Yeah, unfortunately you'll have to write a custom operator or scheduler
until local PVs are available. Another option is to consider running as a
DaemonSet, then you will only have one pod per node and won't need a custom
scheduler.
On Mon, May 8, 2017 at 12:33 AM, vincent gromakowski <
I was thinking of using pod to pod anti affinity and pod to node affinity to
stick pod to nodes depending on local volumes. Should I write a custom
scheduler to map pod to disks and then stick to nodes?
--
You received this message because you are subscribed to the Google Groups
"Kubernetes
We're working on a feature to allow local volumes to be specified as PVs
wth scheduler integration. But it probably won't be in any
production-ready state until at least 1.8.
In the meantime, you might be able to use pod anti-affinity to ensure that
the pods in your statefulset do not end up on
+msau
https://github.com/kubernetes/community/pull/306 is tackling this problem.
Expect an alpha feature that includes scheduling, fault tolerance and
provisioning by v1.8.
On Sun, May 7, 2017 at 12:50 AM, vincent gromakowski <
vincent.gromakow...@gmail.com> wrote:
> Hello
> I would like to
Hello
I would like to have your opinion on statefulset implemented on local volumes.
What would be the best option today ?
- a custom scheduler ?
- a custom scheduler + an operator ?
- an external mapping system ?
I would like to dedicate local disks per pod.
TX
Vg
--
You received this
