Cloudron Forum

Right, the 1GB instance actually has 941MB. I have fixed the setup script accordingly.

ubuntu@ip-172-26-13-174:~$ free -m total used free shared buff/cache available Mem: 941 194 335 0 410 606 Swap: 0 0 0

I remember now. We have this issue with the DNS server (unbound) we use which doesn't work properly with just IPv6 with our configuration. We are working on getting away from unbound altogether because of various other issues, so will get to do this after that.

@susankhanal said in cloudron not accessible using domain name outside the network:

doing curl my.domain.com gives HTML lines from inside the network but doing curl outside the network it says unable to connect

Before debugging further, I would double check the HTML output. Can you check if that looks like Cloudron dashboard code?

Also, is my.domain.com resolving to an internal IP or an external IP?

If it is resolving to an external IP, then most likely port forwarding is not working properly in the router.

@Vortexapps this seems to be VM/deployment related. You have to tell us more about your setup / network for us to help further. nginx is running as you see in the troubleshooting output, so you have to check why it's not reachable by ip on port 443.

@girish Thanks for the quick reply, 👍.

In case anyone else needs this on 22.04, I followed these steps and was able to get the system to boot with full disk encrypted drive with the private key coming from the tpm and no password needing to be typed.

I don't think this uses the same libraries as the new option in 23.10 so I expect that when 24.04 comes out I will do a fresh OS install to use the better supported FDE TPM integration from that version going forward but as a stop gap I am happy with this for now.

Yeah. I know what you mean. Ok I will start afresh 🙂
With this information now it will be ok to configure from scratch.

Many thanks

Jan

Looks like the DNS resolution is failing. Can you explain your environment and check if you can resolve any host from the system via SSH? Is the unbound service working?

One must be more discerning and careful about installing certain driver packages as some come with GUI tweaking utilities which then pull in an entire desktop environment vs only installing the driver component needed.

This was a bug. Fixed in latest package. Thanks for reporting.

@Dr-Inbred Can you try running cloudron-support --troubleshoot and provide it's output here?

@chaitanya it seems there is some networking issue connecting to dockerhub . Can you try pulling down some random docker image?

Fixed now for next release.

Firefox?

@chaddieboy Be safe out there tonight!

Hello I wanted to apologize for the lack of response, as right after I posted this had a family emergency and the holidays and it completely vanished from my mind.

I want to confirm that Port Forwarding was the kicker, as I was following a Doc for Guacamole install instead of Cloudron's full doc for the base install, there were no instructions in the Guac doc to do port forwarding.

Thank you so much!

This has not been tested by us at least and given that 23.04 is not an LTS there is probably no big rush if at all to support it.

@bmann This helped me after updating to 8.0 which broke IPv6 for some reason 🙂 (I'm also on OVHcloud)

@nebulon, after a reboot, I get a connection timeout on the localhost for which my Cloudron resides.

Sorry for the frustrating experience but I have to say in several years of using Cloud infrastructure, this is the first time I hear of someone providing a server on non-port 22. This is why such a special case has not made it to the docs. May I ask what VPS provider this is? We test installation over 30 providers, all use port 22 by default.

@skeats do you need help patching it?