| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Comments:
<0> TodoInTX, are there no differences between databases from a 5.0 server and a 4.1 server?
<1> ok well I've made another change ... the warnings have gone from 9 down to 3
<1> but still getting warning 1329 "No data to FETCH"
<2> tdn: are you copying the files directly or doing a mysqldump from 5.0?
<3> Psykick: Ah, that's something different then...
<3> Sounds like you're iterating a cursor that is empty (for some reason)
<2> tdn: I only have a MySQL 5 and 3.23 box handy to test with.
<0> TodoInTX, well... I copied the database files directly from the backup to a Windows MySQL 5.0 server. Then I am doing mysqldumps and pipe them into mysql on my production server (which is running 4.1).
<1> not using cursors ... at least I don't think I am
<0> TodoInTX, but is mysqldump not needed?
<2> tdn: you may not have a consistent backup then.
<0> TodoInTX, hmm...
<0> TodoInTX, it should be...
<2> mysqldump is perfered method from 5.0 -> 4,1
<1> in my stored procedure I'm using basic select statements that ***ign the selected value to declared integers
<0> TodoInTX, ok.
<0> TodoInTX, that is what I am doing.
<2> mysqldump --compatible=mysql40
<0> TodoInTX, there are 46 databases in the backup. Each and everyone of the other bases are working fine.
<1> I've put the stored procedure on pastebin
<1> http://hashmysql.org/paste/1454
<4> aahm... shouldn't last_insert_id() return only one row?
<5> yes, but only if you use it correct
<0> TodoInTX, do you think that my backup of that database is broken?
<5> don't select it from a table
<4> HarrisonF: so, how do i use it correctly?
<5> SELECT last_insert_Id();
<0> TodoInTX, because it gives me that error with a table not existing when using LOCK TABLES?
<4> HarrisonF: ah, thanks
<1> I think I've figured out what was wrong with the query
<2> Or like I said before... the number of tables you are trying to lock all at once is higher than the number of allowed open files.
<2> tdn: you will probably find that if you dump each database independanty you won't get this error.
<0> TodoInTX, I *am* dumping each database one at a time.
<0> TodoInTX, when I got to this database it failed.
<6> WHO WA HA HA HA HA
<7> i have a list a postcodes with x/y coordinates in meters from a fixed point. I'd like to get all postcodes within z meters of an inputed postcode. can i do something with mysql like: select * from postcodes where sqrt((x1-x2)^2+(y1-y2)^2) < z
<7> basically leave the distance calculation for mysql to do on each of the postcodes.. there about 2800 postcodes
<5> sure, or you could use the GIS stuff
<5> which can handle all of the math for you
<8> there should be a strict sql mode that prevents non grouped, non aggregates to be contained in the select list when aggregating
<5> lsmith: there is
<8> ah goody
<8> then they should turn it on by default :-)
<5> ONLY_FULL_GROUP_BY or something
<5> yea, lets break 90% of the applications out there by default ;)
<8> HarrisonF: great .. more work for someone to get paid for :-)
<9> I am getting the error "Access denied for user: 'user@localhost' (Using p***word: YES)" How can I give this user access?
<10> So.. I've just added an entry to a table, how do I now find out the entry's PK value?
<11> are there any upgrade guides available for mysql? I have version 4 installed via an rpm from my linux distro and I will be installing 5 from an rpm should i be concerned about anything?
<12> crackintosh, y, backup your tables first :)
<13> crackintosh: you should be cocnerned about a lot of things! :)
<13> crackintosh: are you using 4.0 or 4.1 right now?
<11> i am using 4.1
<11> i am going to back up the databases before the upgrade
<11> im going to shutdown mysql
<11> then I am going to install the rpm.
<11> and hope for the best.
<10> I'm using PHP with SQL. How can I get a value from an item I just created?
<14> farrioth: that is a #php question
<14> and its RIGHT in the mysql section on the php docs
<13> crackintosh: you should read the upgrade guide, perhaps?
<13> !m crackintosh upgrade 5.0
<15> crackintosh: Nothing found
<13> !man upgrading 4.1
<15> (Upgrading from Version 4.0 to 4.1) : http://dev.mysql.com/doc/mysql/en/Upgrading-from-4.0.html
<13> !man upgrading 5.0
<15> (Upgrading Replication to 5.0) : http://dev.mysql.com/doc/mysql/en/Replication_upgrade_5.0.html
<13> lol
<13> stupid *** bot
<13> crackintosh: http://dev.mysql.com/doc/mysql/en/Upgrading-from-4.1.html
<11> thank you
<11> weird, I was already reading that document!
<16> Hey, I have a query that appears to run and fail on the master, yet makes it to the slave and kills replication. Any ideas?: http://pastebin.com/556889
<16> btw, I didn't design the db... I know it's horrible (and the maintainer is on vacation for two weeks ;( )
<17> Hello. I am running mysql 5 EL4 RPM on centos 4. My mysqld_safe process is not creating multiple mysqld children. this seems to be hindering system performance
<17> is there a switch for this?
<18> mysqld_safe does not create children
<16> rickhax: perhaps a poor default setting? "show variables like '%max%'"
<11> litheum: im just wondering about upgradeing the rpm file
<17> ok, let me run that
<13> crackintosh: what do you mean?
<17> http://pastebin.com/556897
<17> kibab, can you check it out real quick?
<16> rickhax: seems reasonable... 'mysqladmin processlist'?
<19> )
<16> (or 'show processlist')
<16> Nobody has any ideas on the replication dieing? It shouldn't be able to happen as nobody can connect to the slave, the dump happened during a 'flush tables with read lock;' .... it works fine for about 30 minutes or so, then the same query (or one very similar) runs again.
<16> mysql-5.0.18 on both boxes...
<17> http://pastebin.com/556905
<17> there it is kibab , thanks
<17> my connections are sleeping but its normal
<17> i need my db to handle tons of connections, froma remote server, new ones take like 3-5 sec
<16> You have allowed plenty of connections.... you aren't even close to your max... why do you think it's the db?
<17> for example, when i telnet to port 3306 from my server on the same lan, it takes a while to get a prompt from mysql
<17> but when i telnet to port 80, apache greets me instantly
<17> so i know its in the application. also my php script hangs on mysql_connect()
<17> i strace'd it, and its hanging on the connect() sys call
<16> rickhax: so it doesn't matter what db you connect to?
<17> let me verify
<17> no, it doesnt. its the connection to the mysql server that is lagging
<16> rickhax: if telnet is slow.... then it shouldn't matter...
<17> when I look at PS, i notice there is only one mysqld running under mysqld_safe
<16> strange firewall policy... dropping packets? What happens if you log packets to that port? ethereal/tcpdump? etc.?
<17> when I look at my other servers running mysql 4.1, there are tons of mysqld children running
<17> no firewalls involved
<17> software or hardware
<17> so, i think mysqld can only handle 30-40 connections per process. is this possible?
<17> i think there should be many mysqld children spawned
<16> rickhax: ps auxmm | grep mysql
<17> http://pastebin.com/556918
<16> rickhax: so it's only using threads right now... not extra processes
<17> ah ok. i didnt know that would check for threads
<16> rickhax: hmm... how "busy" is that box?
<17> not busy at all, no other services
<17> only this choking mysql. its a fast box
<17> load average: 0.04, 0.02, 0.00 / -/+ buffers/cache: 161 849
<16> rickhax: hmm... you could try attaching strace to the mysqld process (and have it view children) right before you connect... not sure what else to say, seems really strange?
<17> ok, ill try that. thanks for your advice and help dude
<17> is there a way to switch mysql to multi process rather than multi threaded
<16> rickhax: I don't know that you can control it...
<16> rickhax: almost all of your threads were sleeping anyway.... so it shouldn't need to spawn it.
<16> rickhax: almost sounds like something os/kernel level is going on with it.
<16> hmm.... TCP_NODELAY? not sure.
<16> Well, I have to get home to pick up kids... good luck. I'll be back tomorrow... but hopefully you'll have it answered before then.
<20> if i get errno 2 error and it says a directory doesn't exist which does, does this mean i need to clean caches?
<16> rickhax: I don't suppose there's anything obvious in the error log?
<16> tom7e: sounds like the directory really doesn't exist to me... unless you have filesystem corruption or you have a stale link to it.
<20> ok i must have filesystem corruption or stale link
<20> thanks
<16> tom7e: run fsck on the system and see if that fixes it.
<16> ok... I'm really leaving now ;)
<20> shared hosting, can't run fsck
<20> i will check my links
<6> You go do that.
<6> Then come back with a report.
<6> Join quit Join quit Join quit Join quit Join quit Join quit Join quit Join quit Join quit
<21> litheum: you adorable child, thanks for the valentine's day card
<13> hahaha
<13> yes!
<21> litheum: but what's up with the batman theme?
<13> that's what they had at the drugstore
<21> nice
<13> yep
<21> no thought given to it
<13> they also had polly pocket or something
<13> batman was the best combination of ridiculous and not-completely-out-of-control
<21> so am i the only lucky person to receive such a ****ing ridiculous gift?
<13> dean and bryan were also so blessed
Return to
#mysql
or
Go to some related
logs:
d-link di-704 slow connection
#debian
#debian
evolution junk dapper
#perl
perl-Term-ReadPassword rpm
how to burn .img in ubuntu
mysql alter table add auto_increment
gentoo samba-client ebuild
#lisp
|
|