Problem with Plesk SMTP service

[Andrew_Pa]

Hello to the team.

I realize today that my SMTP doesn't work through a mail client(error 454 4.3.0 Try again later). When I use webmail, everything is fine.

I check the logs and I see this strange behavior.

When I run

/usr/local/psa/bin/service --status smtp
is running

that means that the SMPT service is running.

But When I go to the Services management in the admin Panel the SMTP is not active.

I try to enable it from the panel and also I run

/etc/init.d/postfix restart

With no results.

The problem appear after this :
http://forum.parallels.com/showthread.php?t=263364

Any help would be greatfull

 

[Andrew_Pa]

And from mailog

Sep 27 11:42:25 hotel463 postfix/smtpd[10971]: warning: milter unix:/var/spool/postfix/ctmilter/ctmilter.sock: can't read SMFIC_OPTNEG reply packet header: Connection timed out
Sep 27 11:42:25 hotel463 postfix/smtpd[10971]: warning: milter unix:/var/spool/postfix/ctmilter/ctmilter.sock: read error in initial handshake

 

[Andrew_Pa]

Also I check /var/log/messages and I found

Sep 27 12:12:50 hotel463 postfix/smtpd[14053]: SQL engine 'intentionally disabled' not supported
Sep 27 12:12:50 hotel463 postfix/smtpd[14053]: auxpropfunc error no mechanism available

 

[Andrew_Pa]

Unfortunately I did a server restart. Everything works fine now.

Thank you very much!

 

[Andrew_Pa]

For my bad luck, the problem appears today, with no changes..

Please Is there any solution? Because my server cannot send mail....

 

[Strausmann]

Hello,

i have the same problem. My Plesk Server is a CentOS 5 inkl. Plesk 11 MU 16.
the problem began at noon today with me.
The Micro Updates already reinstalled.

i have see in my postfix logs this error message:

Sep 28 12:40:07 plesk02 postfix-local[3557]: Internal error. Unable to parse e-mail address.
Sep 28 12:40:07 plesk02 postfix-local[3557]: Unable to get sender-domain handlers from the database

Any solutions ?

Best Regards,

Björn

 

[Andrew_Pa]

I gave the same installation as Strausmann mention

Centos 5, 64bit with Plesk 11 MU 16.

 

[IgorG]

http://cafuego.net/2008/09/25/dkim-postfix

 

[Andrew_Pa]

http://cafuego.net/2008/09/25/dkim-postfix

Thank you very much for your answer.

I check my firewall and everything is fine there(Allow from all). I think the problem has something to do with SMTP service in the Panel that cannot be started, while with SSH seems to be running.

After Restart of the server everything was fine till today (after 24 hours, maybe less) that the problem appeared again and the only solution was the Restart again.

The problem appear for first time yesterday.

 

[Andrew_Pa]

After 2 days with no problem, Today the problems appear again.

I check again all the logs and I see that the problem appears because the SMTP service, inside panel, doesn't work.

If I run the commands

/usr/local/psa/admin/sbin/mailmng --stop-service
/usr/local/psa/admin/sbin/mailmng --start-service

The service start for a while, but then It stop again and nobody can send email. If I did a server restart, the problem disappear till its next appearance.

Please can anyone help to solve my problem. It 's very important.

Thanks in advance

 

[Andrew_Pa]

I also check my cpu usage. I see there that "Postifx" service has high cpu usage.
Then I kill the Postfix PID and till now everything is fine.

I think it's the same problem mentioned here
http://forum.parallels.com/showthread.php?t=208784

SO the problem is the the Postfix service stop responding with high cpu usage

 

[Andrew_Pa]

Hello again, Today the same problem.

The process of Postfix has high cpu usage and also I check my mail queue and I see mail that were empty. Empty Sender, Empty Receiver, Empty Header and the date of sending were 1970....

Any suggestions?

 

[JoseViegas]

I'm having this problem with SMTP service being stopped on the control panel and having to constantly stopping and starting it from SSH. Does this thing have a solution?

 

[MichalisZ]

Hello,

we are having the same issue but with different symptoms.
[Centos 6, Plesk 10.4.4 MU47]
Like just Andrew_Pa said at #1, it seems that although the Postfix SMTP is running,

/usr/local/psa/bin/service --status smtp
is running

the Plesk SMTP Service is been reported as down... and even if you click to start it, it does not start.

Our symptoms are the following...
Mail clients can not find the settings ("Autoconfig" eg of thunderbird) of the server and even smtp sending does not seem to work, because our users are reporting problems with sending their email. From the outside world, SMTP does not reply.
On the other hand, just as Andrew_Pa said, mail are being send by the webmail as it should be.

The "strange" thing about this issue is that there is nothing not-ordinary at the logs and it seems that although postifix, psa-pc-remote are running without any problem... something is going terribly wrong.

As for the firewall we have triple-checked and everything is fine. We can connect via telnet to port 25 and get a reply by postfix.
BUT
the main difference to this server and other servers is that, at the others postfix/smtp replies in 1-2 seconds... but in this server it takes up to 20 seconds to replyy, or we even get a connection timeout at the first try.

It seems that something hangs that is been executed before the actual reply of postfix smtp.
[The server does not have any kind of heavy load]

ALso, we have tried switching the MTA to qmail with

/usr/local/psa/admin/sbin/autoinstaller --select-release-current --install-component qmail

and...magically SMTP is working fine!

When we revert back to postfix... we get the same thing... SMTP is up but does not work...

Could this have anything to do with MU47? The problem have started the day MU47 was applied.
Could there is a problem with psa-pc-remote - because the same day it was down (now it is up and running without any kind of issue.)


Sorry for the huge message...but we are trying to get this thing fixed but we have reached an dead-end!
Thank you for your replies!

 

[saatchi]

same problem!!!!

Hi,

I've got the same problem, exactly. Status with shell say it's running but in the panel Postfix is stop and doesn't start. Also I have high CPU usage by Postfix:

3045 postfix 20 0 45292 2936 440 R 100.0(%CPU) 0.0 3073:50 ctmilter.bin

When it seems like this, I restart Postfix, manually:
[root@uitn ~]# /etc/init.d/postfix restart
Shutting down postfix: [ OK ]
Starting postfix: [ OK ]

whit this restart, when I check panel, SMTP server (Postfix) is running. But after about 10 min, it will return and SMTP will be stop working.

My OS is CentOS 6.2 and also my Plesk panel is 11.0.9 Update #29

I couldn't solve it, yet. any idea?

 

[JordiJ]

Hi,

I hope a solution from Parallels to me is exactly the same as the rest of the forum.

Centos 6.2 OS Version 11.0.9 Panel # 31

 

[FabrizioR]

Same problem here

Centos 6.3 -> Plesk 11.0.9 Panel # 31

The error is:

Dec 17 14:22:13 livs postfix/smtpd[1616]: warning: connect to Milter service unix:/var/spool/postfix/ctmilter/ctmilter.sock: No such file or directory

Thanks.

 

[IgorG]

Centos 6.3 -> Plesk 11.0.9 Panel # 31

The error is:

Dec 17 14:22:13 livs postfix/smtpd[1616]: warning: connect to Milter service unix:/var/spool/postfix/ctmilter/ctmilter.sock: No such file or directory

Thanks.

By default /etc/postfix/main.cf should consist following strings:
smtpd_milters = inet:localhost:12768
non_smtpd_milters = inet:localhost:12768

But looks like that your main.cf is modified:
smtpd_milters = inet:localhost:12768 unix var/spool/postfix/ctmilter/ctmilter.sock
non_smtpd_milters = inet:localhost:12768 unix var/spool/postfix/ctmilter/ctmilter.sock

To fix the issue you can try to remove unix var/spool/postfix/ctmilter/ctmilter.sock from /etc/postfix/main.cf and restart postfix. After that all should work fine.

 

[saatchi]

that's it.

thank you very much. I do it and everything is ok, now.

 

[GravuTrad]

Is it possible to reput this file?

How to rebuild milter in plesk to rebuild these kind of file please? (smtp hs)

Thanks