wdrdres3qew5ts21
commented
3 years ago I confirm this was bug if Hub is down or had been destroy but somehow we didn't detach Managed Cluster it will make Klusterlet stuck forever. In the meanwhile if someone encounter this problem you need go Custom Resource Definition then copy whole of Klusterlet then using oc replace -f original-klusterlet.yaml then it will delete both CRD and Klusterlet object instance.
Tested on Openshift 4.7 and using ACM 2.3
original-klusterlet.yaml
kind: CustomResourceDefinition
apiVersion: apiextensions.k8s.io/v1
metadata:
name: klusterlets.operator.open-cluster-management.io
spec:
group: operator.open-cluster-management.io
names:
plural: klusterlets
singular: klusterlet
kind: Klusterlet
listKind: KlusterletList
scope: Cluster
versions:
- name: v1
served: true
storage: true
schema:
openAPIV3Schema:
description: >-
Klusterlet represents controllers on the managed cluster. When
configured, the Klusterlet requires a secret named of
bootstrap-hub-kubeconfig in the same namespace to allow API requests
to the hub for the registration protocol.
type: object
properties:
apiVersion:
description: >-
APIVersion defines the versioned schema of this representation
of an object. Servers should convert recognized schemas to the
latest internal value, and may reject unrecognized values. More
info:
https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
type: string
kind:
description: >-
Kind is a string value representing the REST resource this
object represents. Servers may infer this from the endpoint the
client submits requests to. Cannot be updated. In CamelCase.
More info:
https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
type: string
metadata:
type: object
spec:
description: >-
Spec represents the desired deployment configuration of
Klusterlet agent.
type: object
properties:
clusterName:
description: >-
ClusterName is the name of the managed cluster to be created
on hub. The Klusterlet agent generates a random name if it
is not set, or discovers the appropriate cluster name on
openshift.
type: string
externalServerURLs:
description: >-
ExternalServerURLs represents the a list of apiserver urls
and ca bundles that is accessible externally If it is set
empty, managed cluster has no externally accessible url that
hub cluster can visit.
type: array
items:
description: >-
ServerURL represents the apiserver url and ca bundle that
is accessible externally
type: object
properties:
caBundle:
description: >-
CABundle is the ca bundle to connect to apiserver of
the managed cluster. System certs are used if it is
not set.
type: string
format: byte
url:
description: >-
URL is the url of apiserver endpoint of the managed
cluster.
type: string
namespace:
description: >-
Namespace is the namespace to deploy the agent. The
namespace must have a prefix of "open-cluster-management-",
and if it is not set, the namespace of
"open-cluster-management-agent" is used to deploy agent.
type: string
registrationImagePullSpec:
description: >-
RegistrationImagePullSpec represents the desired image
configuration of registration agent.
type: string
workImagePullSpec:
description: >-
WorkImagePullSpec represents the desired image configuration
of work agent.
type: string
status:
description: Status represents the current status of Klusterlet agent.
type: object
properties:
conditions:
description: >-
Conditions contain the different condition statuses for this
Klusterlet. Valid condition types are: Applied: components
have been applied in the managed cluster. Available:
components in the managed cluster are available and ready to
serve. Progressing: components in the managed cluster are in
a transitioning state. Degraded: components in the managed
cluster do not match the desired configuration and only
provide degraded service.
type: array
items:
description: StatusCondition contains condition information.
type: object
properties:
lastTransitionTime:
description: >-
LastTransitionTime is the last time the condition
changed from one status to another.
type: string
format: date-time
message:
description: >-
Message is a human-readable message indicating details
about the last status change.
type: string
reason:
description: >-
Reason is a (brief) reason for the condition's last
status change.
type: string
status:
description: >-
Status is the status of the condition. One of True,
False, Unknown.
type: string
type:
description: Type is the type of the cluster condition.
type: string
subresources:
status: {}
conversion:
strategy: None
preserveUnknownFields: true
yylin1
dmaneru
Describe the bug A clear and concise description of what the bug is.
After the execution of clean-clusters.sh, klusterlets.operator.open-cluster-management.io has still existed.
To Reproduce Steps to reproduce the behavior:
Expected behavior A clear and concise description of what you expected to happen. The cluster is successfully imported.
Screenshots If applicable, add screenshots to help explain your problem.
Desktop (please complete the following information):
Additional context Add any other context about the problem here.