Container metrics not showing after upgrade to Kubernetes 1.24

[fgiannetti]

Hi all!
Im using Netdata on a Kubernetes cluster and it was working very well, showing node data and all container metrics.
After upgrading Kuberetes to version 1.24 and migrate from Docker to containerd as container runtime, no container is shown aymore.
Is there a config to set or some way to get back the container metrics?

This is my values.yaml file using on Helm install:

image:
  repository: netdata/netdata
  tag: v1.34.1
  pullPolicy: IfNotPresent

initContainersImage:
  repository: alpine
  tag: latest
  pullPolicy: IfNotPresent

service:
  type: LoadBalancer
  port: 80

ingress:
  enabled: false

parent:
  port: 19999
  
  database:
    persistence: true
    storageclass: "gp3"
    volumesize: 16Gi

  alarms:
    persistence: true
    storageclass: "gp3"
    volumesize: 1Gi

  configs:
    netdata:
      enabled: true
      path: /etc/netdata/netdata.conf
      data: |
        [global]
          memory mode = dbengine
          dbengine multihost disk space = 4000
          cleanup obsolete charts after seconds = 300
          delete obsolete charts files = yes

        [plugins]
          cgroups = yes
          tc = no
          enable running new plugins = no
          check for new plugins every = 72000
          python.d = no
          charts.d = no
          go.d = no
          node.d = no
          apps = no
          proc = no
          idlejitter = no
          diskspace = no
          ebpf = no
    stream:
      enabled: true
      path: /etc/netdata/stream.conf
      data: |
        [684C1DEB-A0FC-441F-8803-F497B0469702]
          enabled = yes
          history = 3600
          default memory mode = dbengine
          health enabled by default = auto
          allow from = *
    health:
      enabled: true
      path: /etc/netdata/health_alarm_notify.conf
      data: |
        SEND_EMAIL="NO"
        SEND_SLACK="NO"
        SLACK_WEBHOOK_URL=""
        DEFAULT_RECIPIENT_SLACK=""
        role_recipients_slack[sysadmin]="${DEFAULT_RECIPIENT_SLACK}"
        role_recipients_slack[domainadmin]="${DEFAULT_RECIPIENT_SLACK}"
        role_recipients_slack[dba]="${DEFAULT_RECIPIENT_SLACK}"
        role_recipients_slack[webmaster]="${DEFAULT_RECIPIENT_SLACK}"
        role_recipients_slack[proxyadmin]="${DEFAULT_RECIPIENT_SLACK}"
        role_recipients_slack[sitemgr]="${DEFAULT_RECIPIENT_SLACK}"
    exporting:
      enabled: false
      path: /etc/netdata/exporting.conf
      data: ""
    example:
      enabled: false
      path: /etc/netdata/health.d/example.conf
      data: |
        alarm: example_alarm1
          on: example.random
        every: 2s
        warn: $random1 > (($status >= $WARNING)  ? (70) : (80))
        crit: $random1 > (($status == $CRITICAL) ? (80) : (90))
        info: random
          to: sysadmin

  claiming:
    enabled: false
    token: ""
    rooms: ""
    url: "https://app.netdata.cloud"

child:
  enabled: true
  port: "{{ .Values.parent.port }}"

  updateStrategy: {}

  configs:
    netdata:
      enabled: true
      path: /etc/netdata/netdata.conf
      data: |
        [global]
          memory mode = ram
        [health]
          enabled = no
        [plugin:proc:/proc/net/dev]
          disable by default interfaces matching = eni*
        [plugin:cgroups]
          use unified cgroups = yes
          path to unified cgroups = /sys/fs/cgroup
    stream:
      enabled: true
      path: /etc/netdata/stream.conf
      data: |
        [stream]
          enabled = yes
          destination = netdata:{{ .Values.service.port }}
          api key = 684C1DEB-A0FC-441F-8803-F497B0469702
          timeout seconds = 60
          buffer size bytes = 1048576
          reconnect delay seconds = 5
          initial clock resync iterations = 60
    exporting:
      enabled: false
      path: /etc/netdata/exporting.conf
      data: ""
    go.d:
      enabled: true
      path: /etc/netdata/go.d.conf
      data: |
        modules:
          pulsar: no
          prometheus: yes
    kubelet:
      enabled: true
      path: /etc/netdata/go.d/k8s_kubelet.conf
      data: |
        update_every: 1
        autodetection_retry: 0
        jobs:
          - url: http://127.0.0.1:10255/metrics
          - url: https://localhost:10250/metrics
            tls_skip_verify: yes
    kubeproxy:
      enabled: true
      path: /etc/netdata/go.d/k8s_kubeproxy.conf
      data: |
        update_every: 1
        autodetection_retry: 0
        jobs:
          - url: http://127.0.0.1:10249/metrics

  claiming:
    enabled: false
    token: ""
    rooms: ""
    url: "https://app.netdata.cloud"

Thanks in advance!

[ilyam8]

Hey, @fgiannetti. Consider using:

• latest netdata/helmchart release.
• latest stable Netdata image (v1.42.1).

v1.34.1 is very old.

[fgiannetti]

hi!
Agree that 1.34.1 is very old :)
I've tried 1.42.1 but the same problem still happens

I promise to upgrade soon 😉

[fgiannetti]

Using the last version of the helm chart which includes this PR (netdata/helmchart#371) the containers are shown again.
Thanks for the reply!