451 Error: queue file write error (only when using postfix)

[IanF]

Hi Adam

According to IgorG in post 194 above this issue will be fixed in version 9.3, however his post was made last October and I have since then tried a workaround released on December 10th that still failed to resolve the problems. Assuming that the core mechanics are the same I wouldn't hold out much hope for a total fix.

I'll send a personal message to IgorG and see if he can update this thread with more details.

 

[IgorG]

Guys,

I have called to developers right now and they have confirmed that problem with timeout for postfix-queue is really fixed in 9.3.0 version. Now there is no this timeout and if you meet any problems with attaches in Postfix logs it will be other not timeout postfix-queue related problems. It can be internal Postfix problem when Postfix will shut stalled processes or something else.

 

[MichielK]

updated to 9.3.0 but problem still exists ...

Jan 14 17:40:37 klaver before-queue[3672]: Unprocessed command:
Jan 14 17:40:37 klaver before-remote[3674]: errno: Broken pipe
Jan 14 17:40:37 klaver before-remote[3674]: System error: Broken pipe
Jan 14 17:40:37 klaver before-remote[3674]: Unable to write data into outgoing socket
Jan 14 17:40:37 klaver before-remote[3674]: Unable to translate command answer:
Jan 14 17:40:37 klaver postfix/spawn[3673]: warning: command /usr/lib/plesk-9.0/postfix-queue exit status 255


-r-xr-x--- 1 mhandlers-user root 67432 2009-12-30 02:31 /usr/lib/plesk-9.0/postfix-queue

 

[EnricoS]

Error still exists in Plesk 9.3.0 running on Debian Etch !!!

Out: 220 www3.xtra-media.net ESMTP Postfix (Debian/GNU)
In: EHLO mm-notify-out-21102.amazon.com
Out: 250-www3.xtra-media.net
Out: 250-PIPELINING
Out: 250-SIZE 30720000
Out: 250-VRFY
Out: 250-ETRN
Out: 250-STARTTLS
Out: 250-AUTH LOGIN CRAM-MD5 DIGEST-MD5 PLAIN
Out: 250-ENHANCEDSTATUSCODES
Out: 250-8BITMIME
Out: 250 DSN
In: MAIL
FROM:<RealTimeEmail+NotificationEngine-ssn-de-mat-A4IVI1ZTFRCWR@amazon.com>
SIZE=3009
Out: 250 2.1.0 Ok
In: RCPT TO:<[email protected]>
Out: 250 2.1.5 Ok
In: DATA
Out: 354 End data with <CR><LF>.<CR><LF>
Out: 451 4.3.0 Error: queue file write error
In: QUIT
Out: 221 2.0.0 Bye


also on another machine(debian 5.0 / postfix / plesk 9.3.0) (/var/log/mail.err):
Jan 17 22:05:56 mserver1 before-queue[14522]: Processing handlers...
Jan 17 22:05:56 mserver1 before-queue[14522]: errno: Connection reset by peer
Jan 17 22:05:56 mserver1 before-queue[14522]: System error: Connection reset by peer
Jan 17 22:05:56 mserver1 before-queue[14522]: Unable to write data into outgoing socket
Jan 17 22:05:56 mserver1 before-queue[14522]: Unable to translate command answer: >>=20^M

the remote machine gets the following error(/var/log/mail.info):
host mserver1.de.....[xxx.xxx.xxx.xxx] said: 451 4.3.0 Error: queue file write error (in reply to end of DATA command))

please help!!!

 

[gruntly0007]

Please click one of the Quick Reply icons in the posts above to activate Quick Reply.

 

[ignus]

Igor,

Updated to 9.3.0 and we are still seeing the same problem on CentOS 4.

Sample log entry:
Jan 14 05:29:19 server02 before-queue[6054]: Unprocessed command: gif"
Jan 14 05:29:19 server02 before-remote[6055]: Unprocessed command: gif"
Jan 14 05:29:19 server02 before-queue[6054]: errno: Broken pipe
Jan 14 05:29:19 server02 before-queue[6054]: System error: Broken pipe
Jan 14 05:29:19 server02 before-queue[6054]: Unable to write data into incoming socket
Jan 14 05:29:19 server02 before-queue[6054]: Unable to translate command answer: gif" width=3D"329" height=3D"27"></td>
</tr></table> =20^M
Jan 14 05:29:19 server02 postfix/spawn[5202]: warning: command /usr/lib/plesk-9.0/postfix-queue exit status 255

Jan 14 11:52:46 server02 postfix/smtpd[7987]: timeout after DATA (13997 bytes) from unknown[77.46.165.132]
Jan 14 05:52:46 server02 before-remote[8000]: errno: Broken pipe
Jan 14 05:52:46 server02 before-remote[8000]: System error: Broken pipe
Jan 14 05:52:46 server02 before-remote[8000]: Unable to write data to stream
Jan 14 05:52:46 server02 before-remote[8000]: Some error occured
Jan 14 05:52:46 server02 before-queue[7999]: Lost connection
Jan 14 05:52:46 server02 postfix/spawn[5204]: warning: command /usr/lib/plesk-9.0/postfix-queue exit status 255

Jan 14 13:28:07 server02 postfix/smtpd[20970]: connect from unknown[208.44.56.160]
Jan 14 07:28:08 server02 before-remote[20535]: errno: Broken pipe
Jan 14 07:28:08 server02 before-remote[20535]: System error: Broken pipe
Jan 14 07:28:08 server02 before-remote[20535]: Unable to write data to stream
Jan 14 07:28:08 server02 before-remote[20535]: Some error occured
Jan 14 07:28:08 server02 postfix/spawn[13504]: warning: command /usr/lib/plesk-9.0/postfix-queue exit status 255


Thanks,
Ole

 

[IgorG]

Informed developers about 'Unprocessed command' on 9.3.0. Let's wait their answer.

Guys, It is really huge thread with a lot of mixed different Postfix issues! It is Plesk 9.0 forum's section, but you have posted problems with 9.3.0!
I will close this thread if it remains dustbin for all Postfix related problems.

 

[Antonio Sanchez]

Hi guys,

I don't think this thread has to be closed. This is a crossing thread about one issue with several versions from 9.0.

This thread collects a lot of information about the issue and is very helpfull. The background problem is that Parallels knows the issue many months ago and still has no resolve the problem.

We have to talk with our clients saying Parallels is working on it to solve the problem, but upgrade after upgrade the problem remains.

Please I know Igor, that you are only the messenger, but please tell your development team to solve this issues.

Greetings

Antonio Sanchez

 

[Fee]

I have V, 9.3.0 and still get the error

I have V, 9.3.0 and still get the error

 

[Fee]

I heard that bumping the smtp_proxy_timeout = 600s will over work your server.Maybe that was someone quoting on a legacy machine. I think its amazing that I seen this issue stated back to 05 and it is still not resolved. And I agree with Antonio this thread should remain open until a true fix comes out. This thread is about the same issue and people are stating its still happening in different versions of Postfix.

 

[stefanos]

plesk 9.3 on ubuntu 6.06 and the problem still exists

 

[IgorG]

It is still under developer's investigation, if you means 'Unprocessed command' error. Generally, it is Postfix problem but not Plesk. Why you have not switched to Qmail MTA if you have so many problems with Postfix?

 

[Fee]

I tried to switch to QMail and it opened a flood gate to spam.

 

[NateC]

Igor,

Are there any updates to the 'Unprocessed Command' issue on 9.3.0? I just upgraded last night and this is happening for occasional emails. Quite annoying - didn't happen before the upgrade, so almost certainly a Plesk issue and not a Postfix issue.

I do some custom handling after the Plesk work is done (ie - after this error), so qmail is not an option for me.

 

[Susheel Rao]

hi Igorg,
I will be glad to switch to qmail if you address the header issue posted at http://forum.parallels.com/showthread.php?t=98500&highlight=delivered-to. I can use qmail is this problem is addressed.

- Susheel

 

[IgorG]

I can say only that problem under developer's investigation now. We have received a lot of reports about 'Unprocessed command' problem and bug marked with high priority now. I will update this thread with any useful information as soon as I receive it.

 

[Antonio Sanchez]

Have you got any estimated time?

 

[KirkM]

really a problem

Made the mistake of switching from Qmail (which at least was working to the best degree possible within its significant limitations) to Postfix before I was aware of this issue. That error is constant and absolutely a deal-breaker for my clients. Soooo... I went back to Qmail and now it has all kinds of issues that didn't exist before.

What's worse, if you do a yum update, it re-installs Postfix since it sees it as an "update" to the psa mail driver. So now I need to go back and re-install Qmail AGAIN and put the psa mail driver on the exclude list.

When it comes down to it, there really is no real working mail MTA for the Plesk CP at this time. Would it be better to study up on MTA's and Postfix and install it outside of Plesk? Is that even an option?

I would love to hear what others are doing. I can't imagine I am the only business who is really being hurt by this shortcoming in Plesk. Email is HUGE to my clients and, well, everyone.

Thanks in advance for anything you can share to workaround this critical problem.

 

[IgorG]

I have updated posted already Workaround with new version of patch for fixing 'Unprocessed command' error.
Please try it and update this thread with results.

 

[internezzo]

Hi there

Yesterday we updated 4 servers with the patch with the following results so far:
- "Unprocessed command" errors are gone
- No more "signal 11" errors
- Still got a few "broken pipe" errors on one machine
- A lot less "queue file write error" mails but still getting them!!

It seems that the systems run smoother now, but unfortunately Parallels is still far away from a solid Postfix integration in Plesk...

Best regards
Marcel