Here is my situation. I have a cluster of 6 VM with all running Fedora 28 server. These VM are dedicated to become a Kubernetes HA cluster, each VM being a node with a specific role (etcd, worker or controler). Until now, the cluster was built with rancher but only a few workloads were deployed, mainly to test and learn. So it is not yet production ready and can thus allow changes.
When upgrading from fedora 27 to 28, I decided to try Atomic on one of my machine. What I saw was very exciting and thus decided to install Atomic on all the other VM. My host provider has Atomic 28 enabled, so installing a bare Atomic 28 is easy and done by the provider. Now, I am looking for a simple and robust way to duplicate the changes I have done on my first machine to the others. What came to mind first is to rsync. But when looking at all the articles about atomic host build server, like these ones [0], [1], [2],[3] I wonder now what is the proper way to go. I am looking for advises about how to achieve my goal. [0]http://www.projectatomic.io/docs/compose-your-own-tree/ <http://www.projectatomic.io/docs/compose-your-own-tree/> [1]https://github.com/trishnaguha/build-atomic-host <https://github.com/trishnaguha/build-atomic-host> [2]https://dustymabe.com/2017/10/05/setting-up-an-atomic-host-build-server/ <https://dustymabe.com/2017/10/05/setting-up-an-atomic-host-build-server/> [3]https://trishnag.wordpress.com/2017/03/27/customize-packages-for-atomic-host-ansible-automation/ <https://trishnag.wordpress.com/2017/03/27/customize-packages-for-atomic-host-ansible-automation/>
signature.asc
Description: OpenPGP digital signature
