mysql query not dead after 'kill'
I have run into this a few times now, and am wondering there is a solution: I have two queries: (1) select query on a MERGE table (2) 'alter table' on the same MERGE table. As expected, the select query will lock the 'alter table' query. The alter table query is intended to swap one of the tables in the MERGE set. Now if the 'select' query isn't critical, I would like to kill it if it takes too long. However, as a result I have the 'kill flag' set on the query, for a long time, without it actually dying (e.g. it says in 'send data' status). is there anyway to avoid this issue? E.g. a way to kill the select faster? Usually I am ending up with load running out of control, and in a few cases, mysql is crashing as a result. Also: the status of the 'alter table' is 'rename result table', not 'locked' or 'waiting for table'. -- -- Johannes Ullrich [EMAIL PROTECTED] pgp key: http://johannes.homepc.org/PGPKEYS -- "We regret to inform you that we do not enable any of the security functions within the routers that we install." [EMAIL PROTECTED] -- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: PERL DBI DBD mysql / REMOTE ACCESS
> I added Host 192.168.1.10 user root with passowrd in mysql user table. Did you flush privileges? did you hash the password using the 'password' function? did you type the password correctly ? > > Ideas? What else? > > Thanks. -- ------ Johannes Ullrich [EMAIL PROTECTED] pgp key: http://johannes.homepc.org/PGPKEYS -- "We regret to inform you that we do not enable any of the security functions within the routers that we install." [EMAIL PROTECTED] -- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
mysql 'range' query optimization
This problem is bugging me for a while now: I have a table which has three columns: 'table': start: long integer end: long interg property: varchar(200) Now I am trying to retrieve the 'property' for a 'number': select property from table where number between start and end well, easy enough. Problem: It takes foreve, and according to 'explain', index are sometimes used and sometimes are not used (for identical queries) The table has about 700,000 rows. I have indexes on start, end and (start,end). start and end are unique. 'end' is the primary key. there are supposed to be no overlaps, but there are numbers that are not covered by any range. 'explain' sometimes shows the query using a key, sometimes it doesn't :-(... all of this is kind of 'random'. The table is rather static (<100 updates / day) This problem has haunted me from 3.x days. Currently I am running 4.0.15 Different ways I did try to ask the query: select property from table where end>number order by end asc limit 1 select property from table where number>start and numberhttp://johannes.homepc.org/PGPKEYS -- "We regret to inform you that we do not enable any of the security functions within the routers that we install." [EMAIL PROTECTED] -- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
too many open files, error 24, but max-files is large
on mysql 4.0.10-gamma (rpm install, Redhat advanced server), I am running into 'too many open files' issues ( error 24 ). I am using a rather large merge table (30 distinct tables merged), which is likely the culprit. The error shows up as I have about a dozen of connections. I did increase the number of file handles substantionally (16384) and it looks like this limit is not reached: cat /proc/sys/fs/file-nr 5328475816384 This is a 2.4 kernel, so inodes are allocated automatically and there is no inode-max setting. 'open_files_limit' is set to 0. 'table_cache' is set to 1024 cmd line used to run mysqld: /bin/sh /usr/bin/mysqld_safe --log-slow-queries --datadir=/var/lib/mysql --pid-file=/var/lib/mysql/x440.pid ... /usr/sbin/mysqld --basedir=/ --datadir=/var/lib/mysql --user=mysql--pid-file=/var/lib/mysql/x440.pid --skip-locking--log-slow-queries in other words: I checked ulimit... What am I missing? hardware: dual Xeon system, 8 GByte RAM -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: Fwd: how to handle a 10GB index-file?
Two solutions to your problem: table full errors: check the 'AVG_ROW_LENGTH' and 'MAX_ROWS' option for create and alter table. You can change these on the fly using 'alter table', but it will take quite a time for a table your size (few hours-1 day depending on machine). The exact values for these parameters are not all that important. just make it large enough. Usually, you run into this limit around 4 Gig Byte, but maybe you already set these parameters and need to increase them. Check 'show table status' to verify the current value. To reduce the size of you index file: try a 'repair table'. The 'alter table' will take care of it too. You can also drop and recreate this index. Side note: look at merge tables... they look like a pain at first, but once you get used to it they are quite handy for tables your size. > The data file .MYD of a certain table has a size of 7.3 Gig and the > corresponding index file .MYI has a size of 10Gig. When deleting records from > this table - will the index file automaticly be stripped down? IF not - is > there a way to trim the index-file to use less space?!? > > BTW: how is the maximum size of a index file determined? yesterday, one of > our databases refused to insert more data into the table, because there was > no space left in the index-file (17GB) - but the filesystem had plenty of > space left (SOLARS 8). > > Is the any Documentation regarding this issue? > > thanks, > > Andy > > - -- > [EMAIL PROTECTED] > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.2.1 (GNU/Linux) > > iD8DBQE+ZgGpRrny/uOBVy4RAsPZAJ9XI/OrOuhfwMGEnBpbvBkKHyTtngCghoqJ > KRQc0/zSIAnQQkAG+V0zOns= > =oi3/ > -END PGP SIGNATURE- > > > - > Before posting, please check: >http://www.mysql.com/manual.php (the manual) >http://lists.mysql.com/ (the list archive) > > To request this thread, e-mail <[EMAIL PROTECTED]> > To unsubscribe, e-mail <[EMAIL PROTECTED]> > Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php > > -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: replication "Error updateing slave list" in mysql 4.0.10
> Check the user 'repl' has REPLICATION SLAVE privilege. Ah. that fixed it. Actually, the real reason was that I had not yet updated the mysql tables and the new privileges did not take effect as a result. mysql_fix_privilege_tables , followed by the 'GRANT' command and 'flush privileges' fixed it. Thanks! -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
replication "Error updateing slave list" in mysql 4.0.10
I am having problems setting up replication between two 4.0.10 servers. What I did so far: - generate a dump of the current state of the server using 'mysqldump' (its a mix of mostly innodb tables and some MyISAM tables) - dropped all databases from the slave - imported the dump into the slave using mysql < dump - updated the slave parameters using 'CHANGE MASTER' As I start the slave, it immediatly stops and I am getting the following in the slave error log: 030213 4:36:30 Slave I/O thread: connected to master 'repl@master:3399', replication started in log 'slave-bin.005' at position 86967189 030213 4:36:30 Error updating slave list: Query error 030213 4:36:30 Slave I/O thread exiting, read up to log 'slave-bin.005', position 86967189 not that the master is running on port 3399 (ssh tunnel). I can connect to the master as 'repl' from the slave. looking at the source shows this error in conjunction with 'show slave hosts'. This command returns an empty result on master and host. Not sure what it is supposed to return. I did try a 4.0.5 master first with the same result. Now I upgraded the master to 4.0.10 (slave ran 4.0.10 all along). -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
speedup 'alter table' for large tables
I just had to alter a large (25Gig, 100Million rows) table to increase the max_rows parameter. The 'alter table' query is now running 60+ hours, the last 30+hours it spend in 'repair with keycache' mode. Is there any way to speed up this operation? I realize, it is probably too late now. But next time around it would be handy to know a faster way. - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: encrypted password
the easiest way to do this is to use mysql's own 'password' function. to add a new user use: insert into table (username,passwd) values ('jianping',password('jian1830')) to validate the password: select count(*) from table where username='jianping' and passwd=password('whatwasentered'); or similar... On Tue, 4 Feb 2003 14:58:28 -0500 (EST) Jianping Zhu <[EMAIL PROTECTED]> wrote: > > I have mysql in a redhat machine. I need to use mysql do user > authentication to a website. > I have a table like following. > > +--+--+ > | username | passwd | > +--+--+ > | jianping | jian1830 | > | chichi | jian1830 | > +--+--+ > > I want the passwd field not to be plain text but encrypted. how can i do > that? > > Thanks. > > > > - > Before posting, please check: >http://www.mysql.com/manual.php (the manual) >http://lists.mysql.com/ (the list archive) > > To request this thread, e-mail <[EMAIL PROTECTED]> > To unsubscribe, e-mail <[EMAIL PROTECTED]> > Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php > > -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: tab-delimited text file import
> Does anyone have any hints on how to or where to look to find out how > to import a number of tab-delimited text files with some header info > that reside on a ftp server into a MySQL database using PHP? the 'key' command is 'load data infile'. It is very flexible in handling various delimited file formats. > How about doing this on a daily basis, automagically? well, a little perl script as a cron job? php script will do too. -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: Storage issue
> I'm wondering how well MySQL compress data. I'm about to design a > database which will hold mainly _a lot_ of FLOAT-values, and since I do > not really know how well MySQL compress data or how I could calculate > this I'd really appriciate a little guidance. see chapter 6.2.6 of the mysql manual. > 1 value/minute are stored = 1440 values/day. > 365 days / year. > > We have 100 different tables with 25 columns each. > This makes 100*25*365*1440 = 1 314 000 000 values per year. float: 4 byte, so you need about 5 gigs? But you will need space for indexes and such as well. So just get an 18 Gig drive for the start. -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: Re[4]: mysql 4.0.8- crash on TCP connection
On Thu, 9 Jan 2003 22:56:04 +0200 "Gelu Gogancea" <[EMAIL PROTECTED]> wrote: > > All this is very interesting, BUT i have two binary builds (4.0.7 & > 4.0.8), > Ha,Hayou are gentle. > > > (load avg 10-60 Query/sec), and 4.0.8 crash (in some > > hardware/software) after 2 seconds work :( > Did you see any relationship with 'replication'? I just downloaded 4.0.8. All it did was act as a slave. It crashed after a large 'load table' from the master and now refuses to start. Looks like it crashes as it read the relay-log-info file or just after it does so. at least thats my latest theory after doing more backtrace resolving, stracing and experimenting. did submit one or two bug reports about this. -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: performance tuning
> MySQL 3.23.51 > Linux Kernel 2.4.16 > we do have 1 GB of RAM > the main problem seems to be a table with about 8.597.146 records. Similar situation here (>100 Million rows). things I found that help: - be selective on what rows to index. Try to limit yourself to one row. - increase the key buffer size. (watch memory / swap usage as you do this). - maybe you need more RAM (but see my other post about a question I had about large RAM machines and mysql crashing ;-) ) And more difficult, try to look at your application design and try to come up with 'cache tables' that are generated by a cron job periodically and are used for most queries. -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
mysql 3.23.53a-Max crashes on large memory machine
I am having 'issues' with MySQL running on Redhat Advanced Server on a 8 Gigbyte machine (dual P4-Xeon). After large imports ('load data infile', file size about 1 Gigbyte) into a large table (20-30 GByte, > 100 Million rows), the database crashes. I did try several key_buffer_size settings. The example below had the key_buffer_size set to 6 Gig (but as you see, it shows -50MByte ?). A key buffer size of a bit less then 4 Gig didn't help. I did try a couple of other versions, including 4.0.4 and all of them crashed. The machine is running only one mysql instance at the time. The main issue I try to solve is how to optimize lookups against this one big table. Its index alone is about 10-15 gig... >From the log file: - mysqld got signal 11; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked agaist is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail key_buffer_size=-50335744 record_buffer=209711104 sort_buffer=209715192 max_used_connections=2 max_connections=100 threads_connected=2 It is possible that mysqld could use up to key_buffer_size + (record_buffer + sort_buffer)*max_connections = 3161707 K bytes of memory Hope that's ok, if not, decrease some variables in the equation Attempting backtrace. You can use the following information to find out where mysqld died. If you see no messages after this, something went terribly wrong... Stack range sanity check OK, backtrace follows: 0x806eeb4 0x82d9b38 0x830592f 0x80a194f 0x807598a 0x80742e7 Stack trace seems successful - bottom reached Please read http://www.mysql.com/doc/U/s/Using_stack_trace.html and follow instructions on how to resolve the stack trace. Resolved stack trace is much more helpful in diagnosing the problem, so please do resolve it Trying to get some variables. Some pointers may be invalid and cause the dump to abort... thd->query at (nil) is invalid pointer thd->thread_id=9 - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: R: Mysql Replication
> Then, the master write, update, delete, ecc.. > the slaves answer SELECTs > > If this is impossible, which is the utility to have slaves??? it is possible. Follow the manual for a good start. A couple caveats: - your application has to be able to send the selects to the different (read only) databases on its own. mysql will not help you with the actual query funneling. - be aware that the slave may not be up to date. All replication is done 'in serial'. An update that takes a long time will block all other updates from the master. - if you use 'LOAD DATA INFILE', make sure you do not delete the data file until all slaves have it imported. -- [EMAIL PROTECTED] Collaborative Intrusion Detection join http://www.dshield.org - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
Re: database create fails with error 28
> What the hell is error 28? Where can I find description? > mysqlcheck --all-databases # says everything is ok mysql does come with a little utiltiy, perror, which can be used to translate error numbers into readable messages: $ perror 28 Error code 28: No space left on device Did you check if you have enough disk space? - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
mysql crashes after adding memory.
Yesterday, I increased the memory in my mysql server from 2 GByte to 4 GByte. Here the log file as it died: Number of processes running now: 1 mysqld process hanging, pid 1015 - killed 020808 09:40:12 mysqld restarted 020808 9:40:12 Can't start server: Bind on TCP/IP port: Address already in use020808 9:40:12 Do you already have another mysqld server running on port: 3306 ? 020808 9:40:12 Aborting 020808 9:40:12 /usr/local/mysql/bin/mysqld: Shutdown Complete 020808 09:40:12 mysqld ended The database was under heavy load at the time. The machine is a 'vanilla' dual 1 GHz P-III running RedHat Linux and a 2.4.17 kernel Server version: 3.23.49a-max-log After doing the memory upgrade, I also adjusted some of the parameters, in particular, I increased the key buffer to 2 Gig. The database has a very large (100,000,000 row, 30 GigByte) MyISAM table, and a few smaller myisam and innodb tables. It ran ok for at least half a year (last machine reboot). So far, it does not look like any data was corrupted. Usually, only a small number of client (10 or so) is connected. Here my my.cnf file: set-variable=max_connections=200 set-variable=max_allowed_packet=10M set-variable=key_buffer=2000M set-variable=join_buffer_size=200M set-variable=sort_buffer=150M set-variable=record_buffer=50M set-variable=record_rnd_buffer=50M set-variable=table_cache=200 set-variable=thread_concurrency=4 set-variable=long_query_time=5 innodb_data_home_dir=/usr/local/mysql/innodb innodb_data_file_path=data1/ibdata1:1M;data2/ibdata1:1M set-variable=innodb_buffer_pool_size=50M set-variable=innodb_additional_mem_pool_size=10M innodb_log_group_home_dir=/opt/innodb_logs innodb_log_arch_dir=/opt/innodb_logs innodb_log_archive=1 set-variable = innodb_log_files_in_group=3 set-variable = innodb_log_file_size=3M set-variable = innodb_log_buffer_size=100 innodb_flush_log_at_trx_commit=1 set-variable = innodb_file_io_threads=4 set-variable = innodb_lock_wait_timeout=50 - Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php