Re: [v6ops] draft-vf-v6ops-ipv6-deployment

[Vasilenko Eduard <[e-mail]>]

Hi Jordi,
IMHO: % of IPv6 traffic from IPv6-capable devices are interesting too.
It could be 90%+ with some IPv4aaS service.

But it is not the replacement of overall IPv6 traffic that is needed too. This number is more important for Carrier.
Eduard
-----Original Message-----
From: v6ops [mailto:[e-mail]] On Behalf Of JORDI PALET MARTINEZ
Sent: Monday, March 29, 2021 12:20 PM
To: [e-mail]
Subject: Re: [v6ops] draft-vf-v6ops-ipv6-deployment

HI Eduard,

Life is not so easy!

Even in a cellular network, you may still have an interesting % of users with old phones, even not "smartphones", they may be only 2.5G, etc. This is something key when you work in countries "less developed". This is one of the first parts of the analysis that we do for every customer, questions such as how many phones support IPv6? How many are iOS vs Android vs Windows (others typically are meaningless)? How much time typically take users to get a "new" phone? Are those "new" phones not really new but 2nd hand (so may be still not supporting IPv6)?, etc., etc.

So clearly if you know, for example, that 20% of your customers have outated or non-IPv6 phones, you will not be able to reach more than 80%.

In the case of residential (wireline) services, some users may have replaced the router, and it may not be IPv6-capable (according to the ISP offering), so they need to keep it dual-stack or just IPv4.

Also, some times (both in cellular and wireline), you don't do the IPv6 turn-on in the complete network in a single stage, you may do it per regions, or depending on users contracts, UE capabilities, etc., etc. It is very common an approach such as: New customers get IPv6-only with IPv4aaS + customers with replaced CPEs (failure of the CPE, moving from DSL to GPON, moving to a new location, etc.); then you plan for new service offerings that require a CPE replacement (example triple play, higher bandwidh, better WiFi coverage, 1Gig ports instead of fast ethernet, etc.) - in this case it may mean "customer needs to pay an small extra fee"; then you may want to enforce the replacement of the rest per regions, etc.

By the way, the 4 big mobile carriers in US are already over the 90%. This is reported by ISOC stats, based on measurements.

When I say overnight, I actually should explain it better, and explain that this % is from the users where you enable IPv6-only with IPv4aaS, otherwise, if you still have a % of users that still have older CPEs or UEs, it can't be true.

And yes, agree ... probably the figure is closer to 90% than 75%, at least in the case of mobile providers in stronger economies, which tend to have a faster "update" rate of smartphones.



El 29/3/21 9:33, "v6ops en nombre de Vasilenko Eduard" <[e-mail] en nombre de [e-mail]> escribió:

    Hi Jordi,
    Your last statement is probably too strong.
    About half of Mobile carriers have 464XLAT, but only a couple of them reached 90%+.
    If one would think logically about this fact then one would conclude that 90% is not easy to reach, not overnight.
    I have seen in some RFC (do not remember the number) that 75% is easy to reach (it was supported by numbers), but it was a few years ago - this data should be better now because we have 20%+ CAGR for Webservers IPv6 support.
    Hence, the truth is probably somewhere between 75% and 90%.
    Eduard
    -----Original Message-----
    From: v6ops [mailto:[e-mail]] On Behalf Of JORDI PALET MARTINEZ
    Sent: Sunday, March 28, 2021 11:20 PM
    To: [e-mail]
    Subject: Re: [v6ops] draft-vf-v6ops-ipv6-deployment

    I can confirm that ... just say it in my previous email ...

    Note also that for residential ISPs, reaching IPv6 levels close to 90% is trivial. It happens overnight, because the big volume of traffic to CDNs such as Netflix, Youtube/Google, Facebook, Akamai, etc., etc. which have already IPv6 on.




    El 28/3/21 21:25, "v6ops en nombre de Lencse Gábor" <[e-mail] en nombre de [e-mail]> escribió:

        However, a statement like:

        "For this reason, when IPv4 traffic is vanishingly small (e.g. less than 1%), it would be better to switch to the IPv6-only stage."

        seems to be trivial.

        Can we state something stronger?

        For example:

        "For this reason, when IPv6 increases to a certain limit (e.g. more than 90%), it would be better to switch to the IPv6-only stage."

        Rationale:
        - Introducing an IPv4aaS technology has its costs, but the selling of 
        the lions share of the public IPv4 addresses brings in more money.
        - The maintenance cost of the IPv4aaS solution is less than that of a 
        complete IPv4 network.

        I do not state that it is true, I just ask, if it can be true. Because 
        if it is so, then it could be a better guidance.

        Gábor

        28/03/2021 20:47 keltezéssel, Brian E Carpenter írta:
        > On 28-Mar-21 21:25, JORDI PALET MARTINEZ wrote:
        >> Yes and not … IPv6 in IPv4 (6in4, proto41, etc.) … 6over4 is another protocol.
        >>
        >>   
        >>
        >> Agree, right thing is to use IPv6-only and IPv4aaS, I was just saying that Free was the initiation of 6RD and they were using that. I’m not saying they still use the same or they should keep using the same.
        > But please consider that if an operator is already supporting its customers using classical dual stack or a solid solution like 6rd, there may be no good reason to change for the next ten years or more. Dual stack has no time limit.
        >
        > I think this statement in the draft:
        > "For this reason, when IPv6 increases to a certain limit,
        > it would be better to switch to the IPv6-only stage."
        > is too vague to be useful. Switching costs might be very high, including loss of customers. In fact, the criterion for switching might be as simple as "when IPv4 traffic is vanishingly small."
        >
        >     Brian
        >
        >
        >
        >>   
        >>
        >> Regards,
        >>
        >> Jordi
        >>
        >> @jordipalet
        >>
        >>   
        >>
        >>   
        >>
        >>   
        >>
        >> El 28/3/21 9:14, "v6ops en nombre de [e-mail] <mailto:[e-mail]>" <[e-mail] <mailto:[e-mail]> en nombre de [e-mail] <mailto:[e-mail]>> escribió:
        >>
        >>   
        >>
        >>   
        >>
        >>   
        >>
        >> 6rd is a mode of IPv6 over IPv4, it is opposite to the concept of "IPv4 as a Service" of IPv6-only, so it should be replaced to make IPv6 as a univeral and underlying network protocol gradually.
        >>
        >>   
        >>
        >> Regards
        >>
        >> Chongfeng
        >>
        >>   
        >>
        >>   
        >>
        >>      *From:* JORDI PALET MARTINEZ <mailto:jordi.palet=[e-mail]>
        >>
        >>      *Date:* 2021-03-25 17:15
        >>
        >>      *To:* [e-mail] <mailto:[e-mail]>
        >>
        >>      *Subject:* Re: [v6ops] draft-vf-v6ops-ipv6-deployment
        >>
        >>      Free was using 6RD initially, not sure if they turned into dual-stack, may be with IPv4 via CGN.
        >>
        >>       
        >>
        >>      Regards,
        >>
        >>      Jordi
        >>
        >>      @jordipalet
        >>
        >>       
        >>
        >>      El 24/3/21 17:23, "v6ops en nombre de Alexandre Petrescu" <[e-mail] en nombre de [e-mail]> escribió:
        >>
        >>       
        >>
        >>       
        >>
        >>       
        >>
        >>          Le 24/03/2021 à 16:59, Gabor LENCSE a écrit :
        >>
        >>          > Dear Alex,
        >>
        >>          >
        >>
        >>          > On 3/24/2021 4:12 PM, Alexandre Petrescu wrote: [...]
        >>
        >>          >> Does IPv6 mandate the use of DNS64 and NAT64?
        >>
        >>          >
        >>
        >>          > Of course, not. :)
        >>
        >>       
        >>
        >>          So I agree with you about that.
        >>
        >>       
        >>
        >>          > There are several IPv4 as a Services solutions exist. We have
        >>
        >>          > covered the five most prominent ones 464XLAT, DS-Lite, MAP-E, MAP-T
        >>
        >>          > and lw4o6 in our I-D:
        >>
        >>          > https://tools.ietf.org/html/draft-lmhp-v6ops-transition-comparison-06
        >>
        >>          > > Your ISP is likely using one of them.
        >>
        >>       
        >>
        >>          For clarification - my ISP is called 'Free' (it has freedom features).
        >>
        >>       
        >>
        >>          They offer me paid IPv4 and IPv6 native access at home on ADSL.  It's
        >>
        >>          one publicly routable IPv4 address and an IPv6 /56 prefix globally
        >>
        >>          routable prefix (a 'GUP' if I can say so, not a GUA'(ddress)).
        >>
        >>       
        >>
        >>          Up to now, looking through the configuration interface of my freebox at
        >>
        >>          home I could not see the options that you mention (464XLAT, DSLITE,
        >>
        >>          MAP-E, MAP-T, lw4o6).  One might say that they are there invisible, but
        >>
        >>          I doubt that, I need a proof of it.  How can I check for presence of
        >>
        >>          options 464XLAT, DSLITE, MAP-E, MAP-T or lw4o6?
        >>
        >>       
        >>
        >>          The problems that appear when I try to browse IPv6 sites that absolutely
        >>
        >>          need IPv4 might be because I turned off the IPv4 stack on my computer's
        >>
        >>          interface (Windows Properties on the Interface, check off IPv4).  This
        >>
        >>          operation (turning off IPv4 in a computer) is possible only on Windows,
        >>
        >>          not on linux, AFAIR.  One cant do 'rmmod ipv4' in linux.
        >>
        >>       
        >>
        >>          That also explains the fact that installing IPv4-IPv6 translation boxes
        >>
        >>          (NAT64, 464LAT, etc.) in a network is not sufficient to access IPv4
        >>
        >>          sites from an IPv6-only computer.
        >>
        >>       
        >>
        >>          In order to access IPv4 sites from IPv6-only computers one also needs
        >>
        >>          the IPv4 stack to work ok on that computer and, moreover, it needs some
        >>
        >>          times software features in the Client that support the 64:: notation of
        >>
        >>          IPv6 addresses.  For example, thunderbird (a very modern MUA) does not
        >>
        >>          understand it and gets confused by it.  It takes it for an fqdn, and
        >>
        >>          does not even try to connect the translation boxes.
        >>
        >>       
        >>
        >>          This means that if one wants to migrate more to IPv6 then one has to
        >>
        >>          think about the NAT64 and 464XLAT concepts more outside of the cellular
        >>
        >>          network concept.
        >>
        >>       
        >>
        >>          And yes, I agree with you, NAT64 and 464XLAT are good tools to
        >>
        >>          migrate.  In particular, if one is on a smartphone or other computer
        >>
        >>          using an OS that cant turn off their IPv4 stacks.
        >>
        >>       
        >>
        >>          Alex
        >>
        >>       
        >>
        >>          >
        >>
        >>          > Best regards,
        >>
        >>          >
        >>
        >>          > Gábor
        >>
        >>          >
        >>
        >>          > _______________________________________________ v6ops mailing list
        >>
        >>          > [e-mail] https://www.ietf.org/mailman/listinfo/v6ops
        >>
        >>       
        >>
        >>          _______________________________________________
        >>
        >>          v6ops mailing list
        >>
        >>          [e-mail]
        >>
        >>          https://www.ietf.org/mailman/listinfo/v6ops
        >>
        >>       
        >>
        >>       
        >>
        >>       
        >>
        >>      **********************************************
        >>
        >>      IPv4 is over
        >>
        >>      Are you ready for the new Internet ?
        >>
        >>      http://www.theipv6company.com
        >>
        >>      The IPv6 Company
        >>
        >>       
        >>
        >>      This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.
        >>
        >>       
        >>
        >>       
        >>
        >>       
        >>
        >>      _______________________________________________
        >>
        >>      v6ops mailing list
        >>
        >>      [e-mail]
        >>
        >>      https://www.ietf.org/mailman/listinfo/v6ops
        >>
        >> _______________________________________________ v6ops mailing list [e-mail] https://www.ietf.org/mailman/listinfo/v6ops
        >>
        >>
        >> **********************************************
        >> IPv4 is over
        >> Are you ready for the new Internet ?
        >> http://www.theipv6company.com
        >> The IPv6 Company
        >>
        >> This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.
        >>
        >>
        >> _______________________________________________
        >> v6ops mailing list
        >> [e-mail]
        >> https://www.ietf.org/mailman/listinfo/v6ops
        >>
        > _______________________________________________
        > v6ops mailing list
        > [e-mail]
        > https://www.ietf.org/mailman/listinfo/v6ops
        >

        _______________________________________________
        v6ops mailing list
        [e-mail]
        https://www.ietf.org/mailman/listinfo/v6ops



    **********************************************
    IPv4 is over
    Are you ready for the new Internet ?
    http://www.theipv6company.com
    The IPv6 Company

    This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.



    _______________________________________________
    v6ops mailing list
    [e-mail]
    https://www.ietf.org/mailman/listinfo/v6ops
    _______________________________________________
    v6ops mailing list
    [e-mail]
    https://www.ietf.org/mailman/listinfo/v6ops



**********************************************
IPv4 is over
Are you ready for the new Internet ?
http://www.theipv6company.com
The IPv6 Company

This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.



_______________________________________________
v6ops mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/v6ops