1
0
Fork 0
mirror of https://github.com/kubernetes-sigs/node-feature-discovery.git synced 2025-03-06 08:47:04 +00:00
node-feature-discovery/deployment/base/nfd-crds/cr-sample.yaml

161 lines
5 KiB
YAML
Raw Normal View History

apiVersion: nfd.k8s-sigs.io/v1alpha1
kind: NodeFeatureRule
metadata:
name: my-rule
spec:
rules:
# The following feature demonstrates the capabilities of the matchFeatures and
# matchAny matchers.
- name: "my feature rule"
taints:
- effect: PreferNoSchedule
key: "feature.node.kubernetes.io/special-node"
value: "true"
- effect: NoExecute
key: "feature.node.kubernetes.io/dedicated-node"
labels:
"my-complex-feature": "my-value"
# matchFeatures implements a logical AND over feature matchers.
matchFeatures:
- feature: cpu.cpuid
matchExpressions:
AVX512F: {op: Exists}
- feature: cpu.cstate
matchExpressions:
enabled: {op: IsTrue}
- feature: cpu.pstate
matchExpressions:
no_turbo: {op: IsFalse}
scaling_governor: {op: In, value: ["performance"]}
- feature: cpu.rdt
matchExpressions:
RDTL3CA: {op: Exists}
- feature: cpu.sst
matchExpressions:
bf.enabled: {op: IsTrue}
- feature: cpu.topology
matchExpressions:
hardware_multithreading: {op: IsFalse}
- feature: kernel.config
matchExpressions:
X86: {op: Exists}
LSM: {op: InRegexp, value: ["apparmor"]}
- feature: kernel.loadedmodule
matchExpressions:
e1000e: {op: Exists}
- feature: kernel.selinux
matchExpressions:
enabled: {op: IsFalse}
- feature: kernel.version
matchExpressions:
major: {op: In, value: ["5"]}
minor: {op: Gt, value: ["10"]}
- feature: storage.block
matchExpressions:
rotational: {op: In, value: ["0"]}
dax: {op: In, value: ["0"]}
- feature: network.device
matchExpressions:
operstate: {op: In, value: ["up"]}
speed: {op: Gt, value: ["100"]}
- feature: memory.numa
matchExpressions:
node_count: {op: Gt, value: ["2"]}
- feature: memory.nv
matchExpressions:
devtype: {op: In, value: ["nd_dax"]}
mode: {op: In, value: ["memory"]}
- feature: system.osrelease
matchExpressions:
ID: {op: In, value: ["fedora", "centos"]}
- feature: system.name
matchExpressions:
nodename: {op: InRegexp, value: ["^worker-X"]}
- feature: local.label
matchExpressions:
custom-feature-knob: {op: Gt, value: ["100"]}
# matchAny implements a logical IF over all listed matchers (i.e. at
# least one must match)
matchAny:
- matchFeatures:
- feature: pci.device
matchExpressions:
vendor: {op: In, value: ["8086"]}
class: {op: In, value: ["0200"]}
- feature: usb.device
matchExpressions:
vendor: {op: In, value: ["8086"]}
class: {op: In, value: ["02"]}
pkg/apis/nfd: support label name templating Support templating of label names in feature rules. It is available both in NodeFeatureRule CRs and in custom rule configuration of nfd-worker. This patch adds a new 'labelsTemplate' field to the rule spec, making it possible to dynamically generate multiple labels per rule based on the matched features. The feature relies on the golang "text/template" package. When expanded, the template must contain labels in a raw <key>[=<value>] format (where 'value' defaults to "true"), separated by newlines i.e.: - name: <rule-name> labelsTemplate: | <label-1>[=<value-1>] <label-2>[=<value-2>] ... All the matched features of 'matchFeatures' directives are available for templating engine in a nested data structure that can be described in yaml as: . <domain-1>: <key-feature-1>: - Name: <matched-key> - ... <value-feature-1: - Name: <matched-key> Value: <matched-value> - ... <instance-feature-1>: - <attribute-1-name>: <attribute-1-value> <attribute-2-name>: <attribute-2-value> ... - ... <domain-2>: ... That is, the per-feature data available for matching depends on the type of feature that was matched: - "key features": only 'Name' is available - "value features": 'Name' and 'Value' can be used - "instance features": all attributes of the matched instance are available NOTE: In case of matchAny is specified, the template is executed separately against each individual matchFeatures matcher and the eventual set of labels is a superset of all these expansions. Consider the following: - name: <name> labelsTemplate: <template> matchAny: - matchFeatures: <matcher#1> - matchFeatures: <matcher#2> matchFeatures: <matcher#3> In the example above (assuming the overall result is a match) the template would be executed on matcher#1 and/or matcher#2 (depending on whether both or only one of them match), and finally on matcher#3, and all the labels from these separate expansions would be created (i.e. the end result would be a union of all the individual expansions). NOTE 2: The 'labels' field has priority over 'labelsTemplate', i.e. labels specified in the 'labels' field will override any labels originating from the 'labelsTemplate' field. A special case of an empty match expression set matches everything (i.e. matches/returns all existing keys/values). This makes it simpler to write templates that run over all values. Also, makes it possible to later implement support for templates that run over all _keys_ of a feature. Some example configurations: - name: "my-pci-template-features" labelsTemplate: | {{ range .pci.device }}intel-{{ .class }}-{{ .device }}=present {{ end }} matchFeatures: - feature: pci.device matchExpressions: class: {op: InRegexp, value: ["^06"]} vendor: ["8086"] - name: "my-system-template-features" labelsTemplate: | {{ range .system.osrelease }}system-{{ .Name }}={{ .Value }} {{ end }} matchFeatures: - feature: system.osRelease matchExpressions: ID: {op: Exists} VERSION_ID.major: {op: Exists} Imaginative template pipelines are possible, of course, but care must be taken in order to produce understandable and maintainable rule sets.
2021-05-04 16:30:06 +03:00
apis/nfd: add matchName field in feature matcher terms Extend the format of feature matcher terms (the elements of the arrayspecified under under matchFeatures field) with new matchName field. The value of this field is an expression that is evaluated against the names of feature elements instead of their values (values are matched with the matchExpressions field, instead). The matchName field is useful e.g. in template rules for creating per-feature-element labels based on feature names (instead of values) and in non-template rules for checking if (at least) one of certain feature element names are present. If both matchExpressions and matchName for certain feature matcher term is specified, they both must match in order to get an overall match. Also, in this case the list of matched features (used in templating) is the union of the results from matchExpressions and matchName. An example of creating an "avx512" label if any AVX512* CPUID feature is present: - name: "avx wildcard rule" labels: avx512: "true" matchFeatures: - feature: cpu.cpuid matchName: {op: InRegexp, value: ["^AVX512"]} An example of a template rule creating a dynamic set of labels based on the existence of certain kconfig options. - name: "kconfig template rule" labelsTemplate: | {{ range .kernel.config }}kconfig-{{ .Name }}={{ .Value }} {{ end }} matchFeatures: - feature: kernel.config matchName: {op: In, value: ["SWAP", "X86", "ARM"]} NOTE: this patch changes the corner case of nil/null match expressions with instance features (i.e. "matchExpressions: null"). Previously, we returned all instances for templating but now a nil match expression is not evaluated and no instances for templating are returned.
2022-03-17 18:30:32 +02:00
- name: "avx wildcard rule"
labels:
"my-avx-feature": "true"
matchFeatures:
- feature: cpu.cpuid
matchName: {op: InRegexp, value: ["^AVX512"]}
pkg/apis/nfd: support label name templating Support templating of label names in feature rules. It is available both in NodeFeatureRule CRs and in custom rule configuration of nfd-worker. This patch adds a new 'labelsTemplate' field to the rule spec, making it possible to dynamically generate multiple labels per rule based on the matched features. The feature relies on the golang "text/template" package. When expanded, the template must contain labels in a raw <key>[=<value>] format (where 'value' defaults to "true"), separated by newlines i.e.: - name: <rule-name> labelsTemplate: | <label-1>[=<value-1>] <label-2>[=<value-2>] ... All the matched features of 'matchFeatures' directives are available for templating engine in a nested data structure that can be described in yaml as: . <domain-1>: <key-feature-1>: - Name: <matched-key> - ... <value-feature-1: - Name: <matched-key> Value: <matched-value> - ... <instance-feature-1>: - <attribute-1-name>: <attribute-1-value> <attribute-2-name>: <attribute-2-value> ... - ... <domain-2>: ... That is, the per-feature data available for matching depends on the type of feature that was matched: - "key features": only 'Name' is available - "value features": 'Name' and 'Value' can be used - "instance features": all attributes of the matched instance are available NOTE: In case of matchAny is specified, the template is executed separately against each individual matchFeatures matcher and the eventual set of labels is a superset of all these expansions. Consider the following: - name: <name> labelsTemplate: <template> matchAny: - matchFeatures: <matcher#1> - matchFeatures: <matcher#2> matchFeatures: <matcher#3> In the example above (assuming the overall result is a match) the template would be executed on matcher#1 and/or matcher#2 (depending on whether both or only one of them match), and finally on matcher#3, and all the labels from these separate expansions would be created (i.e. the end result would be a union of all the individual expansions). NOTE 2: The 'labels' field has priority over 'labelsTemplate', i.e. labels specified in the 'labels' field will override any labels originating from the 'labelsTemplate' field. A special case of an empty match expression set matches everything (i.e. matches/returns all existing keys/values). This makes it simpler to write templates that run over all values. Also, makes it possible to later implement support for templates that run over all _keys_ of a feature. Some example configurations: - name: "my-pci-template-features" labelsTemplate: | {{ range .pci.device }}intel-{{ .class }}-{{ .device }}=present {{ end }} matchFeatures: - feature: pci.device matchExpressions: class: {op: InRegexp, value: ["^06"]} vendor: ["8086"] - name: "my-system-template-features" labelsTemplate: | {{ range .system.osrelease }}system-{{ .Name }}={{ .Value }} {{ end }} matchFeatures: - feature: system.osRelease matchExpressions: ID: {op: Exists} VERSION_ID.major: {op: Exists} Imaginative template pipelines are possible, of course, but care must be taken in order to produce understandable and maintainable rule sets.
2021-05-04 16:30:06 +03:00
# The following features demonstreate label templating capabilities
- name: "my system template feature"
labelsTemplate: |
{{ range .system.osrelease }}my-system-feature.{{ .Name }}={{ .Value }}
{{ end }}
matchFeatures:
- feature: system.osrelease
matchExpressions:
ID: {op: InRegexp, value: ["^open.*"]}
VERSION_ID.major: {op: In, value: ["13", "15"]}
- name: "my pci template feature"
labelsTemplate: |
{{ range .pci.device }}my-pci-device.{{ .class }}-{{ .device }}=with-cpuid
{{ end }}
matchFeatures:
- feature: pci.device
matchExpressions:
class: {op: InRegexp, value: ["^06"]}
vendor: {op: In, value: ["8086"]}
- feature: cpu.cpuid
matchExpressions:
AVX: {op: Exists}
pkg/apis/nfd: add variables to rule spec and support backreferences Support backreferencing of output values from previous rules. Enables complex rule setups where custom features are further combined together to form even more sophisticated higher level labels. The labels created by preceding rules are available as a special 'rule.matched' feature (for matchFeatures to use). If referencing rules accross multiple configs/CRDs care must be taken with the ordering. Processing order of rules in nfd-worker: 1. Static rules 2. Files from /etc/kubernetes/node-feature-discovery/custom.d/ in alphabetical order. Subdirectories are processed by reading their files in alphabetical order. 3. Custom rules from main nfd-worker.conf In nfd-master, NodeFeatureRule objects are processed in alphabetical order (based on their metadata.name). This patch also adds new 'vars' fields to the rule spec. Like 'labels', it is a map of key-value pairs but no labels are generated from these. The values specified in 'vars' are only added for backreferencing into the 'rules.matched' feature. This may by desired in schemes where the output of certain rules is only used as intermediate variables for other rules and no labels out of these are wanted. An example setup: - name: "kernel feature" labels: kernel-feature: matchFeatures: - feature: kernel.version matchExpressions: major: {op: Gt, value: ["4"]} - name: "intermediate var feature" vars: nolabel-feature: "true" matchFeatures: - feature: cpu.cpuid matchExpressions: AVX512F: {op: Exists} - feature: pci.device matchExpressions: vendor: {op: In, value: ["8086"]} device: {op: In, value: ["1234", "1235"]} - name: top-level-feature matchFeatures: - feature: rule.matched matchExpressions: kernel-feature: "true" nolabel-feature: "true"
2021-06-18 18:29:08 +03:00
# The following examples demonstrate vars field and back-referencing
# previous labels and vars
- name: "my dummy kernel rule"
labels:
"my.kernel.feature": "true"
matchFeatures:
- feature: kernel.version
matchExpressions:
major: {op: Gt, value: ["2"]}
- name: "my dummy rule with no labels"
vars:
"my.dummy.var": "1"
matchFeatures:
- feature: cpu.cpuid
matchExpressions: {}
- name: "my rule using backrefs"
labels:
"my.backref.feature": "true"
matchFeatures:
- feature: rule.matched
matchExpressions:
my.kernel.feature: {op: IsTrue}
my.dummy.var: {op: Gt, value: ["0"]}
apis/nfd: add matchName field in feature matcher terms Extend the format of feature matcher terms (the elements of the arrayspecified under under matchFeatures field) with new matchName field. The value of this field is an expression that is evaluated against the names of feature elements instead of their values (values are matched with the matchExpressions field, instead). The matchName field is useful e.g. in template rules for creating per-feature-element labels based on feature names (instead of values) and in non-template rules for checking if (at least) one of certain feature element names are present. If both matchExpressions and matchName for certain feature matcher term is specified, they both must match in order to get an overall match. Also, in this case the list of matched features (used in templating) is the union of the results from matchExpressions and matchName. An example of creating an "avx512" label if any AVX512* CPUID feature is present: - name: "avx wildcard rule" labels: avx512: "true" matchFeatures: - feature: cpu.cpuid matchName: {op: InRegexp, value: ["^AVX512"]} An example of a template rule creating a dynamic set of labels based on the existence of certain kconfig options. - name: "kconfig template rule" labelsTemplate: | {{ range .kernel.config }}kconfig-{{ .Name }}={{ .Value }} {{ end }} matchFeatures: - feature: kernel.config matchName: {op: In, value: ["SWAP", "X86", "ARM"]} NOTE: this patch changes the corner case of nil/null match expressions with instance features (i.e. "matchExpressions: null"). Previously, we returned all instances for templating but now a nil match expression is not evaluated and no instances for templating are returned.
2022-03-17 18:30:32 +02:00
- name: "kconfig template rule"
labelsTemplate: |
{{ range .kernel.config }}kconfig-{{ .Name }}={{ .Value }}
{{ end }}
matchFeatures:
- feature: kernel.config
matchName: {op: In, value: ["SWAP", "X86", "ARM"]}