Can't change expired domain name of an app
-
Hey all,
I used to use a domain which expired but is still listed in the domain configuration section and is in use by an app. Currently the app does not start for obvious reasons. Also I'm not able to change the domain name of that app. At least it is not possible from the GUI.
Which steps would you propose to migrate an app to a a new domain name in a scenario as described above?
Cheers
-
First change the DNS provider of that domain in the Cloudron dashboard to
no-opand then you should be able to change the domain of that app again. -
Thanks a lot for your quick reply. I tried just that, but the button to save the changes in the app config is still greyed out.
-
@ctrl are you referring to the location/domain change form and the save button there? Can you try to change the subdomain part as well just to see if you hit some form state/validation issue?
@nebulon Yes, I'm talking about the location form. Neither changing the subdomain field nor selecting another valid TLD from the dropdown allows me to save a new location.
I just checked the form on another app. It works as long as an app is tied to a valid domain.
-
@nebulon Yes, I'm talking about the location form. Neither changing the subdomain field nor selecting another valid TLD from the dropdown allows me to save a new location.
I just checked the form on another app. It works as long as an app is tied to a valid domain.
-
@ctrl looks like a bug. Would it be possible for you to allow us to take a direct look by enabling remote SSH support for us and send us a mail to support@cloudron.io with your dashboard domain?
-
To update here, the root case was that nginx was not succeeding to reload, since that old domain had for some reason a custom wildcard certificate configured. That certificate didn't exist anymore on the disk as such so nginx would refuse to reload. That in turn aborted the apptask to change the location before it changes the location, so kind of a deadlock situation there, since Letsencrypt could also not be used, because that domain was not pointing to that Cloudron anymore.
I have fixed it now by clearing the error marker for that app in the database and change the location directly. Will have to see how we can possibly have a better workaround in the future for such edge cases.
-
To update here, the root case was that nginx was not succeeding to reload, since that old domain had for some reason a custom wildcard certificate configured. That certificate didn't exist anymore on the disk as such so nginx would refuse to reload. That in turn aborted the apptask to change the location before it changes the location, so kind of a deadlock situation there, since Letsencrypt could also not be used, because that domain was not pointing to that Cloudron anymore.
I have fixed it now by clearing the error marker for that app in the database and change the location directly. Will have to see how we can possibly have a better workaround in the future for such edge cases.
