[courier-users] Server hangup after exceeding maximum active

Hello Everybody,

Sometime during the night (IST +0530) my server gets swamped by new
SMTP connections and within no time (within a matter of a minute) I
get warning messages in syslog which says "corieresmtpd: 30 maximum
active connections". After this I see no activity at all, the server
just freezes and I can't even log in.

Sample log:

Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:82.57.26.129]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:201.254.94.70]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:91.165.247.125]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:59.144.40.9]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:59.144.40.9]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:122.168.4.201]
Feb 22 00:36:56 mail courieresmtpd: started,ip=[::ffff:87.240.35.133]
Feb 22 00:38:03 mail courieresmtpd: started,ip=[::ffff:91.165.247.125]
Feb 22 00:39:04 mail courieresmtpd: started,ip=[::ffff:83.30.139.202]
Feb 22 00:39:04 mail courieresmtpd: started,ip=[::ffff:122.168.4.201]
Feb 22 00:39:04 mail courieresmtpd: 26 active connections.
Feb 22 00:39:04 mail courieresmtpd: 30 maximum active connections.
Feb 22 00:40:12 mail courieresmtpd: 30 maximum active connections.
Feb 22 00:41:14 mail courieresmtpd: 30 maximum active connections.
Feb 22 00:42:20 mail courieresmtpd: started,ip=[::ffff:213.140.19.112]
Feb 22 00:42:20 mail courieresmtpd: started,ip=[::ffff:90.20.83.55]

I had scheduled a cron job to log the load average and memory usage,
and the load average stays well below 1 and no swapping happens. One
more thing to note is that these connections are all from a different
IP address and most of them are dynamic. Some of these connections
were rejected just minutes earlier due to listing in spamhaus but they
never got rejected after exceeding the connections.

After this stage (after exceeding the connections) there is no other
logs in the syslog other than the "maximum active connections" warning
by courieresmtpd. Only a reset brings back the server to a working
condition. This happens repeatedly, night after night. I have tried
varying the MAXDAEMON option, increased my RAM to 1GB, disabled
filters (pythonfilter, clamcour), disabled DNS lookups but nothing has
helped. Please help me solve this problem.

I use debian with courier-mta version 0.53.3 and clamcour version
0.3.8, have enabled sbl-xbl.spamhaus.org look-ups.

-Avinash.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?p...rge&CID=DEVDEV
_______________________________________________
courier-users mailing list
courier-users@lists.sourceforge.net
Unsubscribe: https://lists.sourceforge.net/lists/.../courier-users