ingress router pod in pending state and could not access console

Posted on

Hi,

I have installed a new cluster which consists of 1master node and 2worker Nodes.

OCP 4.13

I have disabled a master node to be schedulable .
oc edit schedulers.config.openshift.io cluster

what i have seen that the router pod is trying to be scheduled on the master node with failure to the master is not schedulable.

i have tried to adjust the router pod to run on two worker nodes without any luck by adding the below parameters .

oc project openshift-ingress-operator

oc edit ingresscontroller.operator default

spec:
NodePlacement:
NodeSelector:
macthLabels:
node-role.kubernetes.io/worker: ""
tolerations:
- effect: NoSchedule
operator: Exists

the oc edit command shows ingress controller edited , but in fact these changes are not persists ( not saved ).

i see these added parameters are gone when i run oc edit command again . how can make the 2* worker nodes are be able to run the router pod ?

also when i make the master node able to schedule the pods , the router pod transitioned from pending state to running state but the two console pods stop working and the shows crashing .

any help would be highly appreciated .

Basem

Responses