On Thu, Feb 16, 2017 at 4:22 PM, Ken Dreyer <[email protected]> wrote:

> On Tue, Feb 14, 2017 at 7:39 AM, Aaron Weitekamp <[email protected]>
> wrote:
> > On Mon, Feb 13, 2017 at 6:32 PM, Ken Dreyer <[email protected]> wrote:
> >>
> >> When I have an Atomic Registry running in production, what is the best
> >> way to keep the whole thing up-to-date for security fixes?
> >>
> >> For example, I can use yum-cron to automatically download and install
> >> RPM updates on a traditional system.
> >>
> >> Should I do the following:
> >>
> >>   docker pull openshift/origin-docker-registry
> >>   docker pull openshift/origin
> >>   docker pull cockpit/kubernetes
> >>
> >> ... and then restart the systemd services if any of those has an update?
> >>
> > Yes, that's fine for minor updates. For major upgrades you'll need to run
> > some migration commands[1]. Note: This isn't well-documented or tested
> > outside of OpenShift. A better supported upstream deployment is found
> > here[2].
> >
> > [1]
> > https://docs.openshift.org/latest/install_config/
> upgrading/manual_upgrades.html#updating-policy-definitions
> > [2]
> > https://docs.openshift.org/latest/install_config/install/
> stand_alone_registry.html
>
> Cool, thanks Aaron!
>
> I've written a script to do the pulls+restarts.
> https://github.com/ktdreyer/watch-systemd-containers
>
> Can you help me understand more about what you mean by "better supported"?
>
​By "better supported" I mean there are *a lot* more developers and users
and test automation around the openshift deployment.​

​It's all the same core tech but install/upgrade matters long term.​



> - Ken
>

Reply via email to