Simon, Thank you for good suggestion. I checked the variable. It's value is 28800. I am sure that none of my queries takes 8 hours to complete.
Most probably Django keeps connections open between requests (make sense for performance). That connection is closed after 8 hours but Django (?) assumes it is still open and breaks when query is unsuccessful. Am I right? This problem is compounded (possibly) with related problems. I use FCGI method outlined here: http://wiki.dreamhost.com/index.php/Django. It looks like after exception I have a dead process in memory, which I have to kill manually using "kill -9". Otherwise it is going to be here forever. Well, most probably it is a problem of fcgi.py. Anyway, number of python2.3 processes grows until I hit some predefined limit => system tries to kill these processes => they don't care (should be killed with -9) => they are still around => web site dies, until I log in to shoot the undead processes down. Any help is appreciated. Thanks, Eugene PS: I wish more providers allowed to use mod_python and PostgreSQL... "GrumpySimon" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] > > This error often occurs when MySQL times out because the query's taking > too long. > > At a guess this would be at the MySQL end of things - try running this > command at the mysql prompt: > > SHOW VARIABLES LIKE 'wait_timeout'; > > You should get something like this: > +---------------+----------+ > | Variable_name | Value | > +---------------+----------+ > | wait_timeout | 28800 | > +---------------+-----------+ > 1 row in set (0.00 sec) > > > Where Value is the number of seconds. You can override this by issuing > this command: > set wait_timeout = $largernumber > > More reasons for this error are here: > http://dev.mysql.com/doc/mysql/en/gone-away.html > > --Simon > >
