Package: postfix-policyd Version: 1.80-2.3 Severity: normal Hi,
I had a crash of MySQL a few days ago which destroyed the database of postfix-policyd. I didn't notice, until all my applications complained they couldn't connect to MySQL due to too much open connections. Policyd held 87 open connections. Then I saw that the database was empty and all tables were gone. I guess a select of policyd failed and it assumed the connection is down. Therefore, it had opened a new connection and had thrown away the still open and working connection. If this is the case, can you enhence the code to check what went wrong and only open a new connection, if the connection is really gone? Maybe you could check the availability of the connection with a simple select NULL; that succeeds even if no database is selected. In case of a broken database, you should also print a message to the log, because in my situation I found nothing that indecated any database problem, even through the select/update after the open of a new connection failed. Bye, Jörg -- System Information: Debian Release: unstable/experimental APT prefers unstable APT policy: (900, 'unstable'), (700, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.34 Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash
signature.asc
Description: Digital signature http://en.wikipedia.org/wiki/OpenPGP
