I am getting ~3500 unique "Connection Reset" error messages in the SMTP
log each day. Some are to the same host. But about 2500 of the errors
are to unique hosts.
Example messages (I removed the stack traces that follow each message in
the log...)
10/12/07 03:10:55 ERROR smtpserver: Socket to X24.D-IP06.lipetsk.ru
(195.34.253.24) closed remotely.
10/12/07 03:11:11 ERROR smtpserver: Socket to 118.175.74.69
(118.175.74.69) closed remotely.
10/12/07 03:11:43 ERROR smtpserver: Socket to ipa10.42.91.tellas.gr
(91.140.42.10) closed remotely.
I suspect that much of this is spam. But I'm not sure all of it is.
And even if it's spam, why is the TCPIP layer failing?
This has been going on for quite some time. I'm still getting mail
through the system. But I'm really concerned. This just doesn't seem
right. Can someone explain what is happening here? It doesn't say
'timeout'. And I have trouble believing that 2000-3000 servers each day
are closing connections to my server, and somehow it's all their problem
and not mine.
Basically, is this normal? What is happening in the protocol in
communicating with these thousands of servers that is not correct? If
these aren't "real" errors, then I'd really like to get these messages
out of the logs so I have a reasonable chance of finding the real errors
that I need to deal with. If this is not normal to have this many of
these errors per day, please tell me what is wrong with my server.
BTW.... I'm watching my CPU utilization, and it's barely moving off of
zero most of the day. So this is not resulting from an overloaded
system or anything.
Suggestions? Comments?
Thanks.
Jerry
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
