Assuming you mean to route proposals.fedoraproject.org to be identical to cfp.fedoraproject.org, here is a proposed patch. Please let me know if that's not what you mean :)

diff --git a/fedoraproject.org.template b/fedoraproject.org.template
index fdfddad9..8067f89d 100644
--- a/fedoraproject.org.template
+++ b/fedoraproject.org.template
@@ -355,6 +355,7 @@ pdc.stg             IN      CNAME  wildcard.stg.fedoraproject.org.
 phab.qa.stg         IN      CNAME  wildcard.stg.fedoraproject.org.
 pkgs01.stg          IN    A     38.145.60.17
 pkgs1.stg           IN    CNAME pkgs01.stg
+proposals           IN      CNAME  router-default.apps.ospo-osci.z3b1.p1.openshiftapps.com.
 qa.stg              IN      CNAME  wildcard.stg.fedoraproject.org.
 rabbitmq.stg        IN      CNAME  wildcard.stg.fedoraproject.org.
 raw-updates.coreos.stg IN      CNAME  wildcard.stg.fedoraproject.org.

Hi, I can help rep this change from the Flock staff team. :wave:

@leo Actually it is a little bit different. We are introducing a new host into the mix. Here is what we want:

1. proposals.fedoraproject.org => router-default.apps.ospo-osci.z3b1.p1.openshiftapps.com
2. cfp.fedoraproject.org => elb.apps.ospo-osci.z3b1.p1.openshiftapps.com

I am not sure what we will do with the old system yet. We may deprecate it at a future date, but there are other Fedora events using the old system, e.g. Creative Freedom Summit. So I do not want it to disappear yet.

@jflory7 My apologies, I misunderstood. Here is the new patch :) We can probably get this merged sometime tomorrow at the latest.

diff --git a/fedoraproject.org.template b/fedoraproject.org.template
index fdfddad9..55c1349d 100644
--- a/fedoraproject.org.template
+++ b/fedoraproject.org.template
@@ -295,7 +295,7 @@ bugz.stg            IN      CNAME  wildcard.stg.fedoraproject.org.
 caiapi.stg          IN      CNAME  wildcard.stg.fedoraproject.org.
 calendar.stg        IN      CNAME  wildcard.stg.fedoraproject.org.
 candidate-registry.stg     IN       CNAME  wildcard.stg.fedoraproject.org.
-cfp                 IN      CNAME  router-default.apps.ospo-osci.z3b1.p1.openshiftapps.com.
+cfp                 IN      CNAME  elb.apps.ospo-osci.z3b1.p1.openshiftapps.com.
 codecs.stg          IN      CNAME  wildcard.stg.fedoraproject.org.
 community.stg       IN      CNAME  wildcard.stg.fedoraproject.org.
 copr.stg 300        IN      CNAME  copr-fe-dev.aws.fedoraproject.org.
@@ -355,6 +355,7 @@ pdc.stg             IN      CNAME  wildcard.stg.fedoraproject.org.
 phab.qa.stg         IN      CNAME  wildcard.stg.fedoraproject.org.
 pkgs01.stg          IN    A     38.145.60.17
 pkgs1.stg           IN    CNAME pkgs01.stg
+proposals           IN      CNAME  router-default.apps.ospo-osci.z3b1.p1.openshiftapps.com.
 qa.stg              IN      CNAME  wildcard.stg.fedoraproject.org.
 rabbitmq.stg        IN      CNAME  wildcard.stg.fedoraproject.org.
 raw-updates.coreos.stg IN      CNAME  wildcard.stg.fedoraproject.org.

This looks good to me. Thanks @leo! If we could get it deployed with time to spare before the announcement publishes on Thursday or Friday, there is a greater chance of DNS propagating correctly.

@misc: I think elb.apps.ospo-osci.z3b1.p1.openshiftapps.com is the right hostname but please advise if this is not the case.

@jflory7 I'll bring it up during the daily standup tomorrow, so it should be deployed then!

Thanks @leo! :pray:

@jflory7 @shedrack-sunday
Hey! The DNS has been updated and propagated (at least on my end). However, we now have a SSL issue. If you go to proposals.fedoraproject.org, the certificate only matches api.ospo-osci.z3b1.p1.openshiftapps.com.
So, we have two options:
1. We can use our proxies to reverse proxy to your domain and correcting SSL (the certificate for proposals.fedoraproject.org would be on our end
2. You can get a certificate yourself and update it so it also matches proposals.fedoraproject.org

Let me know what is easier for you!

@leo I think it would be easiest if we could start over with fresh TLS certificates. Is it possible? I might be naïve to what the trade-offs are between these choices, but I don't have a strong opinion.

@jflory7 Sure! It would just mean you handle all the SSL. Clarifying my earlier statement, option 1 would make us handle SSL, and option 2 would let you do it.

I added the route on the openshift cluster for proposals.fedoraproject.org, so https is working now. However, this broke login, so whoever who set that up should change the domain on the auth system as well. I guess that was @jridky ?

I am able to update it, but it would require the change on FAS SSO as well.

BTW have you checked the new scheduling feature in the old system? It is based on Pretalx and currently used for PerfConf

@jflory7 The issue is fixed right? Can this be closed?

Looks like this is working... please reopen if you need anything else. ;)