| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Comments:
<0> it's a good bot, s***y at times
<0> rename a field or the whole table?
<1> a field
<2> funky: it's something like ALTER TABLE tablename CHANGE columnName newColumnName [column_definition]
<1> the name of the column, I dont know the correct term
<0> alter table change column oldname newname type;
<1> ok, thanks iotashan and bobfield
<2> now if someone could just solve my problems, I'd be set :)
<3> RAWR!!!
<2> seems that everyone has this problem, and nobody knows how to fix it >:(
<0> iotashan, it runs from the shell at least?
<3> It's easy to fix.
<2> yes, it runs from the shell
<2> just
<2> just /usr/sbin/mysqld starts it no problem
<2> test
<0> p***
<2> but /etc/init.d/mysql start fails, but without errors
<4> RHEL put the init script in place?
<4> or oyu did?
<2> dunno, to be honest
<2> the RHL box came with 3.something
<4> you dont remember if you put it there or the rpm did?
<0> put a fresh copy of mysql.server there
<2> I updated it to 5.0.18
<2> ok... I see three possible files that you might be refering to
<2> /usr/local/mysql-5.0.18/support-files/mysql.server.sh
<2> /usr/local/mysql-5.0.18/man/mysql.server.1
<2> /usr/share/mysql/mysql.server
<3> The last one.
<0> the one in support-files
<2> lol
<0> don't mind me
<3> Having that in /usr/local indicates you didn't upgrade with the RPMs.
<3> Why?
<2> I have no answer for that
<2> I'm a linux newb
<2> I did "rpm -Uvh MySQL*"
<2> which, from what I understood, was supposed to do an upgrade
<2> OH
<2> I *also* downloaded & unzipped the source to /usr/local
<3> Ahhh
<2> but I haven't compiled or anything
<3> Explains why you had the .sh extension
<2> (sorry, beware the n00b) :)
<2> so what exactly do I copy to where?
<3> Anyway, /usr/share/mysql/mysql.server is the init script you should use, but the RPM should also have already set that up for you
<3> Try the /etc/init.d/mysql start route, and see if your mysqld error log has anything about a failure
<3> Very odd for the init script to fail but mysqld itself to succeed
<2> there was no failure
<2> <2> looking at the /var/lib/mysql/server.err file, I see:
<2> <2> 060309 22:33:13 mysqld started
<2> <2> 060309 22:33:13 mysqld ended
<3> ANy other .err files?
<2> no
<3> RHEL3 or RHEL4?
<2> 4
<3> SELinux is killing it probablyu
<3> setenforce 0
<3> Then try it
<3> See if that changes anything for you
<3> ;)
<3> Don't mess with Bob
<2> that worked
<3> Your SELinux policies are forbidding mysqld from accessing or creating some file, probably the socket file
<2> so is it perminantly fixed, or is the the begining of a long complicated & undocumented fix? :)
<3> It is not permanently fixed until you correct your SELinux policies for mysqld
<2> so just google selinux ?
<3> I'd tell you how to do that but this is where I plug mysql.com/network and shut up
<2> hehe
<2> ah, buying a support contract isn't an option
<2> I wish it was :(
<4> heh
<4> ridiculous
<3> ;)
<2> thanks for pointing me in the right direction, Therion
<3> sure
<5> hi
<6> Hi there
<2> so, theoretically, if I put setenforce 0 in my rc.local, it would just turn off selinux all together?
<6> I cannot seem to find out how to set current_timestamp() or at least the variable value of it as a default for a field... anyone could give me a tip?
<6> alter table something alter column datefield set default current_timestamp(); didn't work
<2> last I checked (which was mysql 3.x) you can't set a default to a function
<0> Shariff, you can't, not directly
<0> now either with a trigger or just call curdate()/curtime() during insert
<6> ah ok thanks!
<0> !man date time functions
<7> (Date and Time Functions) : http://dev.mysql.com/doc/mysql/en/Date_and_time_functions.html
<2> insert triggers are a mess on MSSQL.... are they generally discouraged on mySQL too?
<0> worked okay for me
<0> I encourage them
<2> and they work ok in a high-traffic setup?
<2> LOL
<0> I've used them for million-plus row insertions
<3> iotashan: They aren't so much discouraged as in 5.0 they have some flaws that make them less than as useful as they will be in 5.1.
<3> They function however
<2> very cool to know
<2> I've seen insert trigs bring MSSQL 2k to it's knees
<2> even on a cluster
<8> the reason you cluster MSSQL is so that you have more knees to bring it to when you submit a query
<2> of course, they were poorly written and useless, but that's besides the point :)
<0> only a slight performance degradation for a simple trigger
<2> coolio
<2> I'm very excited about mySQL 5
<0> of course depending on what you do, can slow things down
<4> me too
<4> mysql is growing up ;-)
<3> I had no particular issues with MSSQL other than known limitations (ie: testing for nulls ****ed, in 2000/earlier at least)
<8> MySQL 5 is out already
<8> you should be excited about MySQL 5.1 now
<3> Triggers degrade performance in any DB, to some extent
<3> Darien: 5.2 is more interesting to me than 5.1, but 5.1 has nice stuffssss
<8> being excited about MySQL 5 is so 2005
<0> have to expect that
<2> now if Oracle would stop trying to buy the legs out from under mySQL >:(
<4> haha
<8> Therion: once 5.1 is out then I'll be excited about 5.2
<4> who's trying to win the uber-geek award here?
<4> ;-)
<3> Darien: I'm just interested in some of the things 5.2 will fix that have needed fixing for a very long time now
<3> Darien: The rest is gravy ;)
<0> so why are so many people running 4.0 and 3-something?
<8> because RHEL ****s
<2> and some people are too lazy to use new drivers, and can't figure out old_p***word()
<2> :D
<3> bobfield: MySQL 4.0 and 3-something?
<2> like macromedia.... wish they'd just include the newer drivers
<0> yes that's what I was referring to
<3> 3-something usually because OEM or provider refuses to upgrade
<8> RHEL/FC2 shipped 3.23
<3> 4.0 because it's rock solid and they don't need the newer features
<8> because they're incompetant retards
<8> incompetent
<3> Darien: GPL client library played its part in that too one could ***ume
<2> lol
<8> and there's no *supported* way to upgrade
<8> Therion: Debian had that problem solved before FC2 shipped, AFAIK
<8> anyway, a lot of hosting services STILL sell hosting on FC2
<3> Darien: Well, Debian having a problem solved before someone else is pretty frickin amazing, I have to admit
<8> haha
<3> My wife won't stop yelling "Bitches!"
<3> wtf
<8> Debian solves problems extremely quickly, unless it's around release time, in which case they get put off
<8> Therion: PKers?
<3> They solve security problems quickly
Return to
#mysql
or
Go to some related
logs:
udev intel_agp
stopping apache instances
ubuntu modprobe.d loop
matplotlib boa constructor crash
No module named win32clipboard
2wire608
#gaim
counterstrike varnames
#linuxhelp
#ubuntu
|
|