Here is an image side by side.
i am running the AIO installation on an ubuntu 20.04 server in plain vanilla/updated state, no firewall / router forwading ports 80/443 to the server (10.0.0.11) (i did a test installing apache and i am able to access it from a remote location via the domain name).
I am using a subdomain cloud.mydomain.com which has been pointed to the IP more than 48 hours ago and is resolving perfectly fine to the IP of this installation.
I am able to flick the switch on the firewall router, on and off for ports 443/80 and they take immediate effect. And finally, the test suggested by the nextcloud installer on portchecker.co CONFIRMS, port 443 is open.
I am typing the domain perfectly fine in the input box of the AIO setup page in the screenshot attached.
The ubuntu server is a VM under TrueNAS.
Any ideas where else i could check?
I run the installation TWICE, restoring the server to the initial fresh ubuntu setup.
/etc/hostname and /etc/hosts already setup with the complete subdomain i am using for this server.
Finally, as the last obvious test confirmation, if i shut down the ubuntu server VM where i am running the AIO setup, portchecker.co shows me port 443 is CLOSED. Confirming there was no issue whatsoever to reach the AIO install.
here is another screenshot.
Thank you for the ideas.
I have tried now re-checking that there was not an IP change, and tested using the domain instead of the IP on portchecker.co.
My terminal resolves the subdomain to the right IP, i am copying and pasting, there are no mistakes.
I have installed another server using the same procedure at another location and the whole thing is just running one single command on a fresh installed server, there is no need to configure ip6, individual docker containers or anything, it is all automatic. That other installation went perfectly fine, so i am puzzled here why is till not finding the port, but everything else does,
At this point i am convinced there is a problem with the docker install on github. PERIOD.
I setup a different ubuntu server on a different Virtual enviroment server in the same network, and i had the same issue with port 443 running the AIO docker install.
Then, I cloned the ubuntu VM where i run the docker command for the AIO installation, i booted it, and manually installed apache and the snap certbot application.
I was able to certbot and get the SSL certificate with letsencrypt on the first try, meaning PORT 443 is working fine letting traffic in on the network.
I can access the plain apache server welcome page, with the encrypted padlock using the subdomain that the Nextcloud AIO fails to use on port 443.
I will have to setup nextcloud manually then or deal with the undocumented nextcloud one click install in TrueNAS here…
Hm… Then not sure what is wrong on your truenas instance… In any case you can skip the domain validation with -e SKIP_DOMAIN_VALIDATION=true if you are completely sure that everything is configured correctly.
Thank you szaimen.
I have a TrueNAS server as the host running this VM.
And also, another Proxmox server on a different physical computer on the same network.
The issue is happening on both of them, i do not think is an issue with the VE. And as i mentioned, if i was able to get a cert and open a padlocked apache page on the same VM i am using for AIO, then there should not be issues with AIO.
I will test the extra switch you suggested, and report back
And once the containers got pulled and setup, i got this error message (-e invalid option):
If your server has port 80 and 8443 open and you point a domain to your server, you can get a valid certificate automatially by opening the Nextcloud AIO Interface via:
/usr/bin/start.sh: line 213: exec: -e: invalid option
exec: usage: exec [-cl] [-a name] [command [argument ...]] [redirection ...]
am i supposed to use the switch somewhere else?
Now when i try to access 10.0.0.11:8080 it does not connect.