[Mailman-Users] Mailman needs 20 seconds to complete per email

[Karsten Becker]

WTF?

> Dec 04 09:25:15 2012 qrunner(8680): connect:   ('localhost', 10030)
> Dec 04 09:25:15 2012 qrunner(8680): connect:   (10030, 'localhost')
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '220 mail01.foodmz.local ESMTP Ecologic Institute ready\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply: retcode (220); Msg: mail01.foodmz.local ESMTP Foo ready
> Dec 04 09:25:36 2012 qrunner(8680): connect:   mail01.foodmz.local ESMTP Foo ready
> Dec 04 09:25:36 2012 qrunner(8680): send:   'ehlo mail01.foodmz.local\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-mail01.foodmz.local\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-PIPELINING\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-SIZE 20971520\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-ETRN\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-STARTTLS\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-AUTH PLAIN LOGIN\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-AUTH=PLAIN LOGIN\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-XFORWARD NAME ADDR PROTO HELO SOURCE PORT\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-ENHANCEDSTATUSCODES\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250-8BITMIME\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply:   '250 DSN\r\n'
> Dec 04 09:25:36 2012 qrunner(8680): reply: retcode (250); Msg: mail01.foodmz.local


What is this f***ing Postfix doing...????

Or is it the mailman searching for localhost all the time? But
"localhost" has an entry in /etc/hosts and /etc/nsswitch.conf looks okay.

Regards
Karsten



On 12/04/2012 07:03 AM, Mark Sapiro wrote:
> Larry Stone wrote:
>>
>> On Dec 3, 2012, at 9:22 PM, Mark Sapiro <mark at msapiro.net> wrote:
>>>
>>> Exactly, but I clain the delay is in the SMTP interaction between
>>> OutgoingRunner and Postfix.
>>
>> Karsten did send the additional log line showing the connect did not happen until 9:53:02. 
> 
> 
> I still want to see the smtplib debug output. That will show exactly
> where the delay is and allow focus on exactly that step in the SMTP
> transaction.
>