| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Comments:
<0> hm, no luck. i will just have to do something tricky :P
<0> or something brutally manual :\
<1> the --compatible flag should work fine
<0> i did --compatible=mysql40
<0> but still got You have an error in your SQL syntax.
<0> it's going from 4.1.12 to 4.0.25
<1> right
<1> should work, though
<1> what is the error?
<0> ...near 'DEFAULT CHARSET=latin1 COMMENT='Contains data about active user
<0> same spot, but on a new line number with the new file. so i know it's a bit different (or atleast i can ***ume i do indeed have the new file and the --compatible flag did something)
<1> weird, i can't find a bug about that
<1> is "DEFAULT CHARSET=latin1" even compatible with 4.0?
<1> can't remember
<1> try creating a dummy table with that option in it
<1> if it isn't compatible, you could write a script to remove all ocurrences of that string
<2> Hi people
<2> :-)
<2> I'm trying to move all my databases over from one machine to another, using 'mysqldump -uroot -p --all-databases > msql.sql' then from the new machine typing 'mysql -uroot -p < msql.sql' but I get an error that reads, "ERROR 1064 at line 19: You have an error in your SQL syntax near '-_cms' at line 1". Any hints?
<1> Wyleyrabbit, yeah, use --quote-names on that mysqldump command
<2> jpm, ok, I'll give that a go.
<3> I have a timestamp column in mysql how do i write the current time into it?
<1> Bruce_LeeRoy, UNIX_TIMESTAMP()
<3> ok thanks
<2> jpm_, I get the same error as before.
<1> what's on line #1 then?
<4> 0
<0> jpm_: very weird. now it worked and i used the exact same syntax for importing and it worked.
<2> jpm_, line 1 or line 19?
<0> jpm_: meaning i probably screwed something up before :P
<1> djmccormick, go figure!
<1> Wyleyrabbit, line #1 is where the error is being triggered, no?
<2> jpm_, not sure. It says error at line 19, then error in sql syntax at line 1
<1> right, so show me all 19 lines
<1> there's a paste url on the topic
<2> jpm_, here you go:
<2> http://hashmysql.org/paste/1630
<1> Wyleyrabbit, and you are importing into which version of mysql?
<2> mysql -V
<2> mysql Ver 11.18 Distrib 3.23.58, for redhat-linux-gnu (i386)
<1> damn that's old
<1> erm
<2> jpm_, no kidding
<1> well, it looks like the problem is with that dash on the database name
<2> this is a temporary box my ISP set up for me
<1> and --quote-names is not helping it
<1> how about you opening that dump file manually
<1> and adding quotes around that db name?
<2> with vi?
<3> my timestamp is inserting as 00000000
<3> using UNIX_TIMESTAMP()
<1> Wyleyrabbit, sure, whichever editor you want
<1> Bruce_LeeRoy, crazy!
<1> Bruce_LeeRoy, show me your insert statement
<2> single or double quotes?
<1> Wyleyrabbit, single, i guess
<3> jpm_, whats up with it :(
<1> shouldn't matter
<3> jpm_, my column is timestamp(8) how do i make this longer?
<1> remove the (8) from there
<3> alter table?
<1> right
<1> alter table blah modify column boohoo boohoo timestamp;
<5> hi
<5> i don't understand a where like in 400 000 lines is long ?
<5> SELECT *
<5> FROM `kw`
<5> WHERE `kw` LIKE '%sheik%'
<5> AND `deep` >= '2'
<5> ORDER BY `bid` DESC
<5> LIMIT 0 , 5
<1> i don't understand your question
<5> Showing rows 0 - 2 (3 total, Query took 1.1251 sec)
<5> d select_type table type possible_keys key key_len ref rows Extra
<5> 1 SIMPLE kw range deep deep NULL 080 Using where; Using filesort
<6> LIKE '%sheik%' cannot use an index it will be slow
<5> why this query is so long
<5> kw is an UNIQ field
<1> you are doing a like
<1> what did you expect?
<1> if you do LIKE 'sheik%', then maybe it will use the index
<1> since it can just check the first 5 characters
<1> otherwise, it will need to check all values to see which one match
<1> there's no magic here
<5> it's normal ?
<5> how can i search ?
<5> using a match again with FULL text type ?
<5> against
<1> you can use myisam and then use a fulltext index on that column
<5> but LIKE is for what then ?
<5> jpm_ fulltext is only for search against query right ?
<5> MATCH AGAINST i meant
<5> or i can do also fast LIKE ?
<1> of course
<5> it's still slow
<5> SELECT *
<5> FROM `kw`
<5> WHERE `kw` LIKE '%pill%'
<5> AND `deep` >= '2'
<5> ORDER BY `bid` DESC
<5> LIMIT 0 , 15
<5> Showing rows 0 - 14 (15 total, Query took 1.2227 sec)
<1> did you think it would be faster?
<5> kw FULLTEXT 22854
<1> you need to change your query as well
<5> jpm_ sure 1.2 sec is extermly slow]
<5> a MATCH AGAINST?
<1> http://dev.mysql.com/doc/refman/5.0/en/fulltext-search.html
<1> right
<1> WHERE SEARCH(kw) AGAINST ('pill')
<1> err
<1> WHERE MATCH(kw) AGAINST ('pill')
<5> yeah
<7> hey, i am having this weird problem in which a select statement is returning inconsistent results. could this be a bug in mysql? or am I just going crazy? http://pastebin.com/623992
<7> the select on line 17 returns a different ip than expected
<8> bug ;>
<8> maybe
<8> :D
<7> yeah, check this out: http://pastebin.com/624008
<8> any difference when you put it in ip='1073460'
<8> ?
<8> i mean ' '
<9> wow, I'd have to see that with my own database to believe it. what version of mysq?
<7> mysqladmin Ver 8.41 Distrib 4.1.12, for redhat-linux-gnu on x86_64
<10> i have a question about the user() function. i've been using it to extract the IP address of the client machine and this has worked for hundreds of clients.
<11> interesting..
<8> thats admin polvi
<10> Now, for a couple of clients, the IP part of what user() gives is a hostname instead
<7> dd: huh?
<9> is it repeatable if you dump that table and import it to another db under another name?
<10> the clients are win32 programs using libmysql
<7> vandemar: this table has 48million rows in it... that'll take awhile .. :(
<10> wondering if there is a more reliable way to extract the ip address of the client
<10> other than user()
<11> epictetus, why don't you get your program to retrieve the IP?
<11> Then ammend the changes to the query prior to executing it.
<10> jbrimble: mainly for client machines that have multiple interfaces
<10> i want the ip of the interface they connect to the mysql server over
<11> oh i see.,
<10> I guess I could figure that out if I look at all interfaces and netmasks and gateways, and do my own routing calculation to figure out which interface will be used
<12> Is there a way to get MySQL to output a 5 rows for one entry... i.e. I'm trying to show that there are 5 timeslots available because the event.slots=5 and one person signed up, so I want to output the first row with the person's name, and then 4 blank rows
<13> NOT OK: Runtime error 'Can't insert new RightMedia::Log::Message: DBD::mysql::st execute failed: MySQL server has gone away -- any ideas?
<10> however that would be obnoxious to make work (it needs to work on 2000, xp, 2003, all flavors)
<11> what are you writing the app in?
<10> C
<10> using a gcc variant (bloodshed compiler)
<10> maybe if i poke at the mysql source i can see exactly how the user() function works
<10> and figure out why it is going wrong
<10> or I could just not worry about it.. the bug has only happened on 2 client machines out of hundreds
<10> i can just blame those machines
<10> hehe
Return to
#mysql
or
Go to some related
logs:
unmerge orphans
gentoo libpangoxft firefox
#linux
debian sarge raidtools
BBL PCMCIA.drivers
md2_resync
linux terminal under windows
#perl
+intel 945G +xorg-6 +driver
Suse: rwin
|
|