Paul Bastide

OpenShift Node Feature Discovery Operator – Advanced Configuration

Written by

Paul

in

My teammate and I setup OpenShift: Node Feature Discovery. We wanted to load labels for specific use-cases. I stumbled across Node Feature Discovery: Advanced Configuration, so I edited the NodeFeatureDiscovery resource – nfd-instance.

$ oc -n openshift-nfd edit NodeFeatureDiscovery nfd-instance 
nodefeaturediscovery.nfd.openshift.io/nfd-instance edited

I added to spec.customConfig.klog.addDirHeader and spec.customConfig.klog.v so we can get some nice logging. You’ll want to use the number 3

spec:
  customConfig:
    configData: |
        klog:
          addDirHeader: false
          v: 3

No restart necessary until you start editing the configData. Once you do change the configuration, you’ll see (using oc -n openshift-nfd logs nfd-worker-4v2pw):

I0623 14:19:58.522028       1 memory.go:134] No NVDIMM devices present
I0623 14:39:47.570487       1 memory.go:99] discovered memory features:
I0623 14:39:47.570781       1 memory.go:99]   Instances:
I0623 14:39:47.570791       1 memory.go:99]     nv:
I0623 14:39:47.570798       1 memory.go:99]       Elements: []
I0623 14:39:47.570805       1 memory.go:99]   Keys: {}
I0623 14:39:47.570812       1 memory.go:99]   Values:
I0623 14:39:47.570819       1 memory.go:99]     numa:
I0623 14:39:47.570826       1 memory.go:99]       Elements:
I0623 14:39:47.570833       1 memory.go:99]         is_numa: "false"
I0623 14:39:47.570840       1 memory.go:99]         node_count: "1"
I0623 14:19:58.522052       1 nfd-worker.go:472] starting feature discovery...
I0623 14:19:58.522062       1 nfd-worker.go:484] feature discovery completed
I0623 14:19:58.522079       1 nfd-worker.go:485] labels discovered by feature sources:
I0623 14:19:58.522141       1 nfd-worker.go:485]   {}
I0623 14:19:58.522155       1 nfd-worker.go:565] sending labeling request to nfd-master

You can check your labels at:

$ oc get node -o json  | jq -r '.items[].metadata.labels'
...
{
  "beta.kubernetes.io/arch": "ppc64le",
  "beta.kubernetes.io/os": "linux",
  "cpumanager": "enabled",
  "kubernetes.io/arch": "ppc64le",
  "kubernetes.io/hostname": "worker-1.xip.io",
  "kubernetes.io/os": "linux",
  "node-role.kubernetes.io/worker": "",
  "node.openshift.io/os_id": "rhcos"
}

I then update the configData with labelSources and featureSources so it is using all,-memory, without the memory source. You can see the list of sources at GitHub: node-feature-discovery/source and Feature Discovery: sources

spec:
  workerConfig:
    configData: |
      core:
        labelSources: [-memory,all]
        featureSources: [-memory,all]

Note: There is an additional configuration example at link.

Check your labels and see your labels are there.

$ oc get node -o json  | jq -r '.items[].metadata.labels'
{
  "beta.kubernetes.io/arch": "ppc64le",
  "beta.kubernetes.io/os": "linux",
  "cpumanager": "enabled",
  "feature.node.kubernetes.io/cpu-cpuid.ALTIVEC": "true",
  "feature.node.kubernetes.io/cpu-cpuid.ARCHPMU": "true",
  "feature.node.kubernetes.io/cpu-cpuid.ARCH_2_06": "true",
  "feature.node.kubernetes.io/cpu-cpuid.ARCH_2_07": "true",
  "feature.node.kubernetes.io/cpu-cpuid.ARCH_3_00": "true",
  "feature.node.kubernetes.io/cpu-cpuid.DARN": "true",
  "feature.node.kubernetes.io/cpu-cpuid.DFP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.DSCR": "true",
  "feature.node.kubernetes.io/cpu-cpuid.EBB": "true",
  "feature.node.kubernetes.io/cpu-cpuid.FPU": "true",
  "feature.node.kubernetes.io/cpu-cpuid.HTM": "true",
  "feature.node.kubernetes.io/cpu-cpuid.HTM-NOSC": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IC_SNOOP": "true",
  "feature.node.kubernetes.io/cpu-cpuid.IEEE128": "true",
  "feature.node.kubernetes.io/cpu-cpuid.ISEL": "true",
  "feature.node.kubernetes.io/cpu-cpuid.MMU": "true",
  "feature.node.kubernetes.io/cpu-cpuid.PPC32": "true",
  "feature.node.kubernetes.io/cpu-cpuid.PPC64": "true",
  "feature.node.kubernetes.io/cpu-cpuid.SMT": "true",
  "feature.node.kubernetes.io/cpu-cpuid.TAR": "true",
  "feature.node.kubernetes.io/cpu-cpuid.TRUE_LE": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VCRYPTO": "true",
  "feature.node.kubernetes.io/cpu-cpuid.VSX": "true",
  "feature.node.kubernetes.io/cpu-hardware_multithreading": "true",
  "feature.node.kubernetes.io/kernel-config.NO_HZ": "true",
  "feature.node.kubernetes.io/kernel-config.NO_HZ_FULL": "true",
  "feature.node.kubernetes.io/kernel-selinux.enabled": "true",
  "feature.node.kubernetes.io/kernel-version.full": "4.18.0-305.45.1.el8_4.ppc64le",
  "feature.node.kubernetes.io/kernel-version.major": "4",
  "feature.node.kubernetes.io/kernel-version.minor": "18",
  "feature.node.kubernetes.io/kernel-version.revision": "0",
  "feature.node.kubernetes.io/system-os_release.ID": "rhcos",
  "feature.node.kubernetes.io/system-os_release.OPENSHIFT_VERSION": "4.10",
  "feature.node.kubernetes.io/system-os_release.OSTREE_VERSION": "410.84.202205120749-0",
  "feature.node.kubernetes.io/system-os_release.RHEL_VERSION": "8.4",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID": "4.10",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID.major": "4",
  "feature.node.kubernetes.io/system-os_release.VERSION_ID.minor": "10",
  "kubernetes.io/arch": "ppc64le",
  "kubernetes.io/hostname": "worker1.xip.io",
  "kubernetes.io/os": "linux",
  "node-role.kubernetes.io/worker": "",
  "node.openshift.io/os_id": "rhcos"
}

More posts