Re: [exim] Gmail's new 'suspicious sender' flag

Startseite
Nachricht löschen
Nachricht beantworten
Autor: Ian Eiloart
Datum:  
To: Bill Hayles
CC: <exim-users@exim.org>
Betreff: Re: [exim] Gmail's new 'suspicious sender' flag

On 6 Jul 2011, at 20:48, Bill Hayles wrote:

> Hi, Ian
>
> On Wed, 6 Jul 2011 16:26:03 +0000 in message number <EF12875F-8AA3-4895-B7EE-6F3E02105A69@???>, received here on 06/07/2011 21:27:50, Ian Eiloart <iane@???> said:
>
>>
>> On 6 Jul 2011, at 15:36, Bill Hayles wrote:
>
>>> Which is why my server does NO spam filtering. It's up to the users what to
>>> do about spam; for all I know they may be interested in those offers from
>>> the kind people in the penis enlargement lobby. That's how my users like it.
>>>
>>
>> The trouble with that is that the ideal situation is that spammers simply
>> can't deliver their email. If you check at the MTA, then you can arrange
>> this (false positives and false negatives notwithstanding).
>
> This could lead to tears at bedtime, so I won't pursue it, but one reason
> for my starting my own server was that I don't want any third party deciding
> what mail I should or should not receive. That's entirely up to the end
> user, just as the Post Office don't decide what is and isn't junk mail. My
> users know (and prefer) that it is up to them to deal with spam. As a matter
> of principal I pass through everything.


"Principle". But why not give users the tools to set up spam filters on the server, where smtp time rejection is an option. You're right to leave users in control, but that doesn't mean you have to put mail in their mail boxes.

>> If you leave it to the end user, it's too late to reject the email.
>
> Not necessarily. There are several end-user anti-spam packages, such as
> Mailwasher, which will bounce rejected mail and make it appear that the mail
> was never delivered. However, I think that's a dubious practice as most
> spammers take no notice, it adds to the volume of useless traffic on the net,
> and you could be bombarding somebody unfortunate enough to have an address
> used by the spammers. Therefore I don't do it, and encourage others not to
> either.


Right. Bouncing and rejection are different, where rejection is an option at SMTP time only.

>> As far as the spammer is concerned, the email is delivered.
>
> Do they ever take any notice of bounces?


No, you should not bounce spam because the spammer won't know about it, but SMTP time rejections will be different. A bonnet host won't keep hammering away at a server that consistently refuses its mail. Especially if the sessions are tar-pitted. It's just a waste of time. On the other hand, if you simply accept and deliver the email, then its your resources that are being squandered.

> Given the number of mails
> continually sent to non-existent addresses (which obviously ARE rejected) I
> don't think so. I know I only run a very small server, but the volume of
> mail received by what I term "blunderbuss attacks" far exceeds spam to
> genuine addresses. By blunderbuss what I mean is that the spammer obviously
> has a list of common user names - john@???, peter@??? etc,
> and sends mail in the hope that the address exists. With large mail servers,
> they probably do. For a domain and server with 25 or so accounts, very
> often they don't, but they still arrive despite the bounces.
>
> --
> This is Spain. We do things differently here!
>
> Bill Hayles
> billnot@???
>
>
> --
> ## List details at https://lists.exim.org/mailman/listinfo/exim-users
> ## Exim details at http://www.exim.org/
> ## Please use the Wiki with this list - http://wiki.exim.org/


--
Ian Eiloart
Postmaster, University of Sussex
+44 (0) 1273 87-3148