Different logs output with sigs.k8s.io/controller-runtime v0.12.1

[hatrena]

Currently, with package sigs.k8s.io/controller-runtime v0.11.0 when I run the manager I can see the following dummy logs:

1.656069911530586e+09 INFO controller.GUESTBOOK_CONTROLLER Starting Controller {"reconciler group": "webapp.test.example.com", "reconciler kind": "Guestbook"}

1.6560699115306501e+09 INFO controller.FOO_CONTROLLER Starting EventSource {"reconciler group": "webapp.test.example.com", "reconciler kind": "Foo", "source": "kind source: *v1.Foo"}

1.656069911530703e+09 INFO controller.FOO_CONTROLLER Starting Controller {"reconciler group": "webapp.test.example.com", "reconciler kind": "Foo"}

1.656069911631671e+09 INFO controller.GUESTBOOK_CONTROLLER Starting workers {"reconciler group": "webapp.test.example.com", "reconciler kind": "Guestbook", "worker count": 1}

1.6560699116320908e+09 INFO controller.FOO_CONTROLLER Starting workers {"reconciler group": "webapp.test.example.com", "reconciler kind": "Foo", "worker count": 1}

As you can see each line is corresponding with a specific controller and kind (Guestbook and Foo). so far so good.

---

Now if I upgrade the sigs.k8s.io/controller-runtime package to v0.12.1. All the logs look the same for all controllers and kinds:

1.65607012707837e+09 INFO Starting EventSource {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo", "source": "kind source: *v1.Guestbook"}

1.65607012707843e+09 INFO Starting Controller {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo"}

1.6560701270784159e+09 INFO Starting EventSource {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo", "source": "kind source: *v1.Foo"}

1.6560701270785e+09 INFO Starting Controller {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo"}

1.656070127179373e+09 INFO Starting workers {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo", "worker count": 1}

1.656070127179872e+09 INFO Starting workers {"controller": "FOO_CONTROLLER", "controllerGroup": "webapp.test.example.com", "controllerKind": "Foo", "worker count": 1}

It basically takes info from the last imported controller and uses them for all controllers logs.

Is it supposed to be like this because Kubebuilder is not supporting package sigs.k8s.io/controller-runtime v0.12.1or i am missing something?

Best

[camilamacedo86]

If you upgrade the go mod on your project and then, build a new manager (Operator) with. The new Operator image will be using the new dep added.

Kubebuilder is only a CLI tool to help out you scaffold the projects. Then, if you are looking to see differences in the logs because of https://github.com/kubernetes-sigs/controller-runtime/pull/1827/files you need to check if you do not need to change the default scaffold in the main. go to use it.

[hatrena]

Working with sigs.k8s.io/controller-runtime v0.12.1 is still resulting in the odd logs. Apparently, ths issue is resolved with sigs.k8s.io/controller-runtime v0.12.2.

Thank you

[hatrena]

And i assume this is the related commit that solved the issue: kubernetes-sigs/controller-runtime@19f8d99

[camilamacedo86]

Closing since it is answered.