Daniel Gehriger
daniel.gehriger@linkcad.com
Wed, 22 Nov 2006 21:31:02 +0100
Hi, I'm running the latest DCC as a postfix before-queue filter, as described in the SMTPD_PROXY_README. The DCCIFD_ARGS in the config file is: DCCIFD_ARGS="-Bset:debug=5 -Brelays.ordb.org,any -Bsbl-xbl.spamhaus.org,any -p 127.0.0.1,10023,127.0.0.1/32 -o 127.0.0.1,10026 -SHELO -Smail_host -SSender -SList-ID" After starting dccd / dccproc, everything works fine: > Nov 22 09:15:30 vps183 dccifd[3629]: DNSBL helper process starting on 127.0.0.1,43964 > Nov 22 09:15:34 vps183 dccifd[3257]: 3jq5DV DNSBL helper: hit sender 217.132.229.16 gethostbyname(16.229.132.217.sbl-xbl.spamhaus.org)=127.0.0.4 > Nov 22 09:15:34 vps183 dccifd[3559]: 3jq5DV DNSBL helper answered 1 for sender 217.132.229.16 > Nov 22 09:15:34 vps183 dccifd[3559]: client DNSBL hit 16.229.132.217.sbl-xbl.spamhaus.org After a few minutes of idling, the DNSBL help process exits: > Nov 22 09:32:12 vps183 dccifd[13393]: DNSBL helper process on 127.0.0.1,43964 idle, exiting The problem is that it sometimes doesn't automatically start when required: > Nov 22 09:34:21 vps183 dccifd[6017]: 3jq5Ea no DNSBL helper answer > Nov 22 09:34:21 vps183 dccifd[6017]: 3jq5Ea DNSBL failed for sender 64.40.109.130, 15.0 msg-secs remaining DNSBL lookups continue to fail for anything between a couple of minutes and several hours until the help process starts again: > Nov 22 09:38:01 vps183 dccifd[2034]: DNSBL helper process starting on 127.0.0.1,46780 > Nov 22 09:38:01 vps183 dccifd[2034]: 1pKxst DNSBL helper: hit sender 88.154.101.218 gethostbyname(218.101.154.88.sbl-xbl.spamhaus.org)=127.0.0.4 Why dccproc sometimes fail to start a new help process ? - Daniel