On Sat, 2008-10-04 at 17:16 -0700, Lloyd Tennison wrote:
> Based on To: . Because of AOL redacting email addresses from SPAM reports,
> I thought that if a new Envelope-to: (different name) was added, and the
> user name was split into pieces and then added some weird characters in the
> middle, AOL (and others) would not be able to redact. (From an earlier
> thread, Envelope-To would probably not work.)
yeah, Envelope-To is commonly used by delivery agents, so it's a bad
idea to use it for a different purpose.
> Example:
>
> Call the new header X-Ref
>
> X-Ref: exiqwerm-uqwersrs@???
>
> X-Ref: exim-users@???
>
> This would then work for both mailing lists and individual messages. Any
> thoughts on this? Other options also solicited.
"X-Ref" is easy to mistake for "Xref" header as used in Usenet, so I'd
recommend a different name, like X-VIP, but that's just nitpicking.
since the domain is known from where the complaint comes from (e.g.,
aol.com), or apparent from Received headers if the user has set up
forwardning, you don't really need more information than the local part
in the header. AOL can't redact a lone "richard"...
if you do this, Exim will have to send individual copies of every
message to AOL subscribers. of course, if you're using VERP, you're
doing that already.
from a resources perspective (save global CPU and bandwidth! :-), I
think automating the lookup of the Exim queue id from your logs would be
better. by looking at the delivery time as well as the queue id, you
should usually be able to pinpoint the user accurately.
--
regards, | Redpill _
Kjetil T. Homme | Linpro (_)
