DDNS connection is broken

[Ceejus]

Having a strange issue here. Every 30 minutes (as set to check), I am getting the following error message in my logs:

watchdog: Hostname/IP mapping error! Restart ddns.

IPv6 updates are off. I have a public IPv4 address but my gut instinct is telling me this has something to do with the address being associated with my vlan4094 WAN interface rather than br0 or eth0. Could DDNS be grabbing my IPv4 address from the wrong interface? Perhaps the router itself is assigning my IPv4 address to the wrong interface?

DDNS service is No-IP. When I log into my No-IP account, I can see that indeed, updates are not being sent. Completely baffled by this.

I have a GT-BE98 Pro with my 2.5G WAN/LAN-1 port in use for the WAN connection. Latest stable Merlin firmware 3006.102.6. One user on here suggested that I should use my 10G WAN/LAN-1 port for WAN instead to get it to stop using vlan4094. I am already using that 10G port for a LAN downlink however. Would like to continue using my 2.5G WAN/LAN-1 port for WAN if possible.

 

[Ceejus]

UPDATE:
No matter what port I use for WAN, watchdog keeps reporting a mapping error and restarting DDNS at the verification check intervals. Despite getting a "Registration is successful" message when re-saving settings under WAN > DDNS, watchdog keeps exhibiting this behavior and the No-IP server isn't being reached at all. At my wits end as to what might be causing this issue.

 

[degrub]

Factory reset and start over manually. Verify latest firmware. Check Beta versions for addressing the issue.
Manually configure. Verify after each change that there are no issues.

 

[Ripshod]

Try this in ssh

brctl show

*edit* Forgot to ask, do you definitely have a public IP or are you behind CG-NAT? Post the first two octets of your WAN IP.

 

[Ceejus]

Try this in ssh
brctl show
*edit* Forgot to ask, do you definitely have a public IP or are you behind CG-NAT? Post the first two octets of your WAN IP.

Yes, no CG-NAT. Public IP is 96.238.x.x. Also just updated to the brand new 3006.102.7 firmware and still experiencing the issue.
brctl show (though not sure what help this output would be for this particular issue):

bridge name bridge id STP enabled interfaces
br0 000a.cc28aa324bb0 yes eth0
eth1
eth3
wl0.0
wl0.1
wl0.2
wl1.0
wl1.1
wl2.0
wl2.1
wl2.2
wl3.0
wl3.1
wl3.2
wl3.4
br58 000a.ba28aa324bb3 yes eth0.4
eth1.4
eth3.4
wl0.3
wl0.4
wl1.4
wl2.4
wl3.3

Hoping to avoid yet another full blown factory reset for what appears to be a small bug. It's way too much time and effort to manually configure all of this from scratch again. My gut is telling me I have something misconfigured somewhere.

 

[degrub]

If your gut is correct, then starting from a known good setup is the most likely way to succeed in identifying the issue. But, we can explore some more.

 

[Ceejus]

I switched to DNS-o-Matic (by signing up with OpenDNS and then using the credentials to log into DNS-o-Matic) and unlike with No-IP, DNS-o-Matic manual updates (simply hitting Apply under DDNS) kick off update process and DNS-o-Matic (and thus, No-IP) are both successfully updated.

However, watchdog continues to report a hostname/IP mapping error at each 30-min. interval. Given that /etc/inadyn.conf reports the correct WAN interface and public IP address (which is matched on both DNS-o-Matic and No-IP), I have to believe at this point that there's a bug in the watchdog check.

If my only option here really is a factory reset, that sucks. I'm gonna have to set up a No-IP Linux DUC on my NAS to attempt to work around this issue as I am not sure how to troubleshoot that watchdog check process.

I checked the logs on my Syslog server and to my surprise, this issue has actually been present for months and I just never noticed it (reason being that I previously had IPv6 checks on which was returning a different error).