-
Notifications
You must be signed in to change notification settings - Fork 224
Description
What would you like to be added:
systemd units for running MicroShift on Podman on RHEL in production. This includes creation of the Podman volumes for keeping MicroShift state, least privileges (seccomp et al.), ensuring MicroShift restarts upon failure, etc. We should also suggest an auto-update strategy. That production deployment should use an external CRI-O instance, so MicroShift container restarts/redeployments don't disrupt workloads (however, we may want to also provide an example for the "all-in-one" MicroShift container with CRI-O).
Potentially useful: systemd for Podman, auto-updates, and seccomp for containers.
Why is this needed:
We need a prescriptive way how to do this, so we a) maximise chances users successfully deploy without requiring lots of upfront reading and b) we can provide better test coverage for this scenario.