| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Comments:
<0> i actually can't wait till it gets cold. can't get to dam sleep at night its too hot
<1> csfreak: Sounds a little odd. It might be better to provide access for the subscribers to feed data into the common database.
<0> hows your php knowledge Xgc
<2> Xgc - yeh they do that, but the common database (or the person who uses it) needs to be able to communicate back to the proper subscriber.
<3> csfreak: when you get stuck with queries, this is generally an indication of bad design ;-)
<1> akumanz: That's not my area.
<0> gutts
<4> anyone help meee?:)
<0> whats your problem KaiT
<4> if I have a field called... section... that has a value 1 or 2... is there an easy way in one query to count rows that section=1, and that section=2?
<1> csfreak: Consider creating interfaces in the common database that each subscriber can access, for reading and writing.
<0> yeah..
<0> you mean
<0> you want to get 2 counts in one query?
<1> csfreak: and not one table (or set of tables) per subscriber.
<5> hm i thought a LEFT join should give nulls for missing values?
<4> akumanz: group by, right..?
<0> why do you wanna do it in one query.. just do it in 2 dont complicate things
<4> lol
<0> and just lookup the mysql manual for how to do COUNT
<0> doneee
<4> COUNT(*) i knoww... lol
<0> yeah sweet you're sorted then
<4> k, ya, lol
<0> awesome
<0> i came here for help and i ended up helping
<0> crazy world
<4> well im not listening to you... im doing it in 1 query. :) lol
<0> kiss my a**
<0> jokes
<0> KISS - keep it simple stupid
<1> Kaitlyn: SELECT field1, count(*) FROM table1 WHERE field1 IN (1,2) GROUP BY field1;
<0> if you try to be l33t and do it in 1 query you'll just take twice the time
<0> or Xgc will help like that and you'll be sweet -_-!
<4> Xgc: ya i did something similar to that
<6> union
<1> Kaitlyn: Remove the WHERE clause if you wish to include all values of field1.
<4> Xgc: just realized... if NO rows have field=1 OR none have field=2... then won't get a record and it'll screw up my code! Im always expecting two rows... some way to force the count of 1 even if none are there?
<1> Kaitlyn: Use a LEFT JOIN to handle that case.
<4> hmm.. okay
<7> Hi need some help in install sugar suite 4.01a and mysql
<8> is it true that libmysqlclient always "caches" whole result sets in memory??
<8> http://dev.mysql.com/doc/refman/5.0/en/mysql-store-result.html suggests so
<9> bah
<9> ndbcluster doesn't like me.
<7> I get mysql not available while it is running and accesable by phpmyadmin
<1> Kaitlyn: Something like... SELECT v1.field1, count(t2.field1) FROM (SELECT 1 field1 UNION SELECT 2) v1 LEFT JOIN table2 t2 ON t2.field1=v1.field1 GROUP BY v1.field1;
<4> Xgc: somehow I feel it would be more efficient to do two queries?
<9> When using NDB, do I have to specify the creation of database themselves as using ENGINE=NDB?
<9> Or is it just tables themselves
<1> Kaitlyn: This is going to be pretty efficient. You probably want to avoid round trips to the server.
<5> could somebody have a quick look at http://hashmysql.org/paste/viewentry.php?id=1467 ?
<10> Jax, RIGHT JONI ?
<10> s/JONI/JOIN/
<1> Jax: Show the interface table and inserts.
<5> http://hashmysql.org/paste/viewentry.php?id=1468
<5> at bottom
<1> He has the orientation correct, if the idea is to show all language_tags.
<5> yeah...
<11> jax, how many rows do you see in `interface` where LanguageID = 1?
<5> 2
<11> exactly.
<1> But you will only get 1 and null. But you have no nulls.
<5> ok, i need a query to do what i want :D
<5> i thought i could do that with left join.. guess not
<1> Jax: There are no missing entries.
<1> Jax: What did you want?
<5> i need ALL tags from language_tags in the output
<5> and for missing values in interface, i want a null or a default value.
<1> Jax: Move the WHERE id=1 to the ON clause.
<1> Jax: Remove the WHERE ... is NULL as well.
<1> jax: LEFT JOIN `interface` ON `language_tags`.`TagID` = `interface`.`TagID`
<1> Ooops.
<1> jax: LEFT JOIN `interface` ON `language_tags`.`TagID` = `interface`.`TagID` AND `interface`.`LanguageID` =1
<1> Jax: That's it. No WHERE clause at all.
<1> Earth to Jax!
<9> So... when I create a database in a SQL node which is part of an NDB cluster, why are files created on my SQL node, such as InnoDB and MyISAMI folders, directories, etc etc. Do I need to specify some special create options for database which need to use the NDB cluster engine, or is taht only specified in the tables?
<5> wait
<5> sorry
<1> Jax: You're supposed to be very happy now.
<1> Jax: Just copy/paste. Don't think.
<5> thanks Xgc
<5> worked ;)
<9> Is there a way to pull a list of database present in an NDB cluster using some command-line tool?
<1> Jax: Just try to remember, JOIN criteria really belongs in the ON clause.
<9> I don't expect to be able to see the table data, I just want a list of database in this cluster.
<5> true
<5> i'm dumb most of the time.
<1> Jax: That's probably not true. Wrapping your mind around SQL is sometimes a tough process at first.
<9> hm.
<0> is there anyway to reverse queries run on a mysql db?
<11> transactional rollback
<0> how do you do that
<11> but if you didn't set it up that way to begin with, no.
<0> ffs.
<9> haha
<0> mysql 4.0 btw.. is this a v5 thing
<0> ffssssssssssssss
<11> 4.1 on
<0> ..
<1> Manually, it's better if you make a backup before attempting something you could regret.
<11> using InnoDB and transaction syntax. if you need it, use it. there's no reason not to use 5.0
<0> yeah I made one moronic mistake today.. not thinking about the query before i executed it
<1> akumanz: or just don't commit until you've verified the result.
<0> yeah i fkd up.
<0> lame
<0> whats best way to setup a system of auto hourly mysql database backups?
<9> If I create a database in a SQL node of an NDB cluster, I should see that table have been created from any other SQL node, yes?
<0> ****KKKKKKKKKKKKKKKKKK
<0> i just fkd database even more with another query
<0> i'm gonna cry
<0> WHY AM I NOT THINKING STRAIGHT TODAY
<0> i'm screwed
<0> so screwed.
<12> Calm down.
<0> bro. calm down at what
<0> i have about 6000 images uploaded in database
<3> danq: with 5.0 yes. it has table discovery. not db.
<0> not all of them are named by product id
<9> Oh, I see.
<12> That ****s.
<0> so those ones that aren't named by product id i'm fkd i can't get them back on
<12> But you don't have to explain all your problems to us.
<9> Seems odd that I create the database in the SQL node, and it's not created in the cluster, no? Am I not thinking about it properly here?
<13> hi lokus
<3> danq: a db is something cluster doesn't care about. it starts caring when there's a table. it's a bit quirky that way, but as long as you know...
<9> I see.
<3> anyone want to add their vote to http://bugs.mysql.com/bug.php?id=17501 ?
<9> How does one manage to populate a new SQL node about database information pertaining to the, well, database at hand, if it's not stored in the cluster?
<3> danq: what db info.
<9> well, information that the db actually exists.
<3> danq: create it. then it'll find the tables within
<9> I'd have to do a 'create database' on each SQL node?
<3> right now. yes.
<9> d'oh, ok.
<9> Sorry.
<3> no worries. it's a bit quirky.
<9> I had it working perfectly about two weeks ago, then just stopped.
<9> I'm trying to re-learn, since I didn't take many notes on the subject heh.
<3> guh
<3> define "it"
<9> a nice little cluster.
<9> 4x db nodes and 2x sql nodes and 1 manager, worked well. But I forget many of the details which I'm asing about in here tonight.
<9> Ah ha, success.
<0> screwed everything. gonna go cry
<0> cya
<14> Hi folks!
<14> In running ODBCconfig, I want to configure a new data source that's BDE. Where do I get the driver for that? BTW...I'm running on Fedora Core 4, mysql 4.1.
<14> And I'm running the latest myodbc for FC4.
Return to
#mysql
or
Go to some related
logs:
vncserver red hat hosts.allow
#gentoo
#perl
Unknown symbol devfs_mk_dir
email sent two places sendmail
#javascript
no route to host evolution ubuntu
css bg image inherit
DBAlex
ubuntu + rollback kernel
|
|