Description of problem:

there is a short downtime of the RHODS dashboard while following the procedure to register a cluster to Marketplace. I think it triggers the rolling deployment of the RHODS operator pod, so it is scaled down and up. Doing so, the dashboard is unreachable for few minutes (see attached images). 

This may have a probable impact on the running Jupyter labs. I will check the impact on JL and update this issue

Prerequisites (if any, like setup, operators/versions):

RHODS installed

Steps to Reproduce

1. Follow the marketplace cluster registration process  https://marketplace.redhat.com/en-us/workspace/clusters/add/register 
2. notice that the RHODS operator passes from status "succeeded" o "installing" in the Operators > Installed Operators section
3. notice that a new operator pod is instantiated and the old one is terminated
4. try to reach the RHODS dashboard (you may try few times before hit the error)

Actual results:

• RHODS operator re-deployed
• dashboard not reachable: blank page with the text "Client sent an HTTP request to an HTTPS server"

  Expected results:

no impact on RHODS after cluster registration

Reproducibility (Always/Intermittent/Only Once):

always

Build Details:

RHODS v1.1.1-41

Workaround:

Additional info:

• The registration step which triggers the rhods deployment is "Add the Red Hat Marketplace pull secret to the global pull secret on the cluster"
• there is a note in the registration guide saying that "The update pull secret script will perform a rolling update on all your cluster nodes and update the pull secret" and it suggests to perform this operation during off-hours