Bruno wrote: > Hello,
> My mail server goes down some times and I don't know why. Here are two > logs, first the logwatch log: > --------------------- amavis Begin ------------------------ > 23 messages checked and passed. > **Unmatched Entries** > starting. amavisd-new at domain.com amavisd-new-20030616-p10, > Unicode aware, LANG=en_US: 2 Time(s) > NOTICE: Disconnected from SQL server: 2 Time(s) > lookup_sql: 2006, MySQL server has gone away: 2 Time(s) > TROUBLE in check_mail: creating_partsdir FAILED: DBD::mysql::st > execute failed: MySQL server has gone away at (eval 38) line 238, > <GEN17> line 5073.: 1 Time(s) > TROUBLE in check_mail: creating_partsdir FAILED: DBD::mysql::st > execute failed: MySQL server has gone away at (eval 38) line 238, > <GEN35> line 148.: 1 Time(s) > PRESERVING EVIDENCE in /var/lib/amavis/amavis-20060830T163617-27901: 1 > Time(s) > PRESERVING EVIDENCE in /var/lib/amavis/amavis-20060829T171649-27899: 1 > Time(s) > ---------------------- amavis End ------------------------- > And this is the mail.log error: > --------------------------------------------------- > Aug 14 14:52:14 domain amavis[26931]: (26931-06) lookup_sql: 2013, > Lost connection to MySQL server during query > Aug 14 14:52:14 domain amavis[26931]: (26931-06) NOTICE: Disconnected > from SQL server > Aug 14 14:52:14 domain amavis[26931]: (26931-06) TROUBLE in > check_mail: creating_partsdir FAILED: DBD::mysql::st execute failed: > Lost connection to MySQL server during query at (eval 38) line 238, > <GEN17> line 1968. > Aug 14 14:52:14 domain amavis[26931]: (26931-06) PRESERVING EVIDENCE > in /var/lib/amavis/amavis-20060813T183534-26931 > --------------------------------------------------- > This errors appear "randomly" at different times, but there are some > patterns. Some times, it happened some seconds after rdiff backup > begins. But then other times it has nothing to do with rdiff. > Also, amavis was stoping and not going back up alone, and making al > mail stay in the queue. > If I start again amavis and made a reload of postfix it will proces > queue and work normaly. > I've commented out this line in my.conf > # skip-external-locking > As I read in this forum: > http://www.howtoforge.com/forums/showthread.php?t=3705&page=2 > I've configured my postfix-courier-spamassassin-amavisd with this tutorial: > http://www.howtoforge.net/virtual_postfix_mysql_quota_courier > I'm still having the problem, and every time it goes down I lost my > mails. some of them stay in the queue and after reloading postfix they > are processed, but others just never get through. > I'd really appreciate any advice. > thanks. I would try increasing the default wait_timeout. Looks like you process very few messages, so the connection may time out. I believe newer versions of amavisd-new work around this problem. In my.cnf wait_timeout 60000 http://dev.mysql.com/doc/refman/4.1/en/gone-away.html Gary V ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ AMaViS-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/amavis-user AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3 AMaViS-HowTos:http://www.amavis.org/howto/
