Critical Description
When recreating a pod with no server groups enabled we get this pod stuck in pending with:
- count: 2518
|
eventTime: null
|
firstTimestamp: "2019-03-05T17:19:07Z"
|
involvedObject:
|
apiVersion: v1
|
kind: Pod
|
name: test-couchbase-44bbw-0005
|
namespace: default
|
resourceVersion: "322365"
|
uid: c8cebed7-3f6a-11e9-bb32-42010a8a0215
|
lastTimestamp: "2019-03-05T17:59:09Z"
|
message: '0/9 nodes are available: 9 node(s) didn''t match node selector.'
|
metadata:
|
creationTimestamp: "2019-03-05T17:19:07Z"
|
name: test-couchbase-44bbw-0005.15891fa6d2f6c350
|
namespace: default
|
resourceVersion: "4034"
|
selfLink: /api/v1/namespaces/default/events/test-couchbase-44bbw-0005.15891fa6d2f6c350
|
uid: c8d11f8b-3f6a-11e9-a58f-42010a8a0207
|
reason: FailedScheduling
|
reportingComponent: ""
|
reportingInstance: ""
|
source:
|
component: default-scheduler
|
type: Warning
|
The node selector itself is:
nodeSelector:
|
failure-domain.beta.kubernetes.io/zone: ""
|
Looks like we need to only set it as and when the associated PVC has the same label.