Re: deleting from an InnoDB table with a composite PK 500 chars crashes mysqld

Hi! I was able to repeat this in our computer. mysqld crashes inside the MySQL code. It did not crash if the primary key length was 500 bytes. This looks like a memory overrun bug. I have forwarded this bug report for Monty to check. Regards, Heikki Tuuri Innobase Oy --- Order commercial

Re: deleting from an InnoDB table with a composite PK 500 chars crashes mysqld

Hi! Ok, I found that in unireg.h MAX_KEY_LENGTH is defined as 500, and that is used, for example, in the opt_range.cc PARAM struct as the key value buffer size. A memory overrun will happen if a key length is 500 bytes. InnoDB allows keys up 3500 bytes in size. Fix: change MAX_KEY_LENGTH in

Re: Function in index

Kittiphum Worachat writes: Is it possible to use function such as year() in index key (like use subpart for string field that it work) Nope, you'll have to have a column with just the information that you wish to index (except in the case of strings, of course). Also, you yourself will have

Problems in Foreign Key Syntax

Hi!! I am trying create the following tables:- CREATE TABLE Scenario ( Scenario CHAR(10) NOT NULL PRIMARY KEY, CopyFromScenario CHAR(10)) TYPE = InnoDB; CREATE TABLE Attribute( Attribute CHAR(25) NOT NULL PRIMARY KEY, Type CHAR(25)) TYPE = InnoDB; CREATE TABLE AttributeValue( Attribute

[BUG] 'ENOTCONN' replication dies every once in a while

I can't easily replicate this as it happens only once every few weeks. We are very happy with MySQL otherwise! It is wonderful as a nameserver database backend. We have a server in the US that replicates from a master in The Netherlands. Connectivity is usually very good. However, sometimes we

How can i autostart MYSQL

hi all, How can i autostart MYSQL while starting of my LINUX server. regards, sreedhar - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive)

Re: Is this possible with MySQL?

You may try this and see if it works (didn't test it, just jotted it down here): select t1.textid, t1.textid, t1.textvalue I just don't understand why you did this? Why select t1.textid twice? from texts as t1, texts as t2 Again I don't understand this... Wouldn't t1 and t2 be the same?

Compiling From Source

Hi; I've written this config script: #!/bin/sh cd /usr/local/mysql ./configure \ --with-libwrap=/usr/share/doc/tcp_wrappers \ --with-raid \ --with-thread-safe-client exit 0 Running the script gives me this error: /usr/local/bin/config_mysql: --with-raid: command not found *however* running

Re: Problems in Foreign Key Syntax

Amit, thank you for a bug report. There seems to be a bug in the MySQL parser, because MySQL internally makes the first column of the table AttributeValue as of type VARCHAR(25), though you have declared it as CHAR(25). Since the internal type of the referenced column is CHAR(25) but the

Re: Problems in Foreign Key Syntax

Hi! Ok, I looked in the manual and found the following: http://www.mysql.com/doc/S/i/Silent_column_changes.html If any column in a table has a variable length, the entire row is variable-length as a result. Therefore, if a table contains any variable-length columns (VARCHAR, TEXT, or BLOB),

RE: Is this possible with MySQL?

* Stig Nørgaard Jepsen Is it possible to make a query in MySQL, which could return row 1,3 and 4 based on these conditions? select distinct t1.textkey, if(t2.textid,t2.languageid,t3.languageid) as languageid, if(t2.textid,t2.textid,t3.textid) as textid,

select, replace priority

I have a php file with 1. CREATE TEMPORARY TABLE on tableA; 2. REPLACE into tableA; 3. CREATE TEMPORARY TABLE on tableA; 4. REPLACE into tableA; Can i be sure that the REPLACE in step 2. will be finished before the TEMPORARY table in step 3. is made?

linux and hdparm

I've just checked my IDE parameters for one of my linux mysql servers. Using hdparm -c -d I've discovered that I'm running at the default 16 bit non DMA. I'm assuming that enabling both of these should increase performance, possibly significantly. However, I'm concerned about doing this 1) on

Re: Is this possible with MySQL?

Stig Nørgaard Jepsen wrote: You may try this and see if it works (didn't test it, just jotted it down here): select t1.textid, t1.textid, t1.textvalue I just don't understand why you did this? Why select t1.textid twice? Ugh... That was a mistake -- I actually meant t1.textid,

RE: Is this possible with MySQL?

* Stig Nørgaard Jepsen It's a miracle! Thank you so much. I didn't think it was possible. It seems to work alright. :) The LEFT JOIN and IF() combination is quite powerfull. But i really don't understand what happens. I would be very glad if you could explain for me what's the principle

Re: linux and hdparm

On Saturday 22 Dec 2001 19:42, Gary Huntress wrote: I've just checked my IDE parameters for one of my linux mysql servers. Using hdparm -c -d I've discovered that I'm running at the default 16 bit non DMA. I'm assuming that enabling both of these should increase performance, possibly

Re: deleting from an InnoDB table with a composite PK 500 chars crashes mysqld

hi! Heikki == Heikki Tuuri [EMAIL PROTECTED] writes: Heikki Hi! Heikki Ok, I found that in unireg.h MAX_KEY_LENGTH is defined as 500, and that is Heikki used, for example, in the opt_range.cc PARAM struct as the key value buffer Heikki size. A memory overrun will happen if a key length is

Re: Coredump when running scripts/mysql_install_db

Hi! Michael == Michael Stassen [EMAIL PROTECTED] writes: Michael After applying your patch to my copy of 3.23.46, I get Michael mysqld.cc: In function `void kill_mysql()': Michael mysqld.cc:622: `kill_server_thread' undeclared (first use this function) Michael mysqld.cc:622: (Each

Time Weirdness

I've got a table that's acting as a timesheet, with time began and finished in each record. In order to do up an invoice based on that, I was making a total time field, which was just finish - start. It worked great for all of the records but two, where I had this result: 1 -- Start:

Disk array performance

Does anyone have any experience splitting up index and data files across different RAID arrays? I'm looking at fast disks in RAID0 for the indexes.. and then some slighly slower (rotational speed) disks with some redundancy (RAID?5) for .MYD/.frm files. Has anyone had any experience or