[Mailman-Users] Outbound smtp email problems: (110, 'Connection timed out')

[Matt England]

Some more info about my system:

MTA = patches qmail 1.03
Mailman = 2.1.6rc4

At 5/15/2005 06:36 PM, Brad Knowles wrote:
>         Occasional timeouts are not too unusual, if you're on a busy 
> server.  We see them on the python.org mail system, for example.

That's note the case with my host.  There's virtually zero load on it at 
the moment.  It's a test server.

Also fyi:  all the inbound email to my list is showing up in my pipermail 
archive.

>>  Outbound email from my /bin/mail command is working ok; I am running a
>>  virtual lists domain, however, not sure if that has anything to do with it.
>
>         You could be having some reverse DNS problems on the virtual domain.

I tried that, or at least according to Dave Sill's "The qmail handbook," 
and not difference in behavior.

I'm using tcpserver with qmail-smtpd, and here's my /etc/tcp.smtp:

root at biz2tek 8:47pm [...vpopmail/etc] 136> cat /etc/tcp.smtp
127.:allow,RELAYCLIENT=""
root at biz2tek 8:47pm [...vpopmail/etc] 137>


>>  While I'm hunting down this bug I was wondering if anyone could shed some
>>  light on it for me.
>
>         From the small amount of information you've provided, it's hard 
> to say anything more.  We'd need more detail, especially including data 
> from the MTA logs to see what it was thinking about those attempts.

There's absolutely zero qmail-smtpd log entries (from what I can tell) 
corresponding to a smtp usage from a localhost process.  I'm grasping for 
air here and I find it rather frustrating.

Is there any way I can isolate the outbound/smtp mailman 
process/program/module?  I've got all my other outbound email mechanisms 
working, but I'm wading through a mountain of modules in mailman trying to 
figure out exactly which part is responsible for the outbound mail.

Still digging...any other thoughts/troubleshooting welcome, particularly 
when trying to isolate the outbound email.

-Matt