IPv6 SEO implecations?

Nicholas Meredith nicholas at udhaonline.net
Mon Mar 28 23:37:47 UTC 2011


I would be getting ipv6 connectivity, adding an unknown AAAA record
such as ipv6 or www6; but not www, and do as many comparative ipv4 vs
ipv6 tracerouts from as many route servers as possible. Then you will
have the data you need to actually make an informed decision rather
than just guessing how it will behave. Remove the temp record and add
a real quad for www only if you liked what you saw.

I assume the name servers are also available over ipv6 including glue?

\n

On 29/03/2011, at 9:25, Wil Schultz <wschultz at bsdboy.com> wrote:

> On Mar 28, 2011, at 3:55 PM, Owen DeLong wrote:
>
>>
>> On Mar 28, 2011, at 3:18 PM, Wil Schultz wrote:
>>
>>> I'm attempting to find out information on the SEO implications of testing ipv6 out.
>>>
>>> A couple of concerns that come to mind are:
>>>
>>> 1) www.domain.com and ipv6.domain.com are serving the exact same content.
>>> Typical SEO standards are to only serve good content from a single domain so information isn't watered down and so that the larger search engines won't penalize. So a big concern is having search results take a hit because content is duplicated through two different domains, even though one domain is ipv4 only and the other is ipv6 only.
>>>
>>> 2) Not running ipv6 natively, or using 6to4.
>>> This (potentially) increases hop count and will put content on a slower GRE tunnel and add some additional time for page load times.
>>>
>>> 3) ??? Any others that I haven't thought of ???
>>>
>>> So basically I'd love to set up some sites for ipv6.domain.com via 6to4 as a phase one, and at some point in the near future implement ipv6 natively inside the datacenter, but I'm somewhat concerned about damaging SEO reputation in the process.
>>>
>>> Thoughts?
>>>
>>> -wil
>>
>> If you're worried about SEO, go with native IPv6 and then deploy AAAAs for WWW.domain.foo.
>>
>> It's been working just fine for www.he.net for years.
>>
>> Owen
>>
>
> So far the consensus is to run dual stack natively.
>
> While this definitely is the way things should be set up in the end, I can see some valid reasons to run ipv4 and ipv6 on separate domains for a while before final configuration. For example, if I'm in an area with poor ipv6 connectivity I'd like to be given the option of explicitly going to an ipv4 site vs the ipv6 version.
>
> I'd also like to not damage SEO in the process though. ;-)
>
> -wil




More information about the NANOG mailing list