Re: V6 still not supported
On 23 Mar 2022, at 1:34 AM, Joe Maimon <jmaimon@jmaimon.com <mailto:jmaimon@jmaimon.com>> wrote: ... Since IPv6 was born of the effort to fix the upcoming address shortage visible at the time and to prevent and alleviate the resulting negative effects, the fact that it did not and that globally v4 address shortage is still a problem is a tally of multiple years of failure.
I noticed that no one on NANOG in the nineties predicted the foot-dragging and whining regarding transition from IPv4 toIPv6. We probably should have done so. I, for one, was busy trying to manage interconnects between divisions with their autonomous ref1918 worlds. I applauded the prospect of global unique addressing. So far the technical process has had rocky moments, but it ongoing failure has not happened. Any failure experienced is largely a failure of management/accounts to invest in the future for something that the media can not turn into sound bites and flashy images. This displays a clear lack of enlightened self interest. Even in my home office, over the last nine years I have observed continually increasing IPv6 access for myself and my clients. Comcast has demonstrated that IPv6 has no deleterious effect on typical user experience. Apple and Microsoft have provided admirable support for IPv6 coexisting with IPv4 on end systems. I suggest that it may be more important to deploy solutions to BufferBloat, to the benefit of both IPv4 and IPv6 since it will improve the user experience, than to try to extend IPv4 lifetime, an effort with diminishing returns.
On 3/23/22 12:21 PM, james.cutler@consultant.com wrote:
I suggest that it may be more important to deploy solutions to BufferBloat, to the benefit of both IPv4 and IPv6 since it will improve the user experience, than to try to extend IPv4 lifetime, an effort with diminishing returns.
Given the tremendous growth of video conferencing which strains the upstream, I wonder how many calls ISP's are getting because the "internet is slow" which is attributable to bufferbloat. Is there really anything that ISP can do if they don't supply the ÇPE? What percentage of providers do supply the CPE in the form of cable and dsl modems, etc, that they could solve the problem with a swap out? Mike
It appears that Michael Thomas <mike@mtcc.com> said:
anything that ISP can do if they don't supply the ÇPE? What percentage of providers do supply the CPE in the form of cable and dsl modems, etc, that they could solve the problem with a swap out?
In the US at least, although cable customers can use their own DOCSIS modem and a router, most don't. All the Comcast customers I know have a combo box that is a modem, router, hotspot, and telephone adapter. It has full native IPv6 support, of course. I poked at one and found that some functions like port forwarding are only managed remotely by logging into the Xfinity website. DSL and fiber modems aren't standard like DOCSIS so you have to use the telco's. Anecdotally, when I've rented apartments in the UK and France they all had provider CPE too.
On 3/23/22 1:08 PM, John Levine wrote:
It appears that Michael Thomas <mike@mtcc.com> said:
anything that ISP can do if they don't supply the ÇPE? What percentage of providers do supply the CPE in the form of cable and dsl modems, etc, that they could solve the problem with a swap out? In the US at least, although cable customers can use their own DOCSIS modem and a router, most don't. All the Comcast customers I know have a combo box that is a modem, router, hotspot, and telephone adapter. It has full native IPv6 support, of course. I poked at one and found that some functions like port forwarding are only managed remotely by logging into the Xfinity website.
DSL and fiber modems aren't standard like DOCSIS so you have to use the telco's.
Anecdotally, when I've rented apartments in the UK and France they all had provider CPE too.
I just googled and it seems that DOCSIS 3.1 mandates shapers, though it allows the MSO to turn it on or off. DSL of course is the worst of the bunch and since there is no cablelabs equivalent is a free for all. My ISP is rolling out fiber so it will be interesting to find out whether they roll out modems that implement shapers. My guess: no. Considering they aren't changing the traffic tiers from DSL, sounds like they're a prime candidate for taking lots of "my internet is slow" calls. Mike
Given the tremendous growth of video conferencing which strains the upstream, I wonder how many calls ISP's are getting because the "internet is slow" which is attributable to bufferbloat. Is there really anything that ISP can do if they don't supply the ÇPE? What percentage of providers do supply the CPE in the form of cable and dsl modems, etc, that they could solve the problem with a swap out?
In my experience this is not really a problem of lack of bandwidth (not to say that this is not important) but of queue behavior (the issue is at root one of 'working latency'). So you can solve this for example with AQM. But solving it on the CPE only moves the bottleneck to the LAN/WLAN (in which case use distributed APs, optimally with Ethernet backhaul to the CPE). Next on the horizon is dual queue - which is in discussion at the IETF (TSVWG). Check out the paper at https://www.netforecast.com/wp-content/uploads/NFR5137-Videoconferencing_Int... and see Figure 8. This suggests the network (WLAN to server) has a budget of 130-280 ms of delay (latency), depending on the video conferencing app. See also my paper about AQM deployment at https://arxiv.org/abs/2107.13968 and the recent BITAG paper on the subject at https://www.bitag.org/documents/BITAG_latency_explained.pdf. Also, if you have Mac OS check out the cool new "responsiveness" tool from Apple: https://mailarchive.ietf.org/arch/msg/113attendees/gfvFljIMgsmCTUUPs9TMeBA2w... Jason
james.cutler@consultant.com wrote:
On 23 Mar 2022, at 1:34 AM, Joe Maimon <jmaimon@jmaimon.com <mailto:jmaimon@jmaimon.com>> wrote: ... Since IPv6 was born of the effort to fix the upcoming address shortage visible at the time and to prevent and alleviate the resulting negative effects, the fact that it did not and that globally v4 address shortage is still a problem is a tally of multiple years of failure.
I noticed that no one on NANOG in the nineties predicted the foot-dragging and whining regarding transition from IPv4 toIPv6.
You must not be looking hard enough. There have been many accounts of how Dual Stack as a transition was steamrollered on over naysayers objections and predictions. My personal correspondence on NANOG doesnt date that far back but here are some of the earliest on-topic examples readily available. https://archive.nanog.org/mailinglist/mailarchives/old_archive/2004-10/msg00... https://archive.nanog.org/mailinglist/mailarchives/old_archive/2004-11/msg00... https://archive.nanog.org/mailinglist/mailarchives/old_archive/2005-07/msg00...
We probably should have done so. I, for one, was busy trying to manage interconnects between divisions with their autonomous ref1918 worlds. I applauded the prospect of global unique addressing. So far the technical process has had rocky moments, but it ongoing failure has not happened.
Depends on the definition of the goal. And your definition appears to be some measure of forward momentum. Certainly plenty exists. And that is good news in and of itself. But the internet needed more and long ago already. And that is the bad news.
Any failure experienced is largely a failure of management/accounts to invest in the future for something that the media can not turn into sound bites and flashy images. This displays a clear lack of enlightened self interest.
When you build something and they dont come on schedule, you can blame them or you can consider what you could or should have done differently, especially as it may advise for the future. If you choose the former, there and then is the evidence of the type of thinking that tends to produce these outcomes.
Even in my home office, over the last nine years I have observed continually increasing IPv6 access for myself and my clients. Comcast has demonstrated that IPv6 has no deleterious effect on typical user experience. Apple and Microsoft have provided admirable support for IPv6 coexisting with IPv4 on end systems. I suggest that it may be more important to deploy solutions to BufferBloat, to the benefit of both IPv4 and IPv6 since it will improve the user experience, than to try to extend IPv4 lifetime, an effort with diminishing returns.
Yeah, I think blaming bufferbloat is another example of designers passing the buck and engaging. If the reality consists of large buffers, start engineering with them in mind. Or better yet, fast large ram cheaply and widely available should be considered a welcome and valuable resource. Try to make proper use of it. As for the rest, false dichotomy again. Do both. Or applaud both. Or support both. Or at least, dont oppose either. Joe
participants (5)
-
james.cutler＠consultant.com -
Joe Maimon -
John Levine -
Livingood, Jason -
Michael Thomas