025
Guess not :s

so I can't rename a column that is part of a primary key then?

only connected session will show in processlist

Right, but the state is this:

what version of MySQL?

waawaamilk, try a repair on the database first

nothing to do with pk, but a foreign key elsewhere

on the table even

5.0.32

16871 | unauthenticated user | X.X.X.X:35482 | NULL | Connect | | login | NULL
So it's the "login" step that's taking the time.

that is connecting

#facebook
test what I said there

hmm. found something in show engine innodb status
http://paste.dollyfish.net.nz/b70358
le-strange

shouldn't you have two columns in that FK constraint?
if they're part of one primary key, I mean

um

oh wait, no\

thanks for the help. I assume it works, but can't test it .. they forgot to give me the database username/password/db name

so, that's one of two FK constraints on that table of two primary keys

goodnight

sleep tight Samus_Aran

can I get a dump of the constraints somehow?
I have them in postgres and xmldb form
Foreign-key constraints: "mahara_groumemb_gro_fk" FOREIGN KEY ("group") REFERENCES mahara_group(id) "mahara_groumemb_mem_fk" FOREIGN KEY (member) REFERENCES mahara_usr(id)
gaah
http://paste.dollyfish.net.nz/69b911
so yes, there's one constraint on 'member', and one o the 'community' column which I wish to rename to 'group'

the table names are different waawaamilk, mahara_group vs group

yes, the postgres db is using a prefix
that's where I got the fk dump from

oh right

hmmmm
in the table it references, the default is NULL
whereas there is no default for the group_member table
perhaps this is the problem

unique columns shouldn't be nullable…

le-grrr
I wonder if xmldb is pissing on my database again
hmm, I think it might be. The default is NULL for a whole raft of such columns

when i have two mysql servers master/slave configuration. if the slave looses a connection for an extended period of time (say 1hr)(due to a server shutdown) does when mysql starts again on the slave it just gets all changes since it was last up?

unless you delete logs on the master

which would be stupid to do so i wont be doing that

the documentation is strangely silent… can you drop a constraint on a table in mysql 5?

hi to all

Hi

turns out I need drop foreign key, not drop constraint

Can someone point me to a guide on how to do a binary copy of a database's data files and get another server to attach the database? I need to pull a database off a dead machine's drive, but despite copying it to /var/lib/mysql/database_name, and having it show up with "show databases;" and
display tables with "show tables;", I can't dump with mysqldump.

hi there, I'm having no joy with date conversion..
I can get this far, but I can;t get the time to work! can anyone help..
2', '%d/%m/%Y
dies whenever Itry and put the time in..
I can't find a reference anywhere for the valid dateformat tokens…
narrowed it down
I jsut can't get the minute..

try to find out what your sql backend is (openoffice base it is hsqlbd) and then search the right api.
not all *sql* functions work in a different definition of sql
i.e: my sql, mssql, hsqldb, etc.
Might also want to try to have a ; between the date and time instead of a space.
(makes it easier for the system to distinct them)

fooo-nix thanks

np, but remember, I am a sql new being two.
(two, where I am sorry, I may not imply you are)
There probably is a tokenizer in the sql that won't parse spaces, whilst it is expecting a space, it is getting a :.
(ignores the space, gets 12 in stead of space, yes, that will happen.

foo-nix. I appreciate your help, but I don't think you're right. that function str_to_date expects a format for it to parse
the point of the function is that it can understand dates that are not formatted in a specific way

Have you tried the ; thing?

What is the error?

as long as you proved thedate format.
it works but I can't find the token for minutes
that's all
I don't need them, but it would be nice to know for future

foo-nix: ; is the standard terminator symbol.. I don't think that will help

e.g. %d is day %H is hour
what the hell is minute?

Ahh

It's not %M..

Wel maybe a ,

Minute I think is n

as I say it don't need it..
clearly

try %n or %nn

foo-nix. I appreciate your help, but you're misudnerstanding.
the point of this function is to parse irregular date formats
so saying things like, it needs a space, or semicolon,

I will shut it then

well I aprreciate your being helpful, but it doesn't make sense
don't shut it, thanks for helping

or even %i
Try %i

I get the point, Ilike I said, I am a new being.

and now youve learned something too.: str_to_date lets you specify the formatting…
so we are all learning. YAY!
Thecks %i.. okay

huraay!

spot on!

=P

where did you find that? I searched for a good 10 mins!

I use hsqldb, it uses java, and therefore the java tokenizer, which is often very confusing when spaces are included. (A space is no token for that tokenizer)

Its the time format for PHP

right.. wellt here we go! thanks Thecks

%nn popped to mind for some reason, then I thought PHP and %i
No prob

I am BACK!

foo-nix, I'm typing sql directly into sqlyog, the language I'll ultimately use this query in is irrelevant, I'm afraid. mysql is a db in it's own right, not just php/java's slave :p

PLEASE welcome me!

Wecome winwinwin

infrid good!

please, feel welcome (I mean, wtf?)

f00-nix

Ping?

IT IS foo-nix: YOU NASTY PERSON!
:P

slumming?

how can I add mysqld_safe to startup in freebsd6

Is there a startup script somewhere? /etc/init.d/mysqld ?

f00-nix

no

How did you install it?

i installed from freebsd ports
make config, make make install i belive

I'm fairly certain it has a startup script.
freebsd

its not in rc.d

how about mysql-server ?
find / -name 'mysql-server'

. /usr/local/etc/rc.d/mysql-server
there it is

Question, if i wanna Query my DB and return 1 value (the ID number from the last entry) is this a good way of doing it SELECT id FROM tbl_shipping ORDER BY id LIMIT 1; only thing i dont know is how to order them opposite to what they are so the latest number is @ row 1 so itll work, any
help?

ORDER BY id DESC
ORDER BY id DESC LIMIT 1

thanks

wrong
SELECT_LAST_INSERT_ID() is probably waht you want

That's only if you just inserted the row.

max() ?

im going to assume thats what he just did

That's a bad assumption. You probably should mention that.

that or hes trying to implement auto_increment manually

should have …
Or maybe he just wants the "most recent entry"?

assume a forum db, the most recent post (including edits) may not necessarily be the last ID in the table
but we will see what Giddion55 says
what are you using htis id for?

Let's just assume we shouldn't assume?

If i have a table A and a table B both of which contain fields 'id' is there a way to select all the id's from A that don't appear in B?

a not in b

SELECT a.* FROM a LEFT JOIN b ON a.id = b.id WHERE b.id IS NULL;

Hi!

table1 & table2. Table1 has fewer fields. Table2 additional fields and many records. I would like to copy all the data (and additional fields) to table1. Is there a query I can use to do this?

mysql_connect() [function.mysql-connect]: Lost connection to MySQL server during query in /home/mydomain/public_html/myfile.php on line 20"
Is there a way to fix that?

No, but you have to investigate why the connection gets broken. Could be mysqld crash among others

oh
not my server, so I don't have the tools I need.

haven't seen you ’round in a while.

INSERT INTO new_table (col1, col2 … colN) SELECT * FROM old_table;
Provide a list of common columns only.

I was afraid of that. I have too many cols…

StevenBao! Woah..
That was really a long time ago

yeah.

Where is the mysql password hash physically stored?

In user.MYD

and by default it's only viewable by root?

It's viewable by any user that has access to that table
Default, yes

thanks.

Anyone know of a place to get precompiled windows binaries with the "microsecond patch" (http://www.mysqlperformanceblog.com/2006/09/06/slow-query-log-analyzes-tools/)?

Hmm I seem to have a table that got locked and then for some reason never unlocked
What are my options heh

Problem solved, seemed like I provided the wrong connection settings.

you want it unlocked?

yeah
It's hangin' my scripts!

When I get result from UNION query, will I have different columns in one query's rows and the second's rows?

do you hvae any idea _why_ these tables got locked?

no, not allowed

col2, col3, col4; Table 2: col5, col10 - and i get one row from table 1 and after UNION - table 2
so? isn't possible?

yes, my code dide it.

like, having a mysqldump running ?
usually you only can unlock tables in the same thread as you locked them

but I guess it crashed

http://dev.mysql.com/doc/refman/5.0/en/lock-tables.html
an "unlock tables" could do the trick

tried
But like you said, from a different thread.
The original locking thread is long gone
Thinking about just restarting mysqld
hehe.. I have some windows in my blood I guess.
right. think of it as one query, in a way
one query can only have one set of columns

3´ó¼ºÃ

some people are too eager to leave the channel, without waiting for full answers. why not leave the client connected, to see if anyone has anything to add after a few minutes?

Simple inexperience with IRC is my guess
They don't realize IRC crawwwlllsss
and sometimes the perfect solution comes hours later

they seem to expect everyone to have 100% of their attention fixed on their channel. many of us are doing other things too, and scan the irc client once in a while
anyways, a common solution to SettlerX 's problem is select col2,col3,col4,null as col5,NULL as col10 from table1 UNION ALL select NULL,NULL,NULL,col5,col10 from table2
although i do not know why not just use 2 queries

Ah true, had not considered selecting nulls

Is there a better way to do random selection right now I am running "SELECT * FROM glitters ORDER BY RAND() LIMIT 5" but I get an ERROR 12, which is memory allocation error.

if you get an ERROR 12, you might have quite other problems…

random

http://jan.kneschke.de/projects/mysql/order-by-rand/

read that

bigfoot-: wjhat do you mean?
ok

Hi! I've got some .MYI .MYD and .frm files. Those are from a MySQL 5 server. How may I convert them to MySQL 4.1?

mysqldump + import

The server from which they are from is crashed. I can't use mysqldump on that machine =(
Or can I perform mysqldump on those files?

you can reinstall mysql 5

on my "new" server

any computer, even your desktop PC

what about installing mysql5 on a different machine, use the data-files there, make the dump and reimport them?

that's a good on!
one*

hi, I have some troubles creating a VIEW that have an UNION in it..

and ?

where can I paste the SQL?

rafb.net

http://rafb.net/p/FUophr38.html
it gives me #1064 error

you got more than just a error-number, paste that too
you open ) which you don't close

it's a copy-n-paste error

seekwill thats a bit early fer pinging me

Oh, sorry

I was at home asleep

wait … I'm looking for another way to create te view…

paste everything you get, the way you get it

http://rafb.net/p/oyyP0q77.html
the error is " #1064 - 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 'UNION
SELECT P.id AS idPersona, S.id AS idScheda
FROM persone P
JOIN schede ' at line 4"
I manage to create the view using other JOINs…

another one (thePanz) leaves the channel too soon

is there a way to rename a table ?

!man alter

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

!man rename

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

hi
ERROR 1045 (28000): Access denied for user 'root'@'localhost' (using password: N
i am also geting same error for user which has more previlages

Check your password is being entered correctly

i have not set any password yet

i am also geting same error for user which has more previlages - more privileges from root ?

how can yes by default ubuntu seems to give not more previlages to root
ok it's done. I ran mysql -u root -p and set the password
thanks anyways
can you please any GUI for Mysql administration?
not web based

hi, where can I get this 'sense of humor'? Does it cost much? Will it hurt?

Adnan_, MySQL Administrator and MySQL Query Browser (both by MySQL AB)
look on the mysql site

arthas242, yes it costs. deposit $200 in my paypal account. yes it will hurt.

hmm, very reasonable..
ok, I sent it

arthas242, you have been conned!

you could also try teehee.com, it might not hurt, but it's definatly scary
(and probably expensive)

thANKyo
bye

see ya

Hello everyone..
I have a problem with mysql doing a dump.. the error after the dump states mysqldump: Got error: 2013: Lost connection to MySQL server during query when retrieving data from server

hi

When googling around.. i have not found a clear answer on whats going wrong..
Can someone please help me out? many thanks

There could be a lot of reasons that it lost connection

its only on dumps..
we use the database for a few things like internal wiki , RT and mantis
cd var

How big is the dump?

I think ~4gb
sorry 1gb for a specific database
the problem is only on a specific database

Perhaps you have some big BLOBs and your problem is related to the max_allowed_packet

thanks .. how do i change that max_allowed_packet?

Is there a default time limit for MySQL connections?
Could it be that its reaching the time limit so MySQL closes the connection?

just looking at /var/lib/mysql/rt3 it states the database is 2.6gb..

SHOW VARIABLES LIKE "%timeout%;
Then read in the manual about interactive_timeout and wait_timeout

Max allowed packet is 16mb..
stupid question…. I know mysqldump backs up the structure and content from queries but.. why not backup the filesystem of /var/lib/mysql?

Just a guess - mysqldump output is portable ??

the files might not have all the info in them.. some of it gets written to a log

stmaher, can give problems, mysqldump locks to give a clean copy,

oh ye, thats it backing up the files is fine, but you have to shut down the server to do it or u might get incinsistent data

thanks.. I presume i can try upping the max_allow_packet from 16 to say 32? and /etc/init.d/mysql reload ???

stmaher, it may be dynamic .. set global max_allow_pakcet = 32*1024*1024;
quit, reconnect
see if it allows you
there's a section that shows all the variables and if they can be changed on fly

[mysqld]
set-variable = max_allowed_packet=16M
this is what i have
in /etc/my.cnf

when connecting from localhost OK, but with same username from remote : "Can't connect to MySQL server on 'remote.server.name' (111)" : any idea ?

external

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 and make sure no overrides on the mysqld commandline

thanks : it was the bind-address

I have encoding problem, I insert record in table and in phpmyadmin it looks fine, all symbols are rigth. When I display this with php all not latin characters replaced with question mark. MySQL charset: UTF-8 Unicode (utf8). page charset also utf-8.

check/set your connection charset as well

where do I need to look fot this ?
I have a windows 2003 server, mysql 5.0.41-community-nt

its set during the connection from php

also have these lines in my.ini default-character-set=utf-8
init-connect='SET NAMES utf8'
ok, will do

anyone had libmysqld 5.1.x compiled and running?

when did union come into mysql?
I get "You have an error in your SQL syntax near 'union select 2" with SELECT 1 UNION SELECT 2

libstdc++.so.6

4.0

how i can remove duplicate lines on a table? (Just ID differs)

Add PRIMARY KEY on ID

mh?

oops

ID is different
it is a primary key

on evertything else than ID

but then it crashs
i know this effect

ello everybody

?

wait i try again, actually as far as i remember it doesnt allow me todo this
"duplicate entry"
i just got 2 rows int he table
X_ID and Y_ID
and i want just any combination ones….

What crash you are talking about?
You mean two columns I hope

the adding of a unique index ontop of this fields gives an error then of course: "duplicate entry"
aehm yes

dupes

find them with select count(dupefield) as qty,otherfields from table group by dupefield having qty 1

ALTER IGNORE IGNORE your_tbl ADD UNIQUE KEY (X_ID, Y_ID);
Note IGNORE keyword

ok trying
doesnt eat IGNORE

oops
it's ALTER IGNORE TABLE …
typo above

Hello… is it just me or is the linux mysql query browser missing the regex import tool?
:P

ok
still….
but exactly on my table name?!
ALTER IGNORE my_table, right?
ACH ALTER IGNORE TABLE my_table ok

that max packet size didnt work.. the dump is still hitting 70998029 in size.. then bailing

Did you increase max_allowed_packet at both mysqld and mysqldump side?

mysqldump3

I increased it from 16 to 64mb

in an UPDATE statement i want to modify a field value and just increase it by one. can i do this in a single UPDATE statement like "UPDATE tbl_records SET count=tbl_records.count+1 WHERE …" or so?

y

thx

i installed mysql-4.0.23-0 working fine … i just install "php-mysql-4.4.4-1.0.rhel4.ct" for communicating php files will it work with my installed mysql-4.0.23 or do i need to change it ?? or can any php-mysql version will work with mysql-4.0.23 … or do i need to change somethin?

hi

with database replication master/slave is it possible/recommended to replicate the mysql/privlidges table, or should users be created manually on each different server

Better maintain different users and privileges.

i thought that might be the case, but i thought

how do you comment out a line in .sql file?

usually you give different privileges for Master and Slave

so it's skipped and doesn't cause errors when loading such sq

!man comments

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

can't you just tell me?
i have no browser fired up
is it // or else?
*shrugs*

You have the manual uploaded along with MySQL itself
/* */ or — or #

thank you
and really, if it was all about reading man then this channel isn't really needed
so aren't people like you
as EVERYTHING is somewhere in the man or google
and i do know it
just can't wait when you ask YOUR question to tell you to read man

Asking other people to read themanual for you is not very polite to say the least

salle i'm not asking those who need to read man
and when i'm helping i don't read man either
if i can help, i just do it
it was easier to type /* or # instead of !man comments, wasn't it?

Your words imply people who write manuals are wasting their time you know

but you just like to brag
salle as i said - i will wait till you ask any question here
and i will ask you then wtf don't you read the damned man and waste others time

You will have to wait a lot for me asking for help about MySQL here in this channel

yes, because all you do is tell others to read man
i can setup a bot for that if you like
again, was it easier to type # or !man comments?

!man
For sure
Any other question?
May I ask you one?
Since when you are helping people on this channel?

So I've added a user to mysql. GRANT ALL PRIVILE…., * it comes back and tells me that 0 rows have been effected …
Do I have to restart mysql after adding a new user ?
what's the optimum way to reload/restart it ?

babo it does that, i dunno why

can I just restart the daemon ? Or is there a better way ?
can I just restart the daemon ? Or is there a better way ?