I appreciate all the assistance. The problem is now gone. I haven't changed anything on my end, so I assume the techs on the other end found the problem.
Thank you to everyone for their suggestions.
I am running the 4.01b MercuryE module. About a month or so ago, we begen getting "Connection Refused" errors back from one specific domain (I haven't changed anything in my configuration.) The other domain says that we are the only one they cannot receive mail from. We can get mail to every other domain except theirs. We have been through all the DNS settings and they have all verified. Additionally, I migrated to 4.52 with the same effect.
Session logging returns:
13:32:16.343: --- Wed Oct 24 13:32:16 2007 ---
13:32:16.343: Connect to '206.255.247.19', timeout 60.
13:33:17.359: 8: Socket read timeout
13:33:17.359: --- Connection closed normally at Wed Oct 24 13:33:17 2007. ---
13:33:17.359:
1. What would cause a socket read timeout?
2. They have suggested that we use the HELO command instead of the EHLO command. How do I go about implementing that changing in Mercury?
Thanks in advance for any help that you can provide.
Steve LaVergne
Malvern School District
System Administrator
[quote user="slavergne"]
I am running the 4.01b MercuryE module. About a month or so ago, we begen getting "Connection Refused" errors back from one specific domain (I haven't changed anything in my configuration.) The other domain says that we are the only one they cannot receive mail from. We can get mail to every other domain except theirs. We have been through all the DNS settings and they have all verified. Additionally, I migrated to 4.52 with the same effect.
Session logging returns:
13:32:16.343: --- Wed Oct 24 13:32:16 2007 ---
13:32:16.343: Connect to '206.255.247.19', timeout 60.
13:33:17.359: 8: Socket read timeout
13:33:17.359: --- Connection closed normally at Wed Oct 24 13:33:17 2007. ---
13:33:17.359:
1. What would cause a socket read timeout?
2. They have suggested that we use the HELO command instead of the EHLO command. How do I go about implementing that changing in Mercury?
Thanks in advance for any help that you can provide.
Steve LaVergne
Malvern School District
System Administrator
[/quote]
The system will fallback to HELO when EHLO does not work so that's not the problem. We really need to know the domain where you are having the problem to to provide more help though. FWIW, I can't connect to port 25 of 206.255.247.19 and get a SMTP or ESMTP server so I suspect there is some sort of DNS setup problem.
[quote user="Thomas R. Stephenson"] FWIW, I can't connect to port 25 of 206.255.247.19 and get a SMTP or ESMTP server so I suspect there is some sort of DNS setup problem.[/quote]
For me, I eventually get a banner response
[code]220 barracuda.hsccs.org ESMTP (d6880a56f4eed9215d7f6e8e427dc12c)[/code]
but only after about 95 seconds. After that initial delay, SMTP responses continue normally. I suspect that domain have recently installed/updated the Barracuda device for anti-malware protection which is why Steve had no problems before.
Try increasing the MercuryE timeout from 60 to 120 secs.
I have increased the timeout with no effect. From the beginning, I suspected that it was the Barracuda that was causing the problem, but I cannot convince the people running the domain that something is not set up correctly. I run a Barracuda on my end and have no trouble passing traffic between my mail servers, so I know that there is not any compatibility issues between it and Mercury.
The domain that we are having problems sending to is hsccs.org.
[quote user="slavergne"]
I have increased the timeout with no effect. From the beginning, I suspected that it was the Barracuda that was causing the problem, but I cannot convince the people running the domain that something is not set up correctly. I run a Barracuda on my end and have no trouble passing traffic between my mail servers, so I know that there is not any compatibility issues between it and Mercury.
The domain that we are having problems sending to is hsccs.org.
[/quote]
Not sure what to say, I've no problem at all with this domain. Maybe the problem is one of the hosts within the domain. Could you be very specific about what is bouncing, maybe by using a transcript.
In the transcript, lines marked ">>" are the commands sent by this
mail server, while lines marked "<<" represent the replies from the
remote mail server.
Transcript generated by: tstephenson.com
Message date: Tue, 30 Oct 2007 14:44:40 -0700
Message subject: Test Message
Job identifier: MO006122
Delivery started: Tue, 30 Oct 2007 14:45:00 -0700
Delivery ended: Tue, 30 Oct 2007 14:46:35 -0700
----------- Transcript of delivery process follows ----------------------
Established connection to 206.255.247.19
>> MAIL FROM:<support@tstephenson.com> SIZE=1095
<< 250 Ok
>> RCPT TO:<postmaster@hsccs.org>
<< 250 Ok
>> DATA
<< 354 Start mail input; end with <CRLF>.<CRLF>
>> [Sent 1095 bytes of message data]
<< 250 Ok: queued as B42AE282A7
14:46:35: 95 seconds elapsed, closing connection.
-------------------------------------------------------------------------
Have you checked that your sending IP isn't on any blacklist?
Put it into http://openrbl.org/ - there are others like dnsstuff too.
