hi, we are running a forum with about 600 to 700 people online at the same time plus a shop and other database relying services. because this was to much for our single mysql server we are running a master-slave setup now, both of version 4.0.23. now we get occasional 2013 errors from our forum, lost connection to mysql server during query. we didn't get these on our old setup. the old setup was slow but worked without errors. the appearance of these errors did not start just in time when the new setup was online, the errors started to occur some days ago. they occur only on insert and update statements, so they are produced by our master, and they occur absolutely by chance. everything is okay most of the time, but now and then it occurs. sometimes just one error message, sometimes 1000 of it at once, serveral times every day. our forum delivers us the error messages by email, all other services on our site don't. because everythings fine the most of the time i think our forum is fully compatible with the php and mysql version we are using and the code is okay too. now i'm asking myself if the mysql server may be configured wrong for our needs, especially regarding memory management. on my web search i came across lots of articles dealing with max_allowed_packet or thread_stack. in the server log we get a warning that mysql gets a smaller thread_stack than requested, but most of the answers i found on the web were like "ignore it". the machine is a suse linux kernel 2.6.10. i pasted the parts of our mysql config about which i think that they may be important. help on this and explanations of possible causes is very very much appreciated.
best regards, oliver # The MySQL server [mysqld] port = 3306 socket = /tmp/mysql.sock skip-locking key_buffer = 384M max_allowed_packet = 1M table_cache = 3072 sort_buffer_size = 2M read_buffer_size = 2M myisam_sort_buffer_size = 64M log-slow-queries = /var/log/mysql-slow.log set-variable = long_query_time=10 set-variable = thread_stack=150K connect_timeout=60 wait_timeout=60 interactive-timeout=60 set-variable = max_connections=2000 thread_cache = 8 query_cache_size = 32M # Try number of CPU's*2 for thread_concurrency thread_concurrency = 8 [mysqldump] quick max_allowed_packet = 16M [mysql] no-auto-rehash # Remove the next comment character if you are not familiar with SQL #safe-updates [isamchk] key_buffer = 256M sort_buffer_size = 256M read_buffer = 2M write_buffer = 2M [myisamchk] key_buffer = 256M sort_buffer_size = 256M read_buffer = 2M write_buffer = 2M [mysqlhotcopy] interactive-timeout -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
