Bug 1851680

> Actual results:
> The ClusterLogForwarder could be created successfully, and all the logs are
> sent to the CLO managed ES, not the fluentdserver
> 

How does you ClusterLogging CR look like? Do you have a clusterlogging instance with a log store defined?

I use the below file to create the ClusterLogging CR:

apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
metadata:
  name: "instance"
  namespace: "openshift-logging"
spec:
  managementState: "Managed"
  logStore:
    type: "elasticsearch"
    retentionPolicy: 
      application:
        maxAge: 1d
      infra:
        maxAge: 3h
      audit:
        maxAge: 2w
    elasticsearch:
      nodeCount: 3
      redundancyPolicy: "SingleRedundancy"
      resources:
        requests:
          memory: "2Gi"
      storage:
        storageClassName: "standard"
        size: "20Gi"
  visualization:
    type: "kibana"
    kibana:
      replicas: 1
  collection:
    logs:
      type: "fluentd"
      fluentd: {}

@Qiaoling

Adding a log store in CLO typically sets the "default" output type for ClusterLogFowarder. This is expected behavior. The log store needs to be omitted for a test-case "invalid-test-name". In general using ClusterLogFowarder you should use a collection only ClusterLogging CR, e.g.:

apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
metadata:
  name: "instance"
  namespace: "openshift-logging"
spec:
  managementState: "Managed"
  collection:
    logs:
      type: "fluentd"
      fluentd: {}

Got it.

But I think we should prohibit users from specifying an output named `default` when creating ClusterLogForwarder, because `default` means forwarding logs to ClusterLogging managed ES.

We can prohibit users to do this if they provide a logstore. However, in the absence of logstore the ClusterLogFowarder mechanics will prohibit this case. Good that you raised this topic, because I identified an issue in providing errors as this one in the status field. Basically after merging my PR you should get:

Name:         instance
Namespace:    openshift-logging
Labels:       <none>
Annotations:  API Version:  logging.openshift.io/v1
Kind:         ClusterLogForwarder
Metadata:
  Creation Timestamp:  2020-07-01T09:28:55Z
  Generation:          1
  Managed Fields:
    API Version:  logging.openshift.io/v1
    Fields Type:  FieldsV1
    fieldsV1:
      f:metadata:
        f:annotations:
          .:
          f:kubectl.kubernetes.io/last-applied-configuration:
      f:spec:
        .:
        f:outputs:
        f:pipelines:
    Manager:      oc
    Operation:    Update
    Time:         2020-07-01T09:28:55Z
    API Version:  logging.openshift.io/v1
    Fields Type:  FieldsV1
    fieldsV1:
      f:spec:
        f:inputs:
      f:status:
        .:
        f:conditions:
        f:outputs:
          .:
          f:default:
          f:output[0]:
        f:pipelines:
          .:
          f:invalid-name-testing:
    Manager:         cluster-logging-operator
    Operation:       Update
    Time:            2020-07-01T09:28:57Z
  Resource Version:  91658
  Self Link:         /apis/logging.openshift.io/v1/namespaces/openshift-logging/clusterlogforwarders/instance
  UID:               90407cfc-877f-46fc-aa5d-e214034a40b3
Spec:
  Inputs:
    Name:  application
  Outputs:
    Name:  default
    Type:  fluentdForward
    URL:   fluentdserver.openshift-logging.svc:24224
  Pipelines:
    Input Refs:
      application
    Name:  invalid-name-testing
    Output Refs:
      default
Status:
  Conditions:
    Last Transition Time:  2020-07-01T09:28:57Z
    Message:               all pipelines invalid: [invalid-name-testing]
    Reason:                Invalid
    Status:                False
    Type:                  Ready
  Outputs:
    Default:
      Last Transition Time:  2020-07-01T09:28:57Z
      Message:               no default log store specified
      Reason:                MissingResource
      Status:                False
      Type:                  Ready
    output[0]:
      Last Transition Time:  2020-07-01T09:28:57Z
      Message:               output name "default" is reserved
      Reason:                Invalid
      Status:                False
      Type:                  Ready
  Pipelines:
    Invalid - Name - Testing:
      Last Transition Time:  2020-07-01T09:28:57Z
      Message:               invalid: unrecognized outputs: [default], no valid outputs
      Reason:                Invalid
      Status:                False
      Type:                  Ready
Events:                      <none>

Verified with quay.io/openshift/origin-cluster-logging-operator@sha256:544016b9c2e4d768f090490803b5f4003756aa761b5604c4d9d68ede89b1e507

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (OpenShift Container Platform 4.6 GA Images), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2020:4196