Exim: 421 Lost incoming connection

[kris1351]

What causes this error to start? We had a server begin bouncing mails with that error last night and it had been up and running no issues for a year.

 

[kris1351]

Neither fix worked for us. Here is what happens after the mail has been received and then it just sits for awhile it seems before doing this.

2004-04-29 07:49:16 SMTP command timeout on connection from [200.190.149.113]
2004-04-29 07:49:17 SMTP data timeout (message abandoned) on connection from local process
2004-04-29 07:49:17 1BJAtj-0006V2-Fm <[email protected]>: virtual_sa_userdelivery transport output: An error was detected while processing a file of BSMTP input.
2004-04-29 07:49:17 1BJAtj-0006V2-Fm ** [email protected] R=virtual_sa_user T=virtual_sa_userdelivery: Child process of virtual_sa_userdelivery transport returned 2 from command: /usr/sbin/sendmail
2004-04-29 07:49:17 1BJAyb-0006jQ-10 <= <> R=1BJAtj-0006V2-Fm U=mailnull P=local S=21653
2004-04-29 07:49:17 1BJAtj-0006V2-Fm Completed

 

[esp1]

Do you have checked your firewall rules ?
If you use iptables you can check it by typing iptables -L

 

[kris1351]

Done the /etc/eximlocalsend and the iptables directive listed above.

ACCEPT tcp -- anywhere host.domain.com tcp dpt:smtp

This message was created automatically by mail delivery software.

A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed:

[email protected]
local delivery failed

The following text was generated during the delivery attempt:

------ user2domain.com ------

An error was detected while processing a file of BSMTP input.
The error message was:

421 SMTP incoming data timeout - message abandoned

The SMTP transaction started in line 0.
The error was detected in line 3.
0 previous messages were successfully processed.
The rest of the batch was abandoned.
421 SMTP incoming data timeout - message abandoned Transaction started in line 0 Error detected in line 3

 

[princo]

We are still having this problem :

=================
This message was created automatically by mail delivery software.

A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:

[email protected]
local delivery failed

The following text was generated during the delivery attempt:

------ [email protected] ------

An error was detected while processing a file of BSMTP input.
The error message was:

421 SMTP incoming data timeout - message abandoned

The SMTP transaction started in line 0.
The error was detected in line 3.
0 previous messages were successfully processed.
The rest of the batch was abandoned.
421 SMTP incoming data timeout - message abandoned
Transaction started in line 0
Error detected in line 3

=================


What I did was:
/scripts/eximlocalsend
/scripts/exim4
iptables -A INPUT -i lo -s 0.0.0.0/0 -d 0.0.0.0/0 -j ACCEPT

Only solution was to disable SpamAssassin on that domain, but many users need SpamAssassin. Also message bounce only when sent from some providers, not from all, which makes it hard to test (from whichever provider I try it works, but for some other users it bounces back).

Any ideas about this?

 

[kris1351]

We still have this problem on one server, but no one has a real solution. It does it randomly at that so it makes it hard to troubleshoot.

 

[princo]

It was ok for one week, yesterday this started to happen again

 

[kris1351]

In one of the returns i found this:

Child process of virtual_sa_userdelivery transport returned 2 from command: /usr/sbin/sendmail

That happened each time I sent to one account on another one of our servers. It happens only about once a month or so it was hard to track. I ran the following and it seems to have sorted it out.


rm /usr/sbin/sendmail
/scripts/eximlocalsend
/scripts/exim4
iptables -A INPUT -i lo -s 0.0.0.0/0 -d 0.0.0.0/0 -j ACCEPT


I assume the removing of the sendmail link to exim and rebuilding exim is all I needed to do, but we are resolved for tonight at least.

 

[kris1351]

Oh, we are still using SpamAssasin with that fix. All tests have gone through.

 

[princo]

But sendmail is just a link to exim, it's not real sendmail. Also I know many users use sendmail in their CGIs to send mails, so deleting it would cause a problems...

Even if it works it's work around, I'm wondering what is the real reason for this error?

 

[kris1351]

It recreates the link when you run exim4. It also forces it to review the exim config. It made a big difference so far for some reason. I send emails to an account on this same server every single night and they have been getting bounced for the last 3 hours. I ran the steps above after trying /scripts/eximlocalsend and the iptables commands a couple of hours ago and it fixed it. I can send consistently now.

 

[jackal]

Bump same problem, When will cpanel come out with a fix?

9.4 edge 19

 

[Website Rob]

Correct me if I'm wrong, but are the people having this problem also using SpamAssassin? Although I'm using 9.4 E-11, I do not use SA and do not have these problems. I think only one person has mentioned what they are using (jackal with 9.4 edge 19) but no one is mentioning what they use along with whether or not they use SA.

Just trying to determine if it is a SA problem no matter what WHM version is used.

 

[jackal]

Bob, we are using SA and we just got it working.

 

[princo]

We were using latest stable till few days ago, and now Release, and problem still happens.

You're right it only happens if SA is enabled. Once you disable it on certian domain all problems dissapear.

 

[jackal]

We ran the commands that were mentioned in the thread. Still had problems, we then found one the users was sending email to their isp email address. Their isp domain name was not resolvng correctly and it filled up the q. Once their isp got back working our problem went away.

/scripts/eximlocalsend
/scripts/exim4
iptables -A INPUT -i lo -s 0.0.0.0/0 -d 0.0.0.0/0 -j ACCEPT

We also run MailScanner
we cleaned out the quarantine folder, msglog folder, and the input folder. Restarted exim and mailscanner, User got their isp email working. Problem went away at least for now.

 

[princo]

We also run Mailscanner

This seems to correct the problem for some time:

killall -9 spamd
etc/init.d/cpanel restart
spamd -d

But it still leaves the question why it happens in the first place?

 

[kris1351]

Ok, I found something interesting last night. The one account I was consistently getitng the 421 SMTP error on was my wife's email. Anytime I would send the weekly reports to her they would get bounced. Cpanel techs thought it might be because they were in html, but it would do it randomly on the text versions too.

I took a look in her Cpanel and her account is one of those that has been around through 2 years of Cpanel upgrades. I noticed she had the original main accound and then the [email protected] setup like we used to have. I removed the second account since you don't need that anymore and we have been 100% for 24 hours. Would be interested if anyone else that is seeing this problem is sending to an account that was setup similar.

 

[trparky]

I want an answer from CPANEL NOW! We need a solution for this YESTERDAY!

I have several pissed off customers that expect a fix soon!

CPanel, FIX THIS NOW!

 

[myusername]

Why would anyone open a thread, say they fixed the issue, then not post a fix for it?