Livenessprobe ignores stderrthreshold flag

[chmielas]

/kind bug

What happened?
Not able to decrease log level for livenessprobe.
I want to decrease log level for livenessprobe, but it seems that argument stderrthreshold is ignored in DeamonSet Definiton. After definition update I still see logs on info level :

I0131 08:33:42.730485       1 main.go:71] Health check succeeded
I0131 08:33:44.730166       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:44.730497       1 main.go:71] Health check succeeded
I0131 08:33:46.730098       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:46.730404       1 main.go:71] Health check succeeded
I0131 08:33:48.730168       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:48.730508       1 main.go:71] Health check succeeded
I0131 08:33:50.730125       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:50.730467       1 main.go:71] Health check succeeded
I0131 08:33:52.730125       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:52.730462       1 main.go:71] Health check succeeded
I0131 08:33:54.730176       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:54.730499       1 main.go:71] Health check succeeded
I0131 08:33:56.730123       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:56.730428       1 main.go:71] Health check succeeded
I0131 08:33:58.730153       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:33:58.730508       1 main.go:71] Health check succeeded
I0131 08:34:00.730135       1 main.go:53] Sending probe request to CSI driver "efs.csi.aws.com"
I0131 08:34:00.730462       1 main.go:71] Health check succeeded

What you expected to happen?
To be able to decrease log level from info to error or warning

How to reproduce it (as minimally and precisely as possible)?

Update DeamonSet with the following definition

- name: liveness-probe
  imagePullPolicy: Always
  image: quay.io/k8scsi/livenessprobe:v1.1.0
  args:
    - --csi-address=/csi/csi.sock
    - --health-port=9809
    - --logtostderr=true
    - -stderrthreshold=error
  volumeMounts:
    - mountPath: /csi
      name: plugin-dir

Anything else we need to know?:

Environment

• Kubernetes version (use kubectl version):
  1.14.9-eks
• Driver version:
  1.1.0

[leakingtapan]

Have you tired reducing the log verbosity using - --v=4, something like:

- name: liveness-probe
  imagePullPolicy: Always
  image: quay.io/k8scsi/livenessprobe:v1.1.0
  args:
    - --csi-address=/csi/csi.sock
    - --health-port=9809
    - --v=4
  volumeMounts:
    - mountPath: /csi
      name: plugin-dir

[chmielas]

I tried also with verbosity - --v=4 but log level still remains on the same level.

[radityasurya]

+1

Would be good to disable the logs too since it's flooding the logs

[guitarrapc]

There's livenessprobe v2.0.0 released and health check begin/success logs are suppressed om default.

https://github.com/kubernetes-csi/livenessprobe/releases/tag/v2.0.0

if you are using helm, just upgrade sidecar container image version from v1.1.0 to v2.0.0 will address this issue.

original values.yaml:

aws-efs-csi-driver/helm/values.yaml

Line 15 in 51d19a4

tag: "v1.1.0"

# values.yaml
sidecars:
  livenessProbeImage:
    repository: quay.io/k8scsi/livenessprobe
    tag: "v2.0.0" # <- changed!

[chmielas]

Thank you, after upgrade to v2.0.0 issue is fixed.