My guess is that your IPv6 address changed on the 15 July. I think that you need to log onto your RIPE account and enter the new IPv6 settings under the "Network" tab. ======================================== On 22/01/2022 3:08 pm, Edward Lewis wrote:

Thanks - yes, Cox is my ISP.  But I think all’s well in the ISP IPv6 set up.

I have active IPv6 for other machines in the house. I can ’ssh -6' out of my house using a laptop to a remote VM, over v6.  Checking other devices, a phone on the house wi-fi has an IPv6 address, same /64 prefix as the laptop.  Ditto for a wired desktop.  I didn’t look at everything (that would take all night).

I mention the “wired desktop” as it is “wired" to the same physical ethernet hub as the probe, then wired up to the ISP CPE, etc.

Is it possible that the probe isn’t set up to ask for a v6 address?  It seems to be playing the odd-ball here, all the other devices do IPv6.

...

On Jan 21, 2022, at 20:35, Darin Martin <derwood@naebunny.net> wrote:

Ed, assuming Cox is your ISP this page has info from them regarding IPV6: https://www.cox.com/residential/support/ip-version-6.html

They say that as long as your modem and router are capable, you should get IPV4 and IPV6 addresses at the same time.  Your router, firewall, DHCP and DNS are all a big part of that. Each one of those has to be capable and have it enabled for the Atlas device to get an IPV6 address. If they are, then Cox would be the next one to contact.

On Jan 21, 2022 18:12, Edward Lewis <edlewisjr@cox.net> wrote:

General question…

My probe was doing IPv6 until 15 July 2021 but not it is not.  Is there any way to discover why it is no longer IPv6’ing?

I did change my ISP connection equipment about that time, so something did change on my end.

But IPv6 works for other devices.  When I go to www.ripe.net <http://www.ripe.net/> it tells me I came from an IPv6 address.

The device has been pretty steady - 99.71% the last week, 99.91% for the last month (30days).  It was off-line 29 minutes earlier today (the notice made me think to look), but never off line more than 10 minutes since mid-July.  There’s one stretch of 34d+ of continuous uptime. Run-of-the-mill home connection via a cableTV company.

Is there anyway to ask a probe why it doesn’t have an IPv6 address (via DHCP)?

Ed

note that Probe does not use DHCPv6, therefore if your home network use DHCPv6 [stateful] only, versus RA [stateless] or mixed, you may need to assign IPv6 address manually, as Philip suggested. cheers Jiri

After 3 months, I found time to revisit this...

On Jan 24, 2022, at 03:46, Philip Homburg <philip.homburg@ripe.net> wrote: On 2022/01/22 0:12 , Edward Lewis wrote:

...

General question… My probe was doing IPv6 until 15 July 2021 but not it is not. Is there any way to discover why it is no longer IPv6’ing? If your IPv6 prefix is stable enough, you could manually assign an IPv6 address to the probe and see what happens.

I could I manually assign an IPv6 address to the probe (22382)? What I can’t get my head around is how to directly manage or query [in the sense of managing it, not measurements] the probe. I have an IPv4 address for it, so I can send it packets, but if it has a SSH service, I certainly wouldn’t have a password much less a key. There is no reason I can see keeping the probe from setting up an IPv6 address. On the machines I’ve looked (at least one MacAir and one wired MacMini), IPv6 is apparently fully functional in the home network. I don’t think it is a network environment problem, although, prior to July last year IPv6 worked and I did swap out some CPE equipment at the time IPv6 stopped working. In July I went from a mundane cable modem bridge in front of a NATting WIFI router to a new cable modem which I am running in bridge mode in front of the same NATing WIFI router. This switch did start the IPv6 problems for the probe. What would help is asking the probe what it sees, hence the need for something like management console access. BTW, The SOS History lists AAAA queries but nothing about responses. For 18 April, it shows “no-usb” but for 19 April, that’s cleared. This just confuses me (I did replace the usb stick a few months ago), I can’t tell if the usb is still an issue. I keep staring at the log and it doesn’t seem to tell me anything. This is on https://atlas.ripe.net/probes/22382/#tab-network <https://atlas.ripe.net/probes/22382/#tab-network> PS - As far as the probe’s name "Looks like something Apple would sell”, this comes from a question on a form when I registered the probe that asked for a description to it. And that’s what I thought best described the probe, didn’t realize this would become the name. ;)