FQDN resolving to stale/incorrect IP — no changes on my end, PBX healthy and serving on correct IP

[pfgtech]

Environment:

• Self-hosted 3CX on AWS Lightsail (Debian 12, dedicated/static public IP)
• FQDN: *******.ks.3cx.us (3CX-managed FQDN)

Problem:
Over the weekend the IP that my 3CX FQDN resolves to changed on its own. I made no changes — no reboots, no stop/start, no manual IP or network edits, no restore. The PBX has been running continuously.

Result: all phones show "No Service," and logging into the management console at the FQDN throws "General network error."

What I've confirmed:

• The instance's actual public IP is unchanged and correct (e.g. 52.21.128.185 — confirmed from the box via curl -4 ifconfig.me, and it matches my Lightsail static IP).
• dig +short ******.ks.3cx.us returns a different, stale IP — an AWS address that is not mine and does not respond.
• Browsing directly to my correct IP loads my 3CX login page normally — so the PBX is healthy and serving on the right address.
• Browsing to the IP the FQDN resolves to returns nothing — that address is dead.
• Box-level health is fine: disk ~10% used, no failed systemd units, all 3CX services running, clock in sync, DNS resolution on the box works, IPv4 default route present and outbound v4 confirmed.

Conclusion: This looks like the 3CX-managed FQDN/DNS record drifted to a stale IP on the 3CX side, while my instance and its public IP never changed.

What I've tried:

• Restarted box
• Restarted the management console and config services (3CXPhoneSystemMC01, 3CXCfgServ01) to force the PBX to re-publish its public IP to the FQDN service — dig still returns the stale IP. [update this line with your actual result before posting]

Questions:

1. How do I force the PBX to re-register/re-publish its current public IP to the 3CX FQDN service so the record corrects itself?
2. Is there a known issue with *.ks.3cx.us FQDN records drifting to old AWS IPs?
3. Is there a manual way to correct the FQDN→IP mapping on the 3CX side without paid support?

I don't have a paid support plan, so any pointers on the self-hosted FQDN update mechanism would be much appreciated.

 

[bitn2]

Did you perhaps start up a second instance with the same license and FQDN? Otherwise, try switching to a dynamic IP, wait a moment, and then switch back to a static IP; that should update the FQDN.

 

[KyriacosS_3CX]

Hello,

We can see that the quoted FQDN resolves fine and we can access it on the correct 3. IP, while the 5.. IP belongs to a different PBX. Please recheck.