« Back to projectatomic.io
Ask Your Question

High Availability Kubernetes Cluster on Atomic

asked 2016-01-13 13:37:58 +0000

baboune gravatar image

updated 2016-01-13 13:39:42 +0000


Which options would be recommended to run a High Available (HA) kubernetes cluster using Atomic?

We are considering two different tracks: 1) the kubernetes documented approach with an external load balancer and an etcd CAS to decide on the node acting as a Kubernetes master 2) A pacemaker based approach (corosync pcs) using a virtual IP where pacemeker is deciding which node is acting as a master

Any ideas? Recommendations? Red Hat supported choices?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2016-01-13 18:59:33 +0000

The documented approach for kuberenetes is a proof of concept. If you look at the doc you'll see:

In the future, we expect to more tightly integrate this lease-locking into the scheduler and controller-manager binaries directly, as described in the high availability design proposal

I don't know that anyone has a recommended, production level HA approach for k8s at this point. Love to see otherwise though.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

[hide preview]


Asked: 2016-01-13 13:37:58 +0000

Seen: 9,193 times

Last updated: Jan 13 '16