Open kozmaz87 opened 6 years ago
I guess this is an OpenShift DNS configuration issue. OpenShift expects its applications to be deployed in the
This sounds like Ansible when it runs configures the hostname expectations of OpenShift wrong.
This is not the wrong subdomain. The "app" instance is used for traffic ingress from the Internet. This traffic will mainly be HTTP requests that do not include the *.app subdomain prepended to the .test domain.
By default, ingress traffic will be *.xxx.xxx.gov.uk. The service can be exported manually or edited to use the domain of your choice. This is not a bug.
It cannot be as that field is greyed out on the console.
I can see that I can manually remove and add route but this does not work out of the box.
When you create an app it automatically creates a route that matches the app name +:
openshift_master_default_subdomain=paas.education.gov.uk
Which is actually incorrect, but this should be education.gov.uk
and not app.paas.education.gov.uk.test
. This should be set to match default traffic ingress from the Internet.
I agree. But that means we need to make this setting correct for our current state and change if we transition DNS later.
What other side effects would changing that variable be? Would it affect master controller aspects? Or would it only serve as base for app deployments?
leaving it open until we have a final decision documented here.
The application's virtual hosting is assuming itself to be on the wrong subdomain. It is missing the 'app' subdomain of the address, which makes a new application deploy/run successful but stay unreachable.