Propagation issue on a new domain/app deployment

[ntnsndr]

I am having an issue with setting up a new app on my Cloudron system. I registered the domain on Namecheap two days ago, and immediately set up the domain in Cloudron using the API method, which has always worked fine. It has been over 24 hours. But when I attempt to install a WordPress site (Managed), some weird things are happening. The installation process gets stuck at Installing - Waiting for propagation of EXAMPLE.org (anonymized). And the logs for the app look like this:

Oct 27 09:43:44 box:dns/waitfordns waitForDns: nameservers are ["dns2.registrar-servers.com","dns1.registrar-servers.com"]
Oct 27 09:43:44 box:dns/waitfordns resolveIp: Checking A for EXAMPLE.org at 156.154.133.200
Oct 27 09:43:44 box:dns/waitfordns isChangeSynced: EXAMPLE.org (A) was resolved to 192.64.119.184,193.46.198.90 at NS dns2.registrar-servers.com (156.154.133.200). Expecting 193.46.198.90. Match false
Oct 27 09:43:44 box:dns/waitfordns resolveIp: Checking A for EXAMPLE.org at 2610:a1:1025::200
Oct 27 09:43:44 box:dns/waitfordns isChangeSynced: EXAMPLE.org (A) was resolved to 192.64.119.184,193.46.198.90 at NS dns2.registrar-servers.com (2610:a1:1025::200). Expecting 193.46.198.90. Match false
Oct 27 09:43:44 box:dns/waitfordns waitForDns: EXAMPLE.org at ns dns2.registrar-servers.com: not done
Oct 27 09:43:44 box:dns/waitfordns Attempt 187 failed. Will retry: ETRYAGAIN

Here's where it gets a little weirder. If I renew the domain certificates across Cloudron while it is stuck there, the website starts working; if I go to EXAMPLE.org, I see a WordPress website. But I can't get the installation process unstuck. And if I try reinstalling, the site disappears until I try the renewal again, which once again does not unstuck the installation process.

I attempted to install Surfer at that domain to see if it was a WordPress issue, but the same problem occurred.

Are there other diagnostics you'd recommend? Thanks in advance for your assistance.

[nebulon]

So the issue seems to be that one nameserver does not propagate the change in time, while the other does. This means, depending on which one your system will hit to resolve the domain, it may or may not point to the correct IP address and thus may or may not show the app for you.

This is why Cloudron checks all associated nameservers for full propagation. This seems to be an issue with Namecheap then. If you have a static IP, maybe you can manually change the A and AAAA (if used) records in the Namecheap UI and see if that propagates correctly.

[ntnsndr]

This is strange. I haven't encountered this issue before.

I do have a static IP. I'm comparing the records to the ones that are working on Namecheap. The A Record IP address is the same, and so is the AAAA Record. So I'm not sure what I could change.

Currently the error I'm getting when I navigate to the website is that there is a self-signed security certificate.

I just tried re-establishing the domain on Cloudron and refreshing the certificates, but we are still stuck at the "Waiting for propagation" stage when installing. In my browser, the website is on the page that says "You are seeing this page because the DNS record of magicmountaintalks.org is set to this server's IP but Cloudron has no app configured for this domain."

Update: After a few minutes of trying the install, it reverts back to the "The certificate is not trusted because it is self-signed" error. And if I refresh the certificates, I can access the installed app, but the Cloudron backend is treating the install as having failed, so I can't do any administrative tasks. That seems to me like more likely a Cloudron issue than Namecheap, since the site is accessible.

Is there anything else you can recommend I try?

Thank you.

[nebulon]

So to be clear, the app install or configure task would ensure a valid certificate, so refreshing certificates would likely already find valid certs for the domain, but would reload the nginx reverse proxy and ensuring its config, resulting in delivering the Cloudron placeholder page for that app in question, since the app task never succeeded.

So first we have to get down to the issue, why not all (both) DNS nameservers from namecheap report the correct IP address.

You can check both via SSH on your Cloudron by running those commands:

host -t A <your app domain> 156.154.132.200       # dns1.registrar-servers.com
host -t A <your app domain> 156.154.133.200       # dns2.registrar-servers.com

Are both returning your server's IP correctly while the app task still failing with your mentioned "waitForDns" logs?

Note that Cloudron's app task will only continue if all nameservers associated to a domain are fully propagated, this may not mean that it wouldn't resolve correctly ocally for you already.

[ntnsndr]

Yes, both commands return the same result:

EXAMPLE.org has address 192.64.119.184
EXAMPLE.org has address 193.46.198.90

The bottom IP address there is the one listed under IPv4 network on the Cloudron backend. I am not sure where the first one is coming from. It appears to be a local network IP.

[ntnsndr]

Okay, found the solution. Somehow there was a URL redirect in Namecheap from EXAMPLE.org to www.EXAMPLE.org. When I removed that, the installation succeeded. Thank you! Sorry for the trouble.