[Mailman-Users] Preventing address from being unsubscribed

[Richard Barrett]

At 15:30 03/07/2003, Ed Leafe wrote:
>         I have a list that is pretty high-traffic. I have a separate list 
> that is a sub-set of the big list. To do this, I have an email address 
> that subscribes to the big list, and this address is aliased to a series 
> of python scripts that determine if the message should be passed on to 
> the small list. So far, everything is working great.
>
>         The problem is that some bounces from the small list are being 
> seen as a bounce from this proxy email address, and Mailman disabled the 
> proxy address, cutting off the whole sub-list processing. Is there any 
> way to tell Mailman to *never* remove certain addresses? I.e., to exclude 
> them from bounce processing actions?

If I can paraphrase what I think you have said; some messages sent to 
subscribers of the subordinate list bounce and these bounce messages are 
being returned to the superior list with information indicating they were 
bounced by the subordinate list's email address.

I think you are asking for a means for the superior list to selectively 
ignore those bounce messages.

A more pertinent question would be be why is the superior list should/would 
be seeing those bounce messages.

If the messages your script chooses to pass down to the subordinate list 
are simply posted by it to the subordinate list then any bounce from the 
subordinate list's subscribers should be received by and handled by the 
subordinate list's bounce processing; and MM doesn't normally tell posters 
about bounces resulting from distribution of their posts by the list.

Are you certain that the bounce messages do not actually stem from 
occasional failure of your script to take successful delivery of a message 
from the superior list?

Your MTA logs should show the whether these bounce messages originate 
outside your system and are being passed back to the superior list by some 
means or whether they result from delivery failure to the subordinate list.

>      ___/
>     /
>    __/
>   /
>  ____/
>  Ed Leafe
>  http://leafe.com/
>  http://opentech.leafe.com

------------------------------------------------------------------------------
Richard Barrett                                      http://www.openinfo.co.uk