> On 17 Mar 2015, at 08:20, David Hofstee <da...@mailplus.nl> wrote:
> 
> And that is why the big inbox providers should not allow email over IPv6 
> without DMARC. Then domain reputation can be used reliably.

Um, how does the number of available domains compare with the number of 
available IP addresses?

Perhaps we’ll have to look not at the reputation of the domains or the IP 
addresses, but of the organisations that allocate them to end users. And maybe 
there’s space for high reputation TLDs. For example, I know that I trust 
.ac.uk, .edu, and .coop addresses more than .com addresses (although 
compromised accounts are an issue).

Currently we don’t accept email over IPv6, and I don’t see a business case for 
it, either: senders are going to use IPv4 at least as a fallback for the 
foreseeable future. There’s plenty of space in IPv4 for reputable email senders.

> David Hofstee
> 
> Deliverability Management
> MailPlus B.V. Netherlands (ESP)
> 
> -----Oorspronkelijk bericht-----
> Van: mailop [mailto:mailop-boun...@mailop.org] Namens Rob McEwen
> Verzonden: Tuesday, March 17, 2015 4:07 AM
> Aan: mailop@mailop.org
> Onderwerp: Re: [mailop] Help. Why are my emails being marked as spam by 
> google?
> 
> On 3/16/2015 10:00 PM, Franck Martin wrote:
>> You are likely to be in a world of pain, blacklisting/reputation is 
>> likely to be done at a /64 boundary 
>> cfhttps://www.m3aawg.org/sites/maawg/files/news/M3AAWG_Inbound_IPv6_Po
>> licy_Issues-2014-09.pdf Require a /64 for your machine or change 
>> hoster.
> 
> Likewise, spam filtering is going to be LOT harder in the IPv6 world because 
> (1) sender-IP-blacklists are not nearly as effective, and won't ever be, and 
> (2) spammers, ESPs, and marketers of all various levels of gray/black-hat... 
> won't be nearly as motivated to keep their IPs "clean" 
> since it will be so easy to get new ones, thus making it harder for them to 
> pay a long term price for bad behavior. They will literally laugh at their 
> OLD IPs being on blacklists... because they only planned to use those once 
> anyways, and have already moved on to fresh IPv6 ranges!
> 
> IPv6 unilaterally disarms much of our tools in fighting spam and there is no 
> easy or fast fix for that. Even /64s are going to be easier/cheaper to 
> acquire than getting a single IPv4 IP right now. And getting a /48 won't be 
> that hard to get... then the spammer has 65,000 /64s to  burn through, 
> instead of the 256 IPs they have with an IPv4 /24 block. (some still can't 
> get it into their heads... that 65,000 is a MASSIVELY LARGER number than 256) 
> Therefore, it won't matter that blacklists list the whole /64 block. The 
> spammers will laugh at that, too!
> 
> yes, there are strategies to minimize the damage... but not without paying a 
> steep price... and/or a large loss in QOS.
> 
> Anyone who underestimates the value of sender-IP blacklists... would probably 
> change their mind if, for example, SpamHaus's ZEN list (and
> CBL) completely disappeared for just a few hours. I didn't say, "failed to 
> update"... I mean completely taken offline, mirrors and all. 
> Shock-waves would be felt throughout the Internet... everyone you know would 
> notice a big difference. Some large ISPs' mail systems would crash. Now 
> imagine if ALL sender-IP blacklists... disappeared? ...as may be essentially 
> the case in the IPv6 world, for all practical purposes! 
> Many other content-filtering "lines of defense"... that someone might 
> mistakenly think is a viable alternative... would NOT be able to handle the 
> added load because content filtering tools are order of magnitudes more 
> resource hungry than a sender-IP blacklist that blocks at the connection, 
> before accepting the message.
> 
> And I think this example provided in this thread... may very much be a 
> leading indicator of what is to come if IPv6 is too quickly adopted.
> 

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

_______________________________________________
mailop mailing list
mailop@mailop.org
http://chilli.nosignal.org/mailman/listinfo/mailop

Reply via email to