OpenScanHub project on OpenShift currently does not use imagestreams. So only way to trigger a production deployment is to request an admin to delete and recreate pods. This is a manual step and should be automated through a role.
asap
Just to clarify: The request is to add role that will allow app owners to trigger a new deployment in case they want to just restart some specific pod.
Metadata Update from @zlopez: - Issue tagged with: OpenShift, low-gain, medium-trouble
Metadata Update from @zlopez: - Issue priority set to: Waiting on Assignee (was: Needs Review)
Maybe create a dedicated playbook to restart deployment? For example Koschei project has manual playbooks playbooks/manual/upgrade/koschei-rolling.yml and playbooks/manual/upgrade/koschei-full.yml that can be used by appowners to trigger rolling or full updates as needed.
playbooks/manual/upgrade/koschei-rolling.yml
playbooks/manual/upgrade/koschei-full.yml
@mizdebsk That is not a bad idea, but doing that for each OpenShift project is not great. I assume it will be better to allow appowners to have ability to delete/create pods in their own project.
[backlog refinement] This is still something that would be nice to have. Ideally in a way that will allow us to have something general that could be run for multiple different OpenShift projects. Allowing authorized users to rebuild, rollout project if that is needed.
Log in to comment on this ticket.