Operating system and version (e.g., Ubuntu 24.04):
Ubuntu 26.04 LTS
Web server and version (e.g, Apache 2.4.25):
Nonereplace me
Reverse proxy and version _(e.g. nginx 1.27.2)
None
PHP version (e.g, 8.3):
8.5.4
Is this the first time you’ve seen this error? (Yes / No):
yes
When did this problem seem to first start?
new isntallation
Installation method (e.g. AlO, NCP, Bare Metal/Archive, etc.)
AIO
Are you using CloudfIare, mod_security, or similar? (Yes / No)
NO
Summary of the issue you are facing:
I am working through the official installation instructions for Nextcloud Hub from the github page. I have successfully completed step 3 and am working on step 4. I assume that I did the self-signed certificate, but it feels like I ignored it and went to run the software anyway without a certificate. I hope I did nothing wrong in this step.
Then in step 4 I open the Nextcloud AIO interface successfully.
I don’t quite understand the statement of pointing ports 80 and 8443 to my server. Does this mean my nextcloud servers, using the same IP address as for port 443? And what about the TCP and UDP. Some clarification here would help me a lot.
I am using Adamone software on my Asus router. This among other things gives me a hostname. When I copy this into the domain that will be used for Nextcloud and submit it, I get the following message:
The domain is not reachable on Port 443 from within this container. Have you opened port 443/tcp in your router/firewall? If yes is the problem most likely that the router or firewall forbids local access to your domain. Or in other words: NAT loopback (Hairpinning) does not seem to work in your network. You can work around that by setting up a local DNS server and utilizing Split-Brain-DNS and configuring the daemon.json file of your docker daemon to use the local DNS server.
I have set up port forwarding in the router as follows:
Service_Name External_Port Internal_Port Internal_IP_Address Protocol Source_IP
Nextcloud 443 192.168.1.2 Both
Nextcloud_Talk 3478 192.168.1.2 Both
I also tried getting a domain name from deSEC, using my spectrum-tec.com URL but failed with that as well.
I presume this means my NAT Loopback is not working? Can you give more details on what is required here.
Hi @JakeZwart, welcome! A few clarifications first:
Self-signed certificate (step 3/4): That’s expected. The AIO interface on port 8080 (https://192.168.1.2:8080) always uses a self-signed certificate, and you just need to accept the browser warning. You didn’t do anything wrong there.
Ports 80 and 8443: These are optional. They’re only needed if you want to reach the AIO interface via https://your-domain:8443 with a valid certificate. If you open the AIO interface locally via https://192.168.1.2:8080, you don’t need to forward them at all. The only ports you need to forward to your server (192.168.1.2) are:
443/TCP (Nextcloud): mandatory
443/UDP: optional, only for HTTP/3
3478/TCP and 3478/UDP (Talk/TURN): needed if you want to use Talk
So your forwarding rules with “Both” for 443 and 3478 look correct in principle. Please make sure that both the external and internal port are filled in (443→443, 3478→3478) and that 192.168.1.2 really is the IP of the machine running Docker.
About the error: During domain validation, AIO starts a temporary container (nextcloud-aio-domaincheck) that listens on port 443 on your server. It then tries to connect to your-domain:443 from inside the mastercontainer. The message means that connection failed. NAT loopback is one possible reason, but not the only one. Could you check the following and post the results?
On the server, while you’re on the domain step: sudo docker ps and confirm that nextcloud-aio-domaincheck is running and published on 0.0.0.0:443.
On the server: dig +short your-hostname (or nslookup). Does it return your current public IP? You can compare it with curl -4 ifconfig.me.
Is a firewall active on the Ubuntu server? sudo ufw status. If it is, allow 443/tcp (and 3478 tcp/udp).
From a device outside your network (e.g. your phone with Wi-Fi turned off), open http://your-hostname:443 (http, not https). If you see a short text response, the port forwarding works and the issue is only local access (NAT loopback). If it doesn’t load, the problem is the port forwarding, the firewall, or your ISP blocking/CGNAT.
If step 4 works from outside but validation still fails, your router doesn’t do NAT loopback. Then you have two options:
Regarding deSEC: Could you describe in more detail what you tried and what error you got? AIO’s built-in deSEC integration (the “Don’t have a domain? Get a free one from deSEC” section in the AIO interface) gives you a free *.dedyn.io subdomain. It also automatically enables a local DNS container (dnsmasq) that works around the NAT loopback problem. You then only need to point your router’s DHCP DNS server to your AIO host.
Note: This reply was drafted with the help of AI (Claude) and reviewed by me before posting.
I did not have 443 in the internal and external port. The Asus router said the internal port was optional so I left it blank. That has now been corrected. Likewise the ASUS router says that the Source IP is optional so it is blank as well.
About the error:
Seems fine
nextcloud@nextcloud-Aspire-E5-574:/$ sudo docker ps
[sudo: authenticate] Password:
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
3580511abb0b Package aio-domaincheck · GitHub “/start.sh” 8 minutes ago Up 8 minutes (healthy) 0.0.0.0:443->443/tcp, [::]:443->443/tcp nextcloud-aio-domaincheck
f7a314de57bd Package all-in-one · GitHub “/start.sh” 2 hours ago Up 2 hours (healthy) 0.0.0.0:80->80/tcp, [::]:80->80/tcp, 0.0.0.0:8080->8080/tcp, [::]:8080->8080/tcp, 0.0.0.0:8443->8443/tcp, [::]:8443->8443/tcp, 9000/tcp nextcloud-aio-mastercontainer
nextcloud@nextcloud-Aspire-E5-574:/$
I am working with ADAMnetworks to resolve this issue. So far they seem to think that the IP address changed recently but the problem persists so there might be a problem on their side.
Because of this error I tried the DyDNS method built into AIO. When I ask for a random subdomain slug it gives the error “Unexpected response from deSEC during domain registration (HTTP 400): {“name”:[“Domain registration under dedyn.io is currently suspended.”]}”
If I use a subdomain slug that I would prefer I get the error ““spectrum-tec.dedyn.io” is already taken. Please choose a different subdomain and try again.” This is consistent, no matter subdomain slug I try.
When I log into https://desec.io/ a message comes up that says, “A recent spike with 5-digit numbers of abusive domain registrations required excessive manual handling, which was not sustainable.
.dedyn.io registration is now suspended until further notice. Do NOT email us for exceptions.” I expect that is causing the problem.
Seems fine
142.127.6.26nextcloud@nextcloud-Aspire-E5-574:/$ sudo ufw status
[sudo: authenticate] Password:
Status: inactive
nextcloud@nextcloud-Aspire-E5-574:/$