X-Loop headers
Duke
containing the original recipient's email address. So if you sent an email
to me (don't! ;), autoresponder would reply with
X-Loop: du...@bogus.com
Now, my autoresponder checks for the same thing, namely my own X-Loop
header. But anybody replying to my autoresponder does not include my X-Loop
header. Mail clients do not pass it on and this is semi-ok although it
would be nice to include it so they only get the autoresponse once. The
problem is with other autoresponders which would include their X-Loop
header - since I only check for my own I would autorespond and they would do
the same, creating a mailbomb war.
So, what's the proper way of doing this? Should I not autorespond to
messages with _any_ X-Loop header? All examples I've seen check for their
own X-Loop header. This is no good, for the above-stated reasons. Another
solution would be to pass 2 X-Loop headers, the original one plus yours. It
just doesn't smell good.
David F. Skoll
> The
> problem is with other autoresponders which would include their X-Loop
> header - since I only check for my own I would autorespond and they would
> do the same, creating a mailbomb war.
> So, what's the proper way of doing this?
You keep a list of addresses you've responded to, and you don't auto-respond
to an address that's already been responded to in the last week or so.
This problem was solved by the UNIX vacation(1) program about 20 years ago.
--
David.
Duke
news:LEA_9.54275$j5.171814@news...
> You keep a list of addresses you've responded to, and you don't
auto-respond
> to an address that's already been responded to in the last week or so.
> This problem was solved by the UNIX vacation(1) program about 20 years
ago.
What's the advantage of doing this? It probably seemed a good idea 20 years
ago, doesn't (to me) now. Which seems backwards, resources (like the ones
needed to keep database of addresses) were more limited then while more
abundant now.
Why not just silently ignore any message with any X-Loop (among others)
headers? A smart man once said that the simplest solution is usually the
best, might apply here.
P.S. Thanks for the Followup-To, but I crossposted on purpose since I'm
discussing a general mechanism not just headers.
Per Hedeland
writes:
>"David F. Skoll" <d...@roaringpenguin.com> wrote in message
>news:LEA_9.54275$j5.171814@news...
>> You keep a list of addresses you've responded to, and you don't
>auto-respond
>> to an address that's already been responded to in the last week or so.
>> This problem was solved by the UNIX vacation(1) program about 20 years
>ago.
>
>What's the advantage of doing this?
Uh, it works?
> It probably seemed a good idea 20 years
>ago, doesn't (to me) now. Which seems backwards, resources (like the ones
>needed to keep database of addresses) were more limited then while more
>abundant now.
I'm afraid I can't follow your logic there, if any. But it's at least as
good an idea today as 20 years ago.
>Why not just silently ignore any message with any X-Loop (among others)
>headers?
Because X-Loop doesn't have any defined standard semantics - no headers
starting with X- ever have - which is why it was chosen for the procmail
examples you've seen in the first place. In particular, you can't expect
that other people's auto-responses have an X-Loop header.
> A smart man once said that the simplest solution is usually the
>best, might apply here.
Another smart man said that you should make everything as simple as
possible, but not simpler. Do the world a favour and read the vacation
man page for additional measures to take in order to avoid
inappropriately sending auto-responses.
--Per Hedeland
p...@hedeland.org
those who know me have no need of my name
>I'm writing an autoresponder. I am adding my own unique X-Loop: header,
>The problem is with other autoresponders which would include their X-Loop
>header - since I only check for my own I would autorespond and they would
>do the same, creating a mailbomb war.
>So, what's the proper way of doing this? Should I not autorespond to
>messages with _any_ X-Loop header?
i suggest this. what is the use of your auto-responder responding to an
auto-response (or a mailing list)?
--
bringing you boring signatures for 17 years
Villy Kruse
These used to be marked Precedence: Bulk and the good old vacation(1)
program would not respond to those.
Villy
those who know me have no need of my name
>On 01 Feb 2003 20:45:16 GMT,
>>in comp.mail.headers i read:
>>>Should I not autorespond to messages with _any_ X-Loop header?
>>
>>i suggest this. what is the use of your auto-responder responding to an
>>auto-response (or a mailing list)?
>These used to be marked Precedence: Bulk and the good old vacation(1)
>program would not respond to those.
right. x-loop is strong evidence that the message originated within some
sort of automation, and having automation respond to such a message is pure
insanity. (at least until they attain cognitive abilities.)
Duke
wrote in message news:m1wukhz...@usa.net...
> >These used to be marked Precedence: Bulk and the good old vacation(1)
> >program would not respond to those.
>
> right. x-loop is strong evidence that the message originated within some
> sort of automation, and having automation respond to such a message is
pure
> insanity. (at least until they attain cognitive abilities.)
I'm setting/checking Precedence: bulk/junk/list and will additionally ignore
X-Loop: anything.
Thanks
P.S. Got a lot of good pointers from looking over vacation's source.