starred/nsupdate.info-nsupdate-info

Fork 0

mirror of https://github.com/nsupdate-info/nsupdate.info.git synced 2026-04-25 00:25:58 +03:00

[GH-ISSUE #501] "bad IP received" with Speedport Smart 4 #368

New issue

Open

opened 2026-02-26 10:31:07 +03:00 by kerem · 6 comments

kerem commented

2026-02-26 10:31:07 +03:00

Owner

Originally created by @vinjana on GitHub (Nov 3, 2022).
Original GitHub issue: https://github.com/nsupdate-info/nsupdate.info/issues/501

My router is configured as follows:

Provider: other
Host name: myHostName.nsupdate.info
User name: myHostName.nsupdate.info
Password: bliblablub
Updateserver address: ipv4.nsupdate.info
Protocol: https
Port: 443

I used the configuration of the "Speedport Hybrid" router from your website as template, but entered the value from "Domain name" there in "Host name" here.

API Authentication Result Message: 2022-11-03 12:58:30 api authentication success. [hostname: myHostName.nsupdate.info (given in basic auth)]
Client Result Message: 2022-11-03 12:58:30 myHostName.nsupdate.info - received bad ip address: '2003:e7:7ff:1b82:ca99:b2ff:fe09:860d,80.132.212.228'
Server Result Message: // = empty

So, it seems the authentication worked, but then the IP address information was not provided by the router in the format expected by nsupdate.info.

The router logs report

<Domain existiert nicht (notfqdn)>

i.e. the "domain does not exist". But if I enter something else (e.g. just "myHostName") in the "Host name" field, already after saving the configuration the router reports an authentication failure. So I think that using "myHostName.nsupdate.info" as "Host name" value is correct.

I do not know whether the format of the IPs ($ipv6Address,$ipv4Address) provided by the router is the problem (so the router firmware has a bug), or whether the expectation of nsupdate.info server is incorrect (so the server has a bug). If it is the router, please just indicate. I will then try to file a bug report with the router producer.

Note that the Speedport Smart 4 is one of the two systems that work with hybrid LTE for the German Telekom (T-Com), and the device maybe (?) also used in othere countries. My guess is that others have the same problem and that this is rather an important router model.

Originally created by @vinjana on GitHub (Nov 3, 2022). Original GitHub issue: https://github.com/nsupdate-info/nsupdate.info/issues/501 My router is configured as follows: Provider: other Host name: myHostName.nsupdate.info User name: myHostName.nsupdate.info Password: bliblablub Updateserver address: ipv4.nsupdate.info Protocol: https Port: 443 I used the configuration of the "Speedport Hybrid" router from your website as template, but entered the value from "Domain name" there in "Host name" here. API Authentication Result Message: 2022-11-03 12:58:30 api authentication success. [hostname: myHostName.nsupdate.info (given in basic auth)] Client Result Message: 2022-11-03 12:58:30 myHostName.nsupdate.info - received bad ip address: '2003:e7:7ff:1b82:ca99:b2ff:fe09:860d,80.132.212.228' Server Result Message: // = empty So, it seems the authentication worked, but then the IP address information was not provided by the router in the format expected by nsupdate.info. The router logs report ``` <Domain existiert nicht (notfqdn)> ``` i.e. the "domain does not exist". But if I enter something else (e.g. just "myHostName") in the "Host name" field, already after saving the configuration the router reports an authentication failure. So I think that using "myHostName.nsupdate.info" as "Host name" value is correct. I do not know whether the format of the IPs (`$ipv6Address,$ipv4Address`) provided by the router is the problem (so the router firmware has a bug), or whether the expectation of nsupdate.info server is incorrect (so the server has a bug). If it is the router, please just indicate. I will then try to file a bug report with the router producer. Note that the Speedport Smart 4 is one of the two systems that work with hybrid LTE for the German Telekom (T-Com), and the device maybe (?) also used in othere countries. My guess is that others have the same problem and that this is rather an important router model.

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@ThomasWaldmann commented on GitHub (Nov 3, 2022):

I do not know whether the format of the IPs ($ipv6Address,$ipv4Address) provided by the router
is the problem (so the router firmware has a bug)...

Well, that would be my first guess. The original dyndns standard only addressed giving a single IPv4 address.

The nsupdate.info service extends that to giving a single IPv6 address in the same way.

But I did not yet hear about a "standard" giving a comma separated list of ipv4,ipv6, so maybe ask them by which formal or real-world standard they expect that to be working.

@ThomasWaldmann commented on GitHub (Nov 3, 2022): > I do not know whether the format of the IPs ($ipv6Address,$ipv4Address) provided by the router > is the problem (so the router firmware has a bug)... Well, that would be my first guess. The original dyndns standard only addressed giving a single IPv4 address. The nsupdate.info service extends that to giving a single IPv6 address in the same way. But I did not yet hear about a "standard" giving a comma separated list of `ipv4,ipv6`, so maybe ask them by which formal or real-world standard they expect that to be working.

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@vinjana commented on GitHub (Nov 3, 2022):

O.k. Thanks for the info. I'm going to address this with the router producer (if I can somehow reach them).

@vinjana commented on GitHub (Nov 3, 2022): O.k. Thanks for the info. I'm going to address this with the router producer (if I can somehow reach them).

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@strarsis commented on GitHub (Oct 17, 2023):

+1, when the Speedport Smart 4 has an IPv6 and IPv4 address, it will use them both comma-separated as value in myip.
Most dynamic DNS services do not support this format – or ignore the IPv6 part.
It would be great if nsupdate.info also supported this format!

@strarsis commented on GitHub (Oct 17, 2023): +1, when the Speedport Smart 4 has an IPv6 and IPv4 address, it will use them both comma-separated as value in `myip`. Most dynamic DNS services do not support this format – or ignore the IPv6 part. It would be great if nsupdate.info also supported this format!

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@mstock commented on GitHub (Jul 16, 2025):

It seems like ddclient uses this syntax in newer versions, too. The PR which introduced this (ddclient/ddclient#502) links the dyn.com update API documentation which mentions this syntax with the comma separated addresses (see the 'Note:' in the 'Examples' section).

@mstock commented on GitHub (Jul 16, 2025): It seems like `ddclient` uses this syntax in newer versions, too. The PR which introduced this (ddclient/ddclient#502) links the [dyn.com update API documentation](https://help.dyn.com/remote-access-api/perform-update/) which mentions this syntax with the comma separated addresses (see the 'Note:' in the 'Examples' section).

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@ThomasWaldmann commented on GitHub (Jul 16, 2025):

ddclient: sounds like it can use this syntax, but it can also give just a single IP and one would use 2 separate confs for v4 and v6 in that case.

@ThomasWaldmann commented on GitHub (Jul 16, 2025): ddclient: sounds like it **can** use this syntax, but it can also give just a single IP and one would use 2 separate confs for v4 and v6 in that case.

kerem commented

2026-02-26 10:31:08 +03:00

Author

Owner

@strarsis commented on GitHub (Jul 16, 2025):

This is not an ad, as this is a free service that I am not affiliated with,
but IPv64.net works excellently with these problematic SpeedPort router models.
And no need to manually confirm the subscription, which is another huge plus.

@strarsis commented on GitHub (Jul 16, 2025): This is not an ad, as this is a free service that I am not affiliated with, but [IPv64.net](https://ipv64.net/) works excellently with these problematic SpeedPort router models. And no need to manually confirm the subscription, which is another huge plus.