EINTR in my_connect()

Hi all, Is there any reason to abort connection when error code is EINTR? I tried to patch it as below: --- [EMAIL PROTECTED]:~/src/mysql-5.0.22/libmysqld]$ diff -c client.c client.c.patch *** client.c2006-07-10 12:24:52.0 +0900 --- client.c.patch 2006-07-10

Re: split a datafile

wangxu wrote: No,i don't want to do this. My previous variable is innodb_data_file_path=ibdata1:2G;ibdata2:10M:autoextend. But ibdata2 had exceed 3G! I changed the variable to innodb_data_file_path=ibdata1:2G;ibdata2:2G;ibdata3:10M:autoextend. But Server couldn't start. How can

Re: Setting default character_set_results

Hi Gleb, Please check, if putting of these lines into my.cnf solves the problem: [dbdmysql] default-character-set = ujis Yupe! It works. Thank you very much :) -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:

Setting default character_set_results

Hi, I have a problem with character_set_results on MySQL 4.1. I have this in my my.cnf: [mysqld] default-character-set=ujis default-collation=ujis_japanese_ci [client] default-character-set=ujis [mysql] default-character-set=ujis If I connect from mysql client, and do 'show variables' I get

Re: Character set on 4.1 and ujis support

Hi, Thank you very much for the reply. $dbh-do(SET character_set_results=ujis'); This works! Is there any way I can set this value on MySQL config file, so I don't need to change all my scripts? Right now my my.cnf looks like: [mysqld] default-character-set=ujis

Character set on 4.1 and ujis support

Hi, I just moved my DB from 4.0 to 4.1.11. I used mysqldump to dump the data first, and then inserted it to the new DB. The character set of the data is EUC-JP (ujis). My problem is, I can see the character correctly if I connect to mysql server using mysql client. For example: # mysql -e

MySQL 4.1.10 Crash

Hi, I use MySQL 4.1.10, and it crashes about once a day. I posted about this bug two days ago, but I forgot to put the stack trace result there, sorry for that. Now I put the stack trace result below. My kernel is 2.6.10, RAM is 4 GB. Single CPU with HT (shows up as 2 CPU on linux). The MySQL

mysqld got signal 11

Hi, I have a problem with some of my servers running 4.1.10. It is running on linux 2.6.10, Fedora Core 3 default installation. The mysql I use is from RPM package that mysql.com provides. About once a day, the servers will crash and I find mysqld got signal 11 error on the log file. These

Memory limit?

Hi, I try to install MySQL 4.1.9 (official RPM from mysql.com). My machine is running linux 2.6.9, and it has 4GB of RAM. The problem is MySQL won't start if I set innodb_buffer_pool_size to = 2GB. Here is my ulimit. [EMAIL PROTECTED] mysql]# ulimit -a core file size (blocks, -c) 0 data

Re: Memory limit?

Hi Tobias, On Wed, 9 Feb 2005 14:48:16 +0100 (CET) Tobias Asplund [EMAIL PROTECTED] wrote: I try to install MySQL 4.1.9 (official RPM from mysql.com). My machine is running linux 2.6.9, and it has 4GB of RAM. The problem is MySQL won't start if I set innodb_buffer_pool_size to = 2GB. Here

maximum id of processlist

Hi, I noticed that the number of id when I do show processlist is getting bigger and bigger. Now it is about 444,466,168. What is the maximum number, and what will happen if it runs out of number? --bk -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To

Re: Cannot GRANT REPLICATION SLAVE

Hi Paul, Generally this happens when you upgrade to a newer version of MySQL but don't upgrade the grant tables. See bullet 4 on this page: http://dev.mysql.com/doc/mysql/en/Upgrade.html Arghh OK!!! Thank you very much. It is ok after I run mysql_fix_previlege_table. --bk -- MySQL

Cannot GRANT REPLICATION SLAVE

Hi, I have this weird things happens. kaspia:/var/lib/mysql# mysql --version mysql Ver 12.22 Distrib 4.0.21, for pc-linux-gnu (i386) mysql show grants for root@localhost; +-+ | Grants for [EMAIL PROTECTED]

Re: Cannot GRANT REPLICATION SLAVE

I can grant another previleges, but not replication slave and replication clients. How do I fix this? Thank you in advance. I checked my mysql.user table, and I just found out that it didn't have Repl_slave_priv and Repl_client_priv columns. How is this possible? mysql select * from

Re: Delete duplicate entry

Hi Gerald, try ALTER IGNORE TABLE. Thank you very much. I should have checked the manual first. http://dev.mysql.com/doc/mysql/en/ALTER_TABLE.html IGNORE is a MySQL extension to standard SQL. It controls how ALTER TABLE works if there are duplicates on unique keys in the new table. If IGNORE

Re: Comparing Tables

On Thu, 7 Oct 2004 19:22:23 -0700 Jason Williard [EMAIL PROTECTED] wrote: I know that it is possible to do this, though I don't know how. I have 2 tables containing information about trouble tickets. One of the tables (table2) contains information about every ticket ever received,

Delete duplicate entry

Hi, I have a table that looks like: CREATE TABLE `message_inbox` ( `member_id` mediumint(8) unsigned NOT NULL default '0', `message_id` int(10) unsigned NOT NULL default '0', `new` enum('y','n','replied') NOT NULL default 'y', `datetime` datetime default NULL, KEY `idx_1`

Replication doesn't work under heavy load

Hi, I just noticed that replication on my servers failed when they were under heavy load. I have 1 master and 2 slaves. For example, I have table for all incoming messages. message_inbox | CREATE TABLE `message_inbox` ( `member_id` mediumint(8) unsigned NOT NULL default '0', `message_id`

MySQL Process

Hi, I run Debian 3.1 with kernel 2.6.8-1-686-smp, I wonder why why I do ps -aux, I only can see 1 MySQL process. mysql 4227 0.0 64.6 1849456 1341912 pts/1 S Sep16 4:57 /usr/sbin/mysqld --basedir=/usr --datadir=/var/lib/mysql/ --user=mysql --pid-file=/var/lib/mysql//natuna.pid

SQL_SLAVE_SKIP_COUNTER

Hi, If I run these commands on slave: SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1 SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1 SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1 (3 times and then...) START SLAVE How many counter will be skipped? Is it 3 or just 1? Thank you. Regards, Batara -- MySQL General Mailing

Replication blocked

Hi, I have 2 DB server, running as master and slave. I just add an index to one of my table on master, it took about 12 minutes. During adding the index, I have insert/update queries to other tables. On master this has no problem at all. The problem is, on slave these queries were blocked by the

Re: Problem with Mysql 4.0.18 + Debian

Ryan, Here's the problem. I had the site up several days, with everything running perfectly. Ordinarily there would be about 11 mysql processes running. However, after a few days of running smoothly, the number of mysql processes increases to over 170, and the site crashes. Formerly I'd

Re: Upgrading mySQL 3.23 to 4.0

On Mon, 02 Aug 2004 17:31:03 -0400 [EMAIL PROTECTED] wrote: I am now getting [EMAIL PROTECTED] Desktop]# mysql -u root -pmypasshere ERROR 2002: Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2) any thoughts... once again thanks for the assistance It

Kernel 2.4 and 2.6

Hi, I use MySQL 4.0.20 on my replication (slave) server. I noticed that when I use kernel 2.4.26 SMP, the slave runs very slow (it can't catch up with master). My machine has dual CPUs with HT. But if I change to kernel 2.6.6 SMP, everything just runs fine, the slave can catch up easily with

Re: Performance of big table

Hi Harald, Replacing the index on member_id by a compound index on (member_id, id) might help. Thank you very much for your suggestion. After replacing the index with compound index, the query works very fast now. -- MySQL General Mailing List For list archives:

Performance of big table

Hi, I use InnoDB for my table, and I think I have hit the point where I can't do query optimization any more. The query is as simple as below: SELECT SQL_NO_CACHE diary_id, LEFT(body, 28) AS body , id FROM diary_comment WHERE member_id='343' ORDER BY id DESC LIMIT 15; And it took about 5

ERROR 2013: Lost connection to MySQL server during query

Hi, I am running Debian sarge with MySQL 4.0.20. My problem is I can connect from localhost, but when I try to connect from other host this error comes up: ERROR 2013: Lost connection to MySQL server during query I tried from many clients, included MySQL 4.0 and MySQL 3.23, but they all got same

Re: ERROR 2013: Lost connection to MySQL server during query

I am running Debian sarge with MySQL 4.0.20. My problem is I can connect from localhost, but when I try to connect from other host this error comes up: ERROR 2013: Lost connection to MySQL server during query Sorry I found the answer. I have ALL:ALL in hosts.deny -- MySQL General Mailing

MyISAM and InnoDB table design

Hi, I read it somewhere that InnoDB is faster for table with high read/write concurrency. I have a table look like this: CREATE TABLE diary ( id INT UNSIGNED NOT NULL AUTO_INCREMENT, member_id INT UNSIGNED NOT NULL, title VARCHAR(255) NOT NULL, body TEXT NOT NULL, date DATE NOT NULL,

GROUP BY with MAX

Hi, I have a table that looks like: mysql select * from test3; ++++ | sub_id | date | data | ++++ | 1 | 2004-05-01 | data 001 | | 1 | 2004-05-02 | data 002 | | 1 | 2004-05-03 | data 003 | | 2 |

Grouping problem

Hi, Can I ask MySQL to GROUP BY id on MAX(something). For example, the query below: mysql SELECT diary.date, diary.time, diary.id AS diary_id, diary.title, - link.to_id AS friend_id - FROM link - JOIN diary ON diary.member_id = link.to_id - WHERE link.from_id=10 - ORDER BY

Switch replication master

Hi, I use replication with 1 master and 1 slave. Right now the slave is only used to make a backup. Next week I am going to change my master motherboard. So I need to take it offline for a while (about 90 minutes). I am thinking about switching the master for that 90 minutes, so the slave will

Re: Timestamp and alter table

I want to change the column choose to ENUM('y', 'n', 'weekly') without changing the timestamp. How can I do that? I tried ALTER TABLE as usual, and it didn't affect the timestamp. Sorry I didn't try this from the beginning :) --Batara -- MySQL General Mailing List For list archives:

Timestamp and alter table

Hi, I have a table that looks like: mysql desc test; +---+---+--+-+-+---+ | Field | Type | Null | Key | Default | Extra | +---+---+--+-+-+---+ | timestamp | timestamp(14) | YES | | NULL| | |

Strange behavior with IF?

Hi, Can someone tell me why this query doesn't work? SELECT IF(ISNULL(network.level), 4, network.level) AS level, member.photo_level FROM member LEFT JOIN network ON (network.from_id=101 AND network.to_id=member.id) ORDER BY member.last_login DESC LIMIT 0,3 +---+-+ | level |

Re: Strange behavior with IF?

Hi Diana, SELECT IF(ISNULL(network.level), 4, network.level) AS level, member.photo_level FROM member LEFT JOIN network ON (network.from_id=101 AND network.to_id=member.id) ORDER BY member.last_login DESC LIMIT 0,3 +---+-+ | level | photo_level |

Values for inno_db_data_file_path

Hi, My DB server has 2 GB RAM, and 5 x 36 GB HDD in RAID 5. How can I tune inno_db_data_file_path for best performance? Is it better to split up the data, or just use 1 very big autoextend data? Thank you, Batara -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To

Very big IN ( $value )

Hi, I do my query from Perl DBI, it has a very big IN ( $value ). The query looks like: SELECT id FROM user WHERE sex='m' AND hobby_id='3' AND id IN ( $value ); $value looks like: 102, 233, 445, 543 ... 3443, 332 very long list. It might be about 10,000 id in $value. What is the maximum value

Re: Do I need Innodb?

Hi Ron, My Scenario is if two users open an existing record with data already in the fields. They then add some information to the end of a field with data. Now when they click update the last user/browser gets written to the db, where the first users data is over written. I thought

Re: Do I need Innodb?

On Fri, 16 Jan 2004 13:43:55 -0800 Ron McKeever [EMAIL PROTECTED] wrote: Scenario 1: I have noticed that if two users open a record to edit it in two different browsers and they edit the same field and then click update, the last user/browser gets written to the db, where the first users

Re: MySQL and Perl DBI Persistent Connection

On Fri, 16 Jan 2004 17:54:06 -0500 Syed Ali [EMAIL PROTECTED] wrote: I have a Perl CGI script which uses DBI to access MySQL. I noticed that each query through the Perl CGI script using DBI has a connect, query and then quit. I was wondering if there is a way of keeping the connection

Drop foreign key replication problem

Hi, I am running a master and a slave. When I created a foreign key in master, it looked like this: CONSTRAINT `0_373` FOREIGN KEY (`prefecture_id`) REFERENCES `prefecture` (`id`) But on the slave it was like: CONSTRAINT `0_451` FOREIGN KEY (`prefecture_id`) REFERENCES `prefecture` (`id`) The

What files are these?

I have some very big files in my /var/lib/mysql. [root@nias mysql]# ls -l nias-bin* -rw-rw1 mysqlmysql1073741921 5$B7n(B 8 04:33 nias-bin.001 -rw-rw1 mysqlmysql875448838 5$B7n(B 9 18:31 nias-bin.002 -rw-rw1 mysqlmysql 86 5$B7n(B 9

Re: Linux 7.1 Professional MySQL startup.

Hi Joe, On Thu, 14 Mar 2002 11:00:33 -0800 Joe Krusick [EMAIL PROTECTED] wrote: Hi: Installed the following packages (with no errors) from the Red Hat Linux V7.1 RPM files. It installed MySQL3 3.23.36-1, MySQL-devel3.23.36-1, MySQL-Server3.23.36-1, and MySQLclient9.3.23.22.4. However,

Re: very tough query

Hi Darren, On Tue, 5 Mar 2002 09:42:50 -0800 Daren Cotter [EMAIL PROTECTED] wrote: mysql desc poll_questions; ++-+--+-++- ---+ | Field | Type| Null | Key | Default| Extra |

Slow Query LEFT JOIN

I have some tables, with the rows like below: mysql select count(*) from job_main; +--+ | count(*) | +--+ | 6835 | +--+ 1 row in set (0.00 sec) mysql select count(*) from company_main; +--+ | count(*) | +--+ | 8177 | +--+ 1 row in set

Perl DBI and rollback()

Hello, I think it might not be a right mailing list for this question, but I don't know where else should I post it. If I run Perl DBI's $dbh-connect() without $dbh-disconnect(), MySQL (or is it the Perl DBI module) will run rollback() with an error message: Issuing rollback() for database

Re: Full Text Search and Japanese

On Wed, 20 Jun 2001 15:22:22 -0700 "Ryan Knopp" [EMAIL PROTECTED] wrote: I'm trying to use the Full text Search with the japanese language using either doublebyte or singlebyte character. The problem is it doesn't seem to be working at all for any searches. I use a sql statement like this

Why should not rely on $sth-rows() ?

Hi all, I read in DuBois's MySQL, Perl DBI API Reference part, it is written: $rv = $sth-rows(); Return the number of rows affected by the statement associated with $sth or -1 if an error occured. This method is used primarily for statements that do not return rows. For SELECT statements, you

The table is full: bug in 3.22.32?

Hello, I tried to make a query to MySQL, the version is: mysql Ver 9.38 Distrib 3.22.32, for pc-linux-gnu (i686). The result should be about 4843 rows, but I got this message instead of the result: ERROR 1114: The table 'SQL22e0b00_0' is full When I tried on version: mysql Ver 11.12 Distrib