I went to the production server (Solaris 2.8 on E250) and tarred the data directory for this database. Then I FTPed it over to the test server (Solaris 2.8 on Sparc 20), deleted the files in the same data directory and untarred the files from the production server. The database is running quite fine on the testing server now.
I still suspect some memory limitation on the Sparc 20 preventing the "un-dump" from completing. In re-checking my server data list, the 20 has 1/2 the memory of the E250 (256 Meg on the 20 vs. 512 on the E250 - also 512 on the W2K machine).
Still, it may be a concern to the MySQL developers - not that there are memory limitations, but rather that they resulted in error messages that were both incorrect and not indicative of the real problem.
-Richard
Richard S. Huntrods wrote:
I am having a strange problem reloading data into one copy of MySQL. Details:production server: MySQL - Ver 8.38 Distrib 4.0.5a-beta, for sun-solaris2.8 on sparc development system: MySQL - Ver 8.38 Distrib 4.0.5a-beta, for Win95/Win98 on i32 test system: MySQL - Ver 8.38 Distrib 4.0.5a-beta, for sun-solaris2.8 on sparc production server: Sun E250, Solaris 2.8, 9 gig disk (plenty of free space), 256 M ram. development system: Toshiba laptop, Win2000, 15 gig disk (plenty of free space), 512 M ram. test system: Sun Sparc-20, Solaris 2.8, 9 gig disk (plenty of free space), 256 M ram. All systems are "nominal". Every day, the E250 server runs a cron job to backup the database via the command: mysqldump -u root --lock-tables --databases mydb >CurrentMydb.mysql This has worked (and still works) for over 1.5 years. The database dump is FTPed to the development machine. The database dump is currently 48 Megabytes in size (was 45, but usage logs grow...). The database is rebuilt on the Win2000 server every couple of days using a set of scripts that remove the database and rebuild it from the dump file with the following command: mysql -u root <CurrentMydb.mysql where "CurrentMydb.mysql" is the dump file from the E250 This works fine (even today). However, the same sequence of commands (that has been working for 1.5 years) issued TODAY on the testing server gives the following error: ERROR 1064 at line 430725: You have an error in your SQL syntax. Check the manual that corresponds to your MySQL server version for the right syntax to use near 'VALUES (51172,700,3,0,0,-1,1,'Mon Feb 17 19:59:27 MST 2003')' a However, I've FTPed the dump from the E250 to the Sparc 20 three times (and verified they are identical), and each time I run the rebuild script I get the same type of error - but targeting a different line every time!!! Also - I've used this file on the W2K machine - there is no actual error in the file. The only thing that has changed at all is the size of the database dump (from 45M to 48M). There is plenty of disk space. Has anyone encountered such a problem before, and is there a workaround? Thanks in advance, -Richard
--------------------------------------------------------------------- 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
