Wendell, InnoDB writes a checksum to a database page when it is written to disk. If the checksum does not correspond to the page contents when the page is read back in, you get the below error. Below page 36819 in table registrydb_tn/TBL_AllNames appears to be corrupt, like it says. The checksum is 0, which does not correspond to page contents.
When you encounter this kind of error, the first thing to try is rebooting the computer, like it says below. Next you can try dumping, dropping, and reimporting the corrupt table. The last resort is to recover from a backup. >020306 09:01:16 mysqld restarted >020306 9:01:16 Can't start server: Bind on TCP/IP port: Address already in >use >020306 9:01:16 Do you already have another mysqld server running on port: >3306 ? The last error you got was probably that Linux had not killed the entire mysqld process though mysqld had called exit(1). Then you should look with ps -A what processes are running, and kill -9 them manually. Please send me all error messages and hex dumps you have, from both servers. What version you are running? Linux kernels 2.2 -2.5 seem to have bugs in the i/o system, especially in connection with RAID disks. But let us look first at the hex files. Regards, Heikki -----Original Message----- From: "Wendell Dingus" <[EMAIL PROTECTED]> Newsgroups: mailing.database.mysql Date: Thursday, March 07, 2002 6:16 AM Subject: Innodb problems >I've been using MyISAM tables for a long time on a lot of machines and >(knock on wood) have basically never had a problem. Now though on a larger >server I decided we should use INNODB tables and am having problems. I >didn't realize until today that it's been 'crashing' and recovering >repeatedly for the past few weeks. Today it died though and could not repair >itself. Since I'm doing replication to a second server I just copied >databases from the backup server to the primary and got things back going. >Looking at the log files on the second/backup server it has the same type of >errors in the log file though. Not bad hardware unless both (identical) >servers have the same bad hardware. > >Here's where it crashed and dumped a heck of a lot of hex data into the log >file: > >......................................................................;Inno D >B: End of page dump >InnoDB: Page checksum 1558702454 stored checksum 0 >InnoDB: Page lsn 4 226263974, low 4 bytes of lsn at page end 0 >InnoDB: Page may be an index page where index id is 0 565 >InnoDB: Database page corruption or a failed >InnoDB: file read of page 36819. >InnoDB: You may have to recover from a backup. >InnoDB: It is also possible that your operating >InnoDB: system has corrupted its own file cache >InnoDB: and rebooting your computer removes the >InnoDB: error. > >Number of processes running now: 0 >020306 08:55:21 mysqld restarted >020306 8:55:25 InnoDB: Database was not shut down normally. >InnoDB: Starting recovery from log files... >InnoDB: Starting log scan based on checkpoint at >InnoDB: log sequence number 4 229089185 >InnoDB: Last MySQL binlog file offset 0 465620, file name ./shelby1-bin.001 >020306 8:55:26 InnoDB: Flushing modified pages from the buffer pool... >020306 8:55:26 InnoDB: Started >/usr/sbin/mysqld-max: ready for connections >InnoDB: Database page corruption or a failed >InnoDB: file read of page 36819. >InnoDB: You may have to recover from a backup. >InnoDB: Page dump in ascii and hex (16384 bytes): > len 16384; hex 0000000000008fd30005851f00058 > > >Here's where it gave up the ghost entirely: > >......................................................................;Inno D >B: End of page dump >InnoDB: Page checksum 1558702454 stored checksum 0 >InnoDB: Page lsn 4 226263974, low 4 bytes of lsn at page end 0 >InnoDB: Page may be an index page where index id is 0 565 >InnoDB: and table registrydb_tn/TBL_AllNames index LastName >InnoDB: Database page corruption or a failed >InnoDB: file read of page 36819. >InnoDB: You may have to recover from a backup. >InnoDB: It is also possible that your operating >InnoDB: system has corrupted its own file cache >InnoDB: and rebooting your computer removes the >InnoDB: error. > >Number of processes running now: 0 >020306 09:01:16 mysqld restarted >020306 9:01:16 Can't start server: Bind on TCP/IP port: Address already in >use >020306 9:01:16 Do you already have another mysqld server running on port: >3306 ? >020306 9:01:16 Aborting > >020306 9:01:16 /usr/sbin/mysqld-max: Shutdown Complete > >020306 09:01:16 mysqld ended > > >Here's the my.cnf file: > >[mysqld] >datadir=/var/lib/mysql >socket=/var/lib/mysql/mysql.sock >log-bin >server-id=1 > >default-table-type=innodb >innodb_data_file_path = >ibdata1:2000M;ibdata2:2000M;ibdata3:2000M;ibdata4:2000M;ibdata5:2000M;ibdat a >6:2000M >innodb_data_home_dir = /var/lib/innodb/ >set-variable = innodb_mirrored_log_groups=1 >innodb_log_group_home_dir = /var/lib/iblogs >set-variable = innodb_log_files_in_group=3 >set-variable = innodb_log_file_size=100M >set-variable = innodb_log_buffer_size=16M >innodb_flush_log_at_trx_commit=1 >innodb_log_arch_dir = /var/lib/iblogs >innodb_log_archive=0 >set-variable = innodb_buffer_pool_size=800M >set-variable = innodb_additional_mem_pool_size=40M >set-variable = innodb_file_io_threads=4 >set-variable = innodb_lock_wait_timeout=50 > >[mysql.server] >user=mysql >basedir=/var/lib > >[safe_mysqld] >err-log=/var/log/mysqld.log >pid-file=/var/run/mysqld/mysqld.pid > > > >This is on an IBM xSeries 350 server, dual P3 Xeon 700's, 1.5GB RAM, IBM >ServRAID 4LX RAID controller, RedHat 7.2 with all patches (2.4.9-31 RH RPM >kernel). Any ideas on how to prevent this will be greatly appreciated! >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 > --------------------------------------------------------------------- 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
