> On Jan 14, 2016, at 11:01 AM, Viktor Dukhovni <[email protected]> wrote:
> 
> On Thu, Jan 14, 2016 at 03:10:34PM +0000, Osterweil, Eric wrote:
> 
>>>> DANE Summary
>>>> 16065 DANE enabled zones with TLSA records
>>>> 
>>>> 65 PKIX based Trust Anchor TLSA records (Cert Usage 0)
>>>> 541   PKIX based End Entity TLSA records (Cert Usage 1)
>>>> 266   DANE based Trust Anchor TLSA records (Cert Usage 2)
>>>> 5791  DANE based End Entity TLSA records (Cert Usage 3)
>>> 
>>> 6663
> 
> Ok, so that's 6663 TLSA RRsets, but a much larger number of protected
> zones due to MX indirection.  So I would clearly separate the RRset
> count from the "protected domain" count.

Yep.  The RR counts are listed below the zone count.  I think this is precisely 
what led you to notice.  That is, all of the counts below the zone count are RR 
counts.

> 
>>> 1996  Zones have deployed TLSA for SMTP (Port 25)
> 
> So the missing ~10k "zones" (protected domains) are here, because
> the other ports are rarely (RFC6186 notwithstanding) subject to
> indirection.  

Any of the mail protocols is subject to this indirection, as those DANE records 
are based off the MX record.

> That is you've found 1996 MX hosts with TLSA RRsets?  Or 1996 zones
> with 1 or more MX hosts with TLSA RRsets, or a total of 1996 TLSA
> records for port 25?  I am guessing the latter, because that's what
> makes the "certificate usage" total equal to the "by port" total.

There are 1996 TLSA Resource Records that have the domain name _25._tcp.<domain 
name>.  Each RR at every domain name gets counted separately.  So, if someone 
has two RRs at the same domain name, SecSpider counts two, not one.

> In that case our numbers are similar, I have 10.7k email SMTP
> domains covered by TLSA records of 1564 MX hosts with 2212 TLSA
> RRs (at least, because there are cases where I don't look for any
> TLSA RRs on worse priority MX hosts if a better priority MX hosts
> have no TLSA records).  Of the 10.7k domains 200 have incomplete
> TLSA record coverage in that some MX hosts are not protected, so
> the "domain" is not secured against MITM by attackers who block
> access to the protected MX hosts.

That’s very interesting!

Eric
_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane

Reply via email to