Hi,
that’s because of the default scheduler config, which includes a
serviceAffinity predicates on the “region” label.
In practice, this rule will make sure you don’t have 2 replicas in the same
region, that’s why you have 2 routes in pending status.
Instead of creating a second DC for the router, I would fix the scheduler
config to fit your needs.
Vincent
> On 30 May 2016, at 16:28, Stéphane Klein <cont...@stephane-klein.info> wrote:
>
> Hi,
>
> I've two region in my cluster:
>
> * region A with 2 nodes
> * region B with 2 nodes
>
> Do I need to deploy one "oadm router ..." by region ?
>
> For the moment I've created only one router deploymentconfig and I've:
>
> * replicas = 4
> * 2 router installed with success on region A
> * 2 router in pending status
>
> This is my error log:
>
> 39m 2m 31 {default-scheduler } Warning FailedScheduling
> pod (router-1-4bz0r) failed to fit in any node
> fit failure on node (prod-node-a-2.example.com): PodFitsPorts
> fit failure on node (prod-node-b-1.example.com): Region
> fit failure on node (prod-node-b-2.example.com): Region
> fit failure on node (prod-node-a-1.example.com): PodFitsPorts
>
> Best regards,
> Stéphane
> _______________________________________________
> users mailing list
> users@lists.openshift.redhat.com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users
_______________________________________________
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users
