Unable to load dashboard - fresh install

[d43m0n1k_sek]

Hello, I just created a fresh install, everything seemed to work as expected up until it finished propagating. I now get timeout, and running troubleshot command returns:

Ubuntu: noble 24.04
Processor: Intel(R) Xeon(R) CPU E5-2637 v3 @ 3.50GHz
BIOS CPU @ 0.0GHz x 4
RAM: 8131976KB
Disk: /dev/mapper/ubuntu--vg-ubuntu--lv 9.1G
[OK] node version is correct
[OK] IPv6 is enabled in kernel. No public IPv6 address
[OK] docker is running
[OK] docker version is correct
[OK] MySQL is running
[OK] nginx is running
[OK] dashboard cert is valid
[OK] dashboard is reachable via loopback
[OK] box v8.3.2 is running
[OK] netplan is good
[OK] DNS is resolving via systemd-resolved
[FAIL] Could not load dashboard domain.
Maybe cloudflare proxying is not working. Delete the domain in Cloudflare dashboard and re-add it. This sometimes re-establishes the proxying

I can't delete the entire domain, it's being used by a bunch of other stuff.

More context:

cloudflare zone token set up as per instructions:

dns a record for "my" shows correct external ip and is set to DNS only.

My cloudflare ssl settings are set to full.
There is a cloudron domain key also in dns records, also set to DNS only, but not sur ehow to verify it.

I do have fail2ban installed and running, and I did have ufw enabled with 22, 80 and 443 allowed, I checked security page after the timeout and saw to not enable it, it has since been disabled. Do I have to start all over because of this? Can I rerun install script or activations without reinstalling whole server?

Other tests:

Running - dig +short my.<my domain>.com does return my public ip.

Running - curl -vk https://my.<my domain>.com

• Host my.<my domain>.com:443 was resolved.
• IPv4: <my external ip>
• Trying <my external ip>:443...
• connect to <my external ip> port 443 from <other internal pc> port 39112 failed: Connection timed out
• Failed to connect to my.<my domain>.com port 443 after 135510 ms: Could not connect to server
• closing connection #0
  curl: (28) Failed to connect to my.<my domain>.com port 443 after 135510 ms: Could not connect to server

I am currently behind double nat - modem doesn't allow bridging -

have 80 and 443 fwd from quantum fiber box to router ip as tcp and udp
have 80 and 443 fwd from router to ubuntu server vm ip as tcp and udp

UPnP is disabled on both boxes.

Can also confirm that my plex server is setup the same way, and it fully accessible outside my network (obviously on it's own ports)

during propagation - I tailed the logs and it got stuck on
"2025-05-08T19:29:19.045Z box:provision setProgress: setup - Done"
before connection timed out, no other errors I can see in those logs.

Please advise next steps, thanks in advance!

[d43m0n1k_sek]

also, I just ran
nc -vz my.<mydomain>.com 443
from termux on my phone, and it got a connection succeeded.

[d43m0n1k_sek]

Also, now getting a ssl error on main domain after changing SSL to full from auto - will look into that as well

Looks like changing the ssl mode on the domain broke the old ssl cert - this website is hosted on a cloud provider, so I'm going to temp try and swap back to auto and get this fixed first

edit* Got this fixed

[d43m0n1k_sek]

This post is deleted!

[d43m0n1k_sek]

Confirming no change -

I installed nothing beforehand this go around, even removed and created a new cloudron instance.

Using Netgear Nighthawk RS500 - says it supports hairpin NAT - my guess would then be that the Quantum Fiber modem box doesn't? This seems to be the case, as if I swap my phone over to lte and go to my domainname I do get the setup admin account page. I will try the workaround mentioned in support page and circle back

[d43m0n1k_sek]

Okay coming back with a final update - I created a second ubuntu vm, installed adguard home, did the setup there, assigned it to be the dns server on the router (with cf being backup), added a DNS rewrite to send LAN to local ip when they attempt to go to my.<domain>.com, and so far in testing it's working.

I think we are back in business. I know this post looks more like a blog, me talking to myself and all, but it worked, so if anyone else is having haripin nat issues (my router fully supports it but my Fiber modem does not!), a little extra VM with plenty of benefits beyond getting cloudron up and running was all it took!

This is resolved.

[scooke]

Amazing journey. Thank you for documenting it!

[BrutalBirdie]

@d43m0n1k_sek said in Unable to load dashboard - fresh install:

I know this post looks more like a blog

And that is totally fine, even more, it is very much welcome.
Sharing your experience and walkthrough while having the problem is very beneficial to other users.
So please, keep going!
I do the same e.g. https://forum.cloudron.io/post/43733