Bug 1493921

1. Proposed title of this feature request
Perform an integral upgrade of both openshift components and OS

3. What is the nature and description of the request?
We use the automated upgrade procedure as described here https://docs.openshift.com/container-platform/latest/install_config/upgrading/automated_upgrades.html to upgrade our environments. We run the upgrade playbooks directly as described here https://docs.openshift.com/container-platform/latest/install_config/upgrading/automated_upgrades.html#running-the-upgrade-playbook-directly.

We have all masters and some nodes running on atomic, additionally, we (still) have nodes running on RHEL. Our upgrade procedure looks as follows:

. upgrade atomic on atomic nodes to required version (atomic host upgrade/deploy)
. run upgrade playbooks
. if everything fine, upgrade kernel on RHEL nodes

We think, that upgrading atomic as well as upgrading kernel on RHEL nodes should be handled by upgrade playbooks.

4. Why does the customer need this? (List the business requirements here)

We would like to have that for several reasons:

- The upgrade to e.g. version 3.4 requires docker 1.12, so it is clear that atomic needs to be updated.
- It is more efficient if all upgrades are done from "one place", e.g. upgrading OpenShift requires node restart, upgrading kernel requires node restart. Therefore, at the moment, each node is restarted more than once, because OpenShift and kernel are upgraded separately. If the upgrade scripts would handle both, one restart would suffice

5. How would the customer like to achieve this? (List the functional requirements here)
Through an ansible playbook that involve all the above mentioned upgrades.

7. Is there already an existing RFE upstream or in Red Hat bugzilla?
No BZ, although there is a trello card about this general upgrade procedure:
https://trello.com/c/qJap7zzj/30-epic-master-upgrade-improvements-for-ops-36