Zen, the printout looks like the thread reading in the transaction system header from the first data file would have stuck.
Please send to me the printouts from other crashes. Maybe this is a bug in the trx system header handling. Though, my first guess is that this is an OS/drivers/hardware bug and the thread has hung inside a call of pread(). Unfortunately, the InnoDB Monitor thread has also hung on one of the reserved semaphores, and we do not get more detailed diagnostic printout below. I added to MySQL-4.0 now more diagnostics. It will always print the number of pending pread() and pwrite() calls if there is a long semaphore wait. Best regards, Heikki Tuuri Innobase Oy Foreign keys, transactions, and row level locking for MySQL InnoDB Hot Backup - a hot backup tool for InnoDB which also backs up MyISAM tables http://www.innodb.com/order.php Register now for the 2004 MySQL Users Conference! http://www.mysql.com/events/uc2004/index.html ----- Original Message ----- From: "ZenShadow" <[EMAIL PROTECTED]> Newsgroups: mailing.database.myodbc Sent: Thursday, March 11, 2004 12:58 AM Subject: Database crash, lock held too long > --=-4vGO7MeWwD4jujspp0iW > Content-Type: text/plain > Content-Transfer-Encoding: quoted-printable > > Folks, > > We've been running a fairly high-traffic mysql server for some time, > and we've been occasionally seeing messages like this in the logs: > > InnoDB: Error: semaphore wait has lasted > 600 seconds > InnoDB: We intentionally crash the server, because it appears to be > hung. > 040309 18:20:36 InnoDB: Assertion failure in thread 2623773248 in file > sync0arr.c line 934 > InnoDB: Failing assertion: 0 > InnoDB: We intentionally generate a memory trap. > InnoDB: Send a detailed bug report to [EMAIL PROTECTED] > > This is a fairly critical issue for us (the latest crash borked > replication, which is never fun), so if anyone has an answer it would be > most appreciated. I've attached the full relevant log snippet and the > my.cnf for anyone who's willing to take a look. > > I'd provide more information, but it has so far been impossible to tell > what's causing it. This server averages >2000 queries per second (25% > of which is inserts and updates, I'd guess), and is also the replication > master for another database. It seems to be random so far. > > This system is running on a dual Xeon (2.4GHz + hyperthread) with 4GB of > memory, Two U160 SCSI arrays (configured for 1TB @ RAID1+0), and the > stock RedHat 2.4.20-8smp kernel (the box is pretty much a stock RedHat 9 > install). > > Again, any insight would be most appreciated. > > --ZS > -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]