Cloudron Forum

Fixed in support as outlined in https://forum.cloudron.io/topic/12300/how-to-change-dns-settings-of-my-server/18

@joseph Thanks, that was it! It's all working now.

Thanks, it's working !

The only message was only related to PTR. And there was no other message. Maybe it was a special case or a problem with the request and the question was whether from my server or on the provider's side.

After I clicked sync DNS, I left it for another 24 hours just to be sure, because the proprogation could have been prolonged. After 24 hours there was still a problem , then I did a record analysis and noticed as I wrote above one record A was not updated.

In general, the situation embraced. You can close the topic 😄 only waiting for the improvement of the results from PTR. Generally I have changed but FCrDNS has not been updated.

Looks like that was the fix, now it doesn't change the A and AAAA records of the email server, and Cloudron can still use my External SMTP connection while having the Mail Server Location set to "cloudron.blockbluemedia.com".

i think the unbound DNS worked; seems ok now. thanks!!

I just waited 24 hours and now everything is working 👍

https://my.hussainwh.info works for me. Note that for https://hussainwh.info to work, you have to install an app on that domain. When there is no subdomain , it's called the bare/naked/apex/root domain .

@victorlinares94 GoDaddy has put some new limitations on API access. Please contact their support.

https://www.reddit.com/r/godaddy/comments/1bl0f5r/am_i_the_only_one_who_cant_use_the_api/

" We wanted to inform you that we have recently updated our Domain API requirements. As part of this update, customers are now required to have 50 or more domains in their account to utilize the API. Unfortunately, as you currently only have 1 domain in your account, access to the API is blocked for you. "

See also:

https://www.reddit.com/r/godaddy/comments/1chs1j8/godaddy_access_denied_via_apicall/

That fixed it, the home IP wasn't whitelisted. Thanks for your help

Ah ok. I guess, let us know if it happens again!

@jdaviescoates Says me, since www. tells people to expect the main brand website. naked doesn't tell them anything, eg: SaaS apps often use naked for their login, and www. for the marketing site.

Check out pretty much any big brand you can think of that ought to know, www.google.com etc.

I'm sure I read a deeper article on the whys somewhere, but I was in your shorter is better camp for a long time until I realised there were good reasons for the www. being the destination.

Cookies containment from subdomains and use of CDNs are a couple of reasons I remember off the top of my head. I'm sure there's others.

@girish The terminal command comes back correctly. I don't have time right now to upload a file. But I did upload a test index.html file the other day and it worked fine. I added my app locations to the child nameservers page and I'm not getting the TXT file download anymore. I'm only seeing the characters included in the app browser address if I am in the settings section of the apps. I am still getting issues loading pages if I just type in my regular domain and I get the untrusted certificate warning.

@fisi I think @nebulon responded to your ticket from support. If this is the same support ticket, the reason is that your server has 2 NICs and connected to the internal and external network. This is not a problem but the routes are set up incorrectly in ubuntu. The default route is going via the internal gateway. I think you have to adjust the netplan to fix the routing situation .

(I hope I am replying about the correct tickte 🙂 ! )

@luckow yes. See this comment - https://forum.cloudron.io/post/83868

https://git.cloudron.io/cloudron/box/-/commit/0dfadc59228978f82ae3aa2ba4be2b421e785e02 was the fix . You can always apply it locally in /home/yellowtent/box/src/dns/digitalocean.js . Then, systemctl restart box and renew all certs.

@matix131997 This was a bug in the UI code (setupdns view code). Fixed in https://git.cloudron.io/cloudron/box/-/commit/003bc457bf95a5b555a83d76b36199f3acec8ece

As a workaround:

Setup the DNS as manual or wildcard for a start Then, after you setup admin etc, go to Domains View and change it to Bunny. This view does not have this bug.

it just went away, i think the router had a DNS cache thats flushed now. It's no longer being set as the DNS server via DHCP, so I think its resolved.

Thank you all, I can finally acess my app.

Please update to Cloudron 7.6.4 where it is fixed already. The DigitalOcean API behavior had changed recently.

@sixfold_distract said in Link subdomain on squarespace to Cloudron: questions:

So, scenario 1 is just a simple, case-by-case subdomain like mail.example.com or forum.example.com. Can I even do this via Cloudron?

yes, of course. Just choose the Manual DNS provider. You can just set up normal A/AAAA records for the apps. For the email server, if you go to the Status section, it will tell you what DNS records to set up.