Report instance group 'routing-api' has no jobs definederror in cf-operator logs when helm upgrade to 2.7.1

[aqan213]

Describe the bug
When upgrade kubecf from 2.6.1 to 2.7.1 . I checked the cf-operator logs, found the following error:

# kubectl logs -f cf-operator-quarks-fc47d85dc-s9qnt -n cf-operator
2020-12-30T03:52:16.731Z	DEBUG	controller-runtime.manager.events	recorder/recorder.go:52	Warning	{"object": {"kind":"Secret","namespace":"kubecf","name":"bpm.routing-api-v1","uid":"401c3e4f-368a-46b5-a445-715149c5d356","apiVersion":"v1","resourceVersion":"41747469"}, "reason": "BPMApplyingError", "message": "Failed to apply BPM information: building containers failed for instance group routing-api: instance group 'routing-api' has no jobs defined"}
2020-12-30T03:52:16.731Z	ERROR	controller	controller/controller.go:246	Reconciler error	{"controller": "bpm-controller", "name": "bpm.routing-api-v1", "namespace": "kubecf", "error": "failed to apply BPM information: building containers failed for instance group routing-api: instance group 'routing-api' has no jobs defined"}
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).reconcileHandler
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:246
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).processNextWorkItem
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:218
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).worker
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:197
k8s.io/apimachinery/pkg/util/wait.BackoffUntil.func1
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:155
k8s.io/apimachinery/pkg/util/wait.BackoffUntil
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:156
k8s.io/apimachinery/pkg/util/wait.JitterUntil
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:133
k8s.io/apimachinery/pkg/util/wait.Until
	/go/src/code.cloudfoundry.org/quarks-operator/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:90

It reported instance group 'routing-api' has no jobs defined. Then I checked the desired-manifest-v3, it contains the routing-api instance group without jobs.

# kubectl get secrets -n kubecf desired-manifest-v3 -o json | jq -r '.data."manifest.yaml"' | base64 --decode
..........
..........
- azs: null
  env:
    bosh:
      agent:
        settings: {}
      ipv6:
        enable: false
  instances: 2
  jobs: []
  name: routing-api
  properties:
    quarks: {}
  stemcell: default
  update:
    canaries: 1
    canary_watch_time: 30000-1200000
    max_in_flight: "1"
    serial: false
    update_watch_time: 5000-1200000
  vm_resources: null

As a workaround I added the following lines to remove routing-api instance group

- type: remove
   path: /instance_groups/name=routing-api

What confused me is I have set features.routing_api.enabled to false. Why it still add the routing-api instance group?

To Reproduce
I did not helm install but tried helm upgrade to 2.7.1.

Expected behavior
Since the features.routing_api.enabled was set to false, it should be no routing-api instance group.

Environment

• CF Version
• Component/Plugin Name
• Others

Additional context
Add any other context about the problem here.