#mysql
20 October 2007
Total 24 pages. You are browsing page 2/24.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last
00:32 <****> fuse: You won't know until you ask the question. Each minute you wait could be one less minute someone here has to look.
00:32 <****> :-)
00:33 <****> Fine, fine....The prepare doesnt generate an error, and the bind seems to work, but the execute causes a segfault
00:33 <****> http://rafb.net/p/5af9Pf94.html
00:33 <****> there is a portion of the C code where the bind,execute happens
00:34 <****> The stmt is initialized outside of that function.
00:34 <****> omfg
00:34 <****> i found my error i think
00:34 <****> wait, maybe no
00:35 <****> yeah, i have no idea what the isssue is
00:41 <****> fuse, your binds use a handle stmt, but execute uses mime_stmt
00:41 <****> i fixed that
00:41 <****> still dies
00:42 <****> _insert (event_id=81, k=0x87cf181 "Date", klen=4, v=0x87cf187 "Fri, 19 Oct 2007 22:45:15 GMT", vlen=29)
00:42 <****> that is the last frame in gdb that is code that i wrote.
00:42 <****> cli_stmt_execute (), and thi sis the very last frame
00:42 <****> themo|ester, stop that ffs
00:43 <****> I just am just wondering if my binds are incorrect at some point
00:43 <****> http://rafb.net/p/p4bZmh25.html
00:43 <****> that is the actual code that is being tested
00:44 <****> gnari: i pasted that code a bit ago, i've fixed a couiple things since then
00:44 <****> hey all.... anyone knows a good application to convert a .sql file to diagram?
00:44 <****> Is IN or NOT IN faster?
00:44 <****> field types are BIGINT UNSIGNED,VARCHAR, VARCHAR
00:46 <****> fuse, probably not the issue, but do you set is_unsigned anywhere?
00:46 <****> BIGUNS
00:47 <****> beebeee, depends
00:48 <****> gnari: Well, let's assume that we do the obvious reversal of the sub-SELECT.
00:50 <****> beebeee, if the list is extreemely long, and does not contain 43, but the first element is 42, then a "WHERE 42 IN (...)" will be faster than a "WHERE 43 NOT IN (...)"
00:52 <****> Hrm.
00:52 <****> Quick version question. I have some MySQL books from 2004, when 4.0 was new. I've I'm learning the basics of MySQL on a 5.0.45 system, will there be any issues, other than I won't know about the new stuff?
00:52 <****> IN really is an advanced language construct.
00:53 <****> leitz you should be fine
00:53 <****> just remember that the timestamp format changed between 4.0. and 4.1
00:54 <****> beebeee, not really. X IN(e1,e2,...en) is a macro for ((X = e1) OR (X = e2)...OR (X = en) )
00:55 <****> gnari: But... it accepts a sub-SELECT...
00:55 <****> beebeee, yes
00:56 <****> Thanks!
00:57 <****> beebeee, just remember there can be unexpected effects if the subselect in a NOT IN() returns NULLs
00:58 <****> gnari: Yeah... that's one thing that bothers me.
00:58 <****> (My)SQL is so damn powerful. It's almost like the app has no work to do. Heh.
00:59 <****> gnari: just on the event_id
01:01 <****> fuse, i do not see it set. i am talking about your "my_bool is_unsigned", not bind[0].is_unsigned
01:04 <****> fuse, i do not know the API, but are you sure you should set it to &is_unsigned ?
01:05 <****> Is it possible to get some user statistics out of mysql? e.g. how many queries each user performed, etc?
01:06 <****> Fiskah, no
01:07 <****> So there is no other way than spying the process list? :)
01:07 <****> fuse, i read the docs as you should do bind[0].is_unsigned=1;
01:07 <****> spying on the process list won't even give you that
01:08 <****> Am I right in thinking that non-admins can effectively only insert and edit but not delete grant rules?
01:08 <****> no
01:09 <****> I'm getting an error and I'm not sure what is causing the problem. If anyone could help, much apprecated. http://pastebin.com/d29087f04
01:09 <****> ebergen: so how can I remove them?
01:09 <****> you have to be a super user
01:09 <****> ...that's what I meant with admins
01:10 <****> I have this error please help me http://pastebin.mysql-es.org/179
01:10 <****> flexo, the relevant part only
01:10 <****> non admins can't modify the grant rules at all
01:10 <****> http://pastebin.com/d44668a0d better?
01:11 <****> ebergen: I have some random revokes on my tables that I apparently set myself. The best I can do is set all privileges?
01:11 <****> ebergen: of course you can
01:11 <****> what?
01:11 <****> revokes you set?
01:11 <****> I'm confused
01:12 <****> flexo, why the parens?
01:12 <****> gnari: do you have a link for the unsigned stuff?
01:12 <****> ebergen: not all of course, but some of the things that SHOW GRANTS; shows are things you may have set as a general user with access to the relevant database
01:12 <****> oh, I dunno, someone told me to try that to fix it...I didn't have them before
01:12 <****> ebergen: ...right?
01:12 <****> I have this error please help me http://pastebin.mysql-es.org/179
01:12 <****> ebergen: mostly table-level permissions in my case
01:12 <****> scarfboy are you talking about general users being able to modify their own tables or the global grant tables
01:13 <****> I have this error please help me http://pastebin.mysql-es.org/179
01:13 <****> gnari: btw, it still crashes with unsigned=1
01:13 <****> fuse, i just read http://dev.mysql.com/doc/refman/5.0/en/c-api-prepared-statement-datatypes.html
01:13 <****> ebergen: well, I assume the permissions are actually *stored* in mysql.{user,host,db,tables_priv} and so
01:13 <****> that's actually where they are stored
01:14 <****> ebergen: The user I log in as can't access those, but can still grant/revoke to effectively change some rules there
01:14 <****> no, you can't
01:14 <****> if you can you really screwed something up
01:15 <****> ebergen: you're telling me a user with full permissions for a specific database cannot change table/column level permissions at all?
01:15 <****> greetings all .. I'm reading the InnoDB docs at http://dev.mysql.com/doc/refman/5.0/en/innodb-configuration.html where the 2GB limit is mentioned for 32-bit GNU/Linux x86 systems. I'm running FreeBSD 5.4 on dual Xeon servers .. is this limit relevant for this setup as well?
01:16 <****> 2 GB limit?
01:16 <****> It's 4 GB!
01:16 <****> hey .. I'm just reading the docs, lol
01:16 <****> it's a 4GB process limit on 32bit systems, right?
01:16 <****> ebergen: it's our IT department with the love for anal, fine-grained permissions that set this up, maybe they set it up qider than I though. (Me, I say network connection is enough for app security:) )
01:16 <****> gnari: yeah, just read through it, you are definitely correct, but it is crashing in the same spot.
01:17 <****> fuse, have you tried to change the query so as to avoid the first bind? that way you will see if it connected to the binding of the long long
01:17 <****> virgomoon: 4GB process limit is rare because of the kernel/user split, but yes.
01:17 <****> scarfboy a user without the grant option won't be able to change the permissions for anything
01:18 <****> ebergen: I have the grant option for my own database
01:18 <****> gnari: hm
01:18 <****> k .. I'll monkey around with it, then
01:18 <****> thx
01:18 <****> ebergen: but obviously no access to mysql.*
01:18 <****> gnari: i am doing another prepared statement that is almost identical, and it works fine with the long long
01:19 <****> ebergen: so does this amount to only being able to add+edit and not delete grant rules?
01:19 <****> ok I see what you're saying now
01:19 <****> ebergen: yeah, I can see the misunderstanding, sorry
Total 24 pages. You are browsing page 2/24.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last