content large_stringlengths 3 20.5k | url large_stringlengths 53 192 ⌀ | branch large_stringclasses 4
values | source large_stringclasses 50
values | embeddings listlengths 384 384 | score float64 -0.21 0.65 |
|---|---|---|---|---|---|
# Traefik & HTTP Provide your [install configuration](../overview.md) via an HTTP(S) endpoint and let Traefik do the rest! ## Configuration Example You can enable the HTTP provider as detailed below: ```yaml tab="File (YAML)" providers: http: endpoint: "http://127.0.0.1:9000/api" ``` ```toml tab="File (TOML)" [provider... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/others/http.md | master | traefik | [
-0.05407603457570076,
0.04211290180683136,
-0.0980408564209938,
-0.06570347398519516,
-0.055774953216314316,
-0.11901724338531494,
-0.046331070363521576,
0.02321658656001091,
0.006639493629336357,
0.023478347808122635,
0.025136033073067665,
-0.040959324687719345,
-0.036169715225696564,
-0.... | 0.076849 |
# File The file provider lets you define the [install configuration](../overview.md) in a YAML or TOML file. It supports providing configuration through a single configuration file or multiple separate files. !!! info The file provider is the default format used throughout the documentation to show samples of the confi... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/others/file.md | master | traefik | [
-0.05192676931619644,
-0.011626765131950378,
-0.03522087633609772,
-0.0701659619808197,
0.021784601733088493,
-0.025164619088172913,
-0.005238788668066263,
0.05756013095378876,
-0.056613121181726456,
-0.017295822501182556,
0.021188635379076004,
0.06373055279254913,
0.018876733258366585,
-0... | 0.005929 |
# Traefik & AWS ECS ## Configuration Example You can enable the ECS provider with as detailed below: ```yaml tab="File (YAML)" providers: ecs: {} ``` ```toml tab="File (TOML)" [providers.ecs] ``` ```bash tab="CLI" --providers.ecs=true ``` ## Configuration Options | Field | Description | Default | Required | |:------|:-... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/others/ecs.md | master | traefik | [
-0.05030565336346626,
0.05878818780183792,
-0.06403148174285889,
-0.007567839231342077,
0.008398937061429024,
-0.06760857254266739,
-0.0064131030812859535,
0.028268497437238693,
0.05891243368387222,
0.007973571307957172,
0.07611307501792908,
-0.07419218122959137,
0.012607280164957047,
-0.0... | 0.05668 |
```toml # With logical AND and OR, with precedence set by parentheses. constraints = "Label(`a.label.name`, `valueA`) && (Label(`another.label.name`, `valueB`) || Label(`yet.another.label.name`, `valueC`))" ``` ```toml # Includes only containers having a label with key `a.label.name` and a value matching the `a.+` regu... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/others/ecs.md | master | traefik | [
-0.04549681022763252,
0.0802462100982666,
-0.0334588848054409,
-0.07523217052221298,
-0.006400623824447393,
-0.0014197026612237096,
0.07189604640007019,
0.010224109515547752,
0.04173044487833977,
-0.052696362137794495,
0.06505361199378967,
-0.08724455535411835,
0.07018870860338211,
-0.0310... | 0.014722 |
Traefik provides some Kubernetes Custom Resources, such as `IngressRoute`, `Middleware`, etc. When using KubernetesCRD as a provider, Traefik uses [Custom Resource Definition](https://kubernetes.io/docs/concepts/extend-kubernetes/api-extension/custom-resources/) to retrieve its routing configuration. Traefik Custom Res... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-crd.md | master | traefik | [
-0.005242965649813414,
0.019098209217190742,
-0.03582969680428505,
-0.015812955796718597,
-0.026295974850654602,
-0.0766192302107811,
0.002700931392610073,
0.02321036346256733,
0.0557861328125,
0.053766265511512756,
-0.01364954188466072,
-0.07091841101646423,
-0.046148527413606644,
-0.0524... | 0.150187 |
are allowed. When set to `[]`, every cross-provider reference is rejected. | [] | No | | [`providers.kubernetesCRD.nativeLBByDefault`](#opt-providers-kubernetesCRD-nativeLBByDefault "#opt-providers-kubernetesCRD-nativeLBByDefault") | Allow using the Kubernetes Service load balancing between the pods instead of the one ... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-crd.md | master | traefik | [
-0.05356458202004433,
0.012959674000740051,
-0.02016066387295723,
-0.0029154059011489153,
-0.027945900335907936,
-0.029165254905819893,
0.03291502222418785,
-0.025605536997318268,
-0.028092864900827408,
0.04806935042142868,
-0.04255286604166031,
-0.07313114404678345,
-0.00797957181930542,
... | 0.154285 |
# Traefik & Knative The Traefik Knative provider integrates with Knative Serving to provide advanced traffic management and routing capabilities for serverless applications. [Knative](https://knative.dev) is a Kubernetes-based platform that enables serverless workloads with features like scale-to-zero, automatic scalin... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/knative.md | master | traefik | [
-0.01663314923644066,
0.00898743886500597,
-0.03238590434193611,
-0.04611041024327278,
-0.06573578715324402,
-0.09233752638101578,
0.0027622722554951906,
-0.006109355483204126,
0.018752869218587875,
0.04728462174534798,
-0.025459827855229378,
-0.03261275216937065,
-0.039304111152887344,
-0... | 0.215868 |
Namespace of the public Kubernetes service. | | ### `endpoint` The Kubernetes server endpoint URL. When deployed into Kubernetes, Traefik reads the environment variables `KUBERNETES\_SERVICE\_HOST` and `KUBERNETES\_SERVICE\_PORT` or `KUBECONFIG` to construct the endpoint. The access token is looked up in `/var/run/secr... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/knative.md | master | traefik | [
-0.06557921320199966,
0.014052682556211948,
-0.05807941406965256,
0.0009587768581695855,
-0.06542449444532394,
-0.0928763672709465,
0.03493407368659973,
-0.0013441067421808839,
0.13915537297725677,
-0.004050291143357754,
-0.0006440697470679879,
-0.06987866759300232,
-0.013689286075532436,
... | 0.211763 |
# Traefik & Kubernetes with Gateway API The Kubernetes Gateway provider is a Traefik implementation of the [Gateway API](https://gateway-api.sigs.k8s.io/) specification from the Kubernetes Special Interest Groups (SIGs). This provider supports Standard version [v1.5.1](https://github.com/kubernetes-sigs/gateway-api/rel... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-gateway.md | master | traefik | [
-0.05236411467194557,
-0.0028416833374649286,
-0.02386639267206192,
-0.008827227167785168,
-0.03952315077185631,
-0.07953396439552307,
0.0010925718815997243,
0.012625220231711864,
0.06461912393569946,
0.02204519882798195,
-0.0379892997443676,
-0.016507048159837723,
-0.06518593430519104,
-0... | 0.168948 |
used to copy the service `loadbalancer.status` (containing the service's endpoints IPs) to the Gateway `status.addresses`. | "" | No | | [`providers.kubernetesGateway.` `statusAddress.service.name`](#opt-providers-kubernetesGateway-statusAddress-service-name "#opt-providers-kubernetesGateway-statusAddress-service-name"... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-gateway.md | master | traefik | [
-0.0905078649520874,
-0.032959967851638794,
-0.018138814717531204,
-0.026227613911032677,
-0.059682395309209824,
-0.02839501202106476,
-0.015360728837549686,
-0.037818003445863724,
0.06933125108480453,
0.03374089300632477,
-0.051566336303949356,
-0.020756695419549942,
0.004758245311677456,
... | 0.110187 |
# Traefik & Ingresses with NGINX Annotations This provider is a Kubernetes Ingress controller that manages access to cluster services by supporting the [Ingress](https://kubernetes.io/docs/concepts/services-networking/ingress/) specification. It also supports many of the [ingress-nginx](https://kubernetes.github.io/ing... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress-nginx.md | master | traefik | [
-0.11434043943881989,
0.029441403225064278,
-0.03931018337607384,
0.007748504634946585,
0.00783004891127348,
-0.05585617199540138,
0.004471412859857082,
-0.06580384075641632,
0.010441956110298634,
0.07010235637426376,
-0.015583928674459457,
-0.010325802490115166,
-0.03307849168777466,
-0.0... | 0.237185 |
kubernetesIngressNginx: # -- Enable Kubernetes Ingress NGINX provider enabled: true # Namespace discovery # -- Namespace the controller watches for updates to Kubernetes objects # When using rbac.namespaced, it will watch helm release namespace and namespaces listed in this array namespaces: - default # OR use namespac... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress-nginx.md | master | traefik | [
-0.11695536226034164,
0.010242147371172905,
-0.014349449425935745,
0.029247529804706573,
0.017101561650633812,
-0.020303284749388695,
0.07616356760263443,
-0.05775820463895798,
0.07853624224662781,
0.06410962343215942,
-0.03937896341085434,
-0.10077545046806335,
0.0073668030090630054,
-0.0... | 0.178099 |
`publishService`](#opt-providers-kubernetesIngressNGINX-publishService "#opt-providers-kubernetesIngressNGINX-publishService") | Service fronting the Ingress controller. Takes the form `namespace/name`. | "" | No | | [`providers.` `kubernetesIngressNGINX.` `publishStatusAddress`](#opt-providers-kubernetesIngressNGINX-p... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress-nginx.md | master | traefik | [
-0.04051563888788223,
0.02831539325416088,
-0.03282686695456505,
0.029276810586452484,
-0.09222545474767685,
0.013220211490988731,
0.023242004215717316,
-0.02662421204149723,
0.032055627554655075,
0.10273002833127975,
-0.011175447143614292,
-0.05913255736231804,
0.021986620500683784,
-0.03... | 0.156687 |
[`nginx.ingress.kubernetes.io/proxy-next-upstream-timeout`](../../../../routing-configuration/kubernetes/ingress-nginx/#opt-nginx-ingress-kubernetes-ioproxy-next-upstream-timeout) annotation. | 0 | No | | [`providers.` `kubernetesIngressNGINX.` `upstreamKeepaliveTimeout`](#opt-providers-kubernetesIngressNGINX-upstreamK... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress-nginx.md | master | traefik | [
-0.06807812303304672,
0.054155461490154266,
-0.0028126821853220463,
0.047230299562215805,
-0.014968528412282467,
-0.0181744322180748,
0.016917100176215172,
-0.023042302578687668,
0.07649451494216919,
0.03489406779408455,
-0.006077949423342943,
-0.005839704070240259,
-0.006783674471080303,
... | 0.15422 |
# Traefik & Kubernetes The Traefik Kubernetes Ingress provider is a Kubernetes Ingress controller; i.e, it manages access to cluster services by supporting the [Ingress](https://kubernetes.io/docs/concepts/services-networking/ingress/) specification. ??? warning "Ingress Backend Resource not supported" Referencing back... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress.md | master | traefik | [
-0.05481182411313057,
0.0023523608688265085,
-0.06155145913362503,
0.014301127754151821,
-0.02072485163807869,
-0.024849971756339073,
0.03326916694641113,
0.008320549502968788,
0.04433509707450867,
0.07340555638074875,
-0.01801588200032711,
-0.0397857241332531,
-0.012905227951705456,
-0.01... | 0.178146 |
| | [`providers.kubernetesIngress.crossProviderNamespaces`](#opt-providers-kubernetesIngress-crossProviderNamespaces "#opt-providers-kubernetesIngress-crossProviderNamespaces") | List of namespaces from which Ingresses or Services are allowed to use `traefik.ingress.kubernetes.io/router.middlewares`, `traefik.ingress.k... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/providers/kubernetes/kubernetes-ingress.md | master | traefik | [
-0.08054642379283905,
-0.014929077588021755,
-0.02352888323366642,
-0.04087217152118683,
-0.0024886643514037132,
-0.0638836920261383,
0.04523507133126259,
-0.06041688844561577,
0.009615221060812473,
0.030407080426812172,
0.0020965987350791693,
-0.07289088517427444,
-0.015167372301220894,
-... | 0.139301 |
# Tracing Visualize the Requests Flow {: .subtitle } The tracing system allows developers to visualize call flows in their infrastructure. Traefik uses [OpenTelemetry](https://opentelemetry.io/ "Link to website of OTel"), an open standard designed for distributed tracing. ## Configuration Example To enable the tracing:... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/tracing.md | master | traefik | [
-0.05558714643120766,
0.04635850712656975,
-0.04489302262663841,
-0.017184186726808548,
-0.017851071432232857,
-0.1740412712097168,
0.017638199031352997,
-0.014352360740303993,
0.06431785225868225,
-0.019149774685502052,
-0.024281494319438934,
-0.08469691872596741,
-0.049032121896743774,
-... | 0.16809 |
| Yes | | [`tracing.otlp.grpc.tls.ca`](#opt-tracing-otlp-grpc-tls-ca "#opt-tracing-otlp-grpc-tls-ca") | Path to the certificate authority used for the secure connection to the OpenTelemetry Collector, it defaults to the system bundle. | "" | No | | [`tracing.otlp.grpc.tls.cert`](#opt-tracing-otlp-grpc-tls-cert "#opt-tr... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/tracing.md | master | traefik | [
-0.04758760705590248,
0.03418751433491707,
-0.06626204401254654,
0.022344615310430527,
-0.03225582465529442,
-0.11519496887922287,
0.021188605576753616,
-0.0037210064474493265,
-0.003284631296992302,
-0.045482221990823746,
0.060312945395708084,
-0.056743890047073364,
0.0017249928787350655,
... | 0.002348 |
# Metrics Traefik provides metrics in the [OpenTelemetry](#open-telemetry) format as well as the following vendor specific backends: - [Datadog](#datadog) - [InfluxDB2](#influxdb-v2) - [Prometheus](#prometheus) - [StatsD](#statsd) Traefik Proxy has an official Grafana dashboard for both [on-premises](https://grafana.co... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.09248203039169312,
0.0003420731227379292,
-0.0667402595281601,
0.016216980293393135,
-0.08235818892717361,
-0.0795903280377388,
-0.03168031945824623,
-0.003955651540309191,
0.057662494480609894,
0.013263135217130184,
-0.026736391708254814,
-0.11947111785411835,
-0.05498386546969414,
-0.... | 0.155338 |
| - | No | | [`metrics.otlp.grpc.insecure`](#opt-metrics-otlp-grpc-insecure "#opt-metrics-otlp-grpc-insecure") | Allows exporter to send metrics to the OpenTelemetry Collector without using a secured protocol. | false | Yes | | [`metrics.otlp.grpc.tls.ca`](#opt-metrics-otlp-grpc-tls-ca "#opt-metrics-otlp-grpc-tls-ca") ... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.02747025527060032,
0.05012647435069084,
-0.08241496235132217,
0.0349179208278656,
-0.05144058167934418,
-0.09663375467061996,
0.03566690906882286,
-0.0013498197076842189,
0.01859895884990692,
-0.02765745483338833,
0.06500250846147537,
-0.1072062999010086,
0.006885559298098087,
0.0493103... | 0.032999 |
points. | true | No | | [`metrics.influxDB2.addRoutersLabels`](#opt-metrics-influxDB2-addRoutersLabels "#opt-metrics-influxDB2-addRoutersLabels") | Enable metrics on routers. | false | No | | [`metrics.influxDB2.addServicesLabels`](#opt-metrics-influxDB2-addServicesLabels "#opt-metrics-influxDB2-addServicesLabels") | E... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.021723730489611626,
-0.035690803080797195,
-0.052334003150463104,
0.056723013520240784,
0.0006897794082760811,
-0.016510948538780212,
0.053691402077674866,
-0.035728614777326584,
0.01075576152652502,
-0.044583071023225784,
0.03458838909864426,
-0.06384820491075516,
0.05640782415866852,
... | 0.028445 |
metrics for internal resources (e.g.: `ping@internals`). | false | No | | [`metrics.statsD.addEntryPointsLabels`](#opt-metrics-statsD-addEntryPointsLabels "#opt-metrics-statsD-addEntryPointsLabels") | Enable metrics on entry points. | true | No | | [`metrics.statsD.addRoutersLabels`](#opt-metrics-statsD-addRoutersLabel... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.030773092061281204,
-0.06295589357614517,
-0.1002519428730011,
0.05050140619277954,
-0.0006579632754437625,
-0.03505624830722809,
0.011837156489491463,
0.000036830329918302596,
-0.01408687885850668,
0.0022596241906285286,
0.013554261066019535,
-0.03196589648723602,
0.024169625714421272,
... | 0.052291 |
follows [official OpenTelemetry semantic conventions v1.23.1](https://github.com/open-telemetry/semantic-conventions/blob/v1.23.1/docs/http/http-metrics.md). #### HTTP Server | Metric | Type | [Labels](#labels) | Description | |----------|-----------|-------------------------|------------------| | [`http.server.request... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
0.024632230401039124,
0.04033661261200905,
-0.06302201002836227,
-0.026747196912765503,
-0.05991659685969353,
-0.1448698192834854,
-0.05532607063651085,
0.021463792771100998,
0.004187587182968855,
0.0018362924456596375,
0.03352109715342522,
-0.051994506269693375,
0.0016073761507868767,
-0.... | 0.123964 |
| `code`, `method`, `protocol`, `entrypoint` | Request processing duration histogram on an entrypoint. | | [`traefik\_entrypoint\_requests\_bytes\_total`](#opt-traefik-entrypoint-requests-bytes-total-2 "#opt-traefik-entrypoint-requests-bytes-total-2") | Count | `code`, `method`, `protocol`, `entrypoint` | The total siz... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.04715120047330856,
0.06175042316317558,
-0.11321938037872314,
0.023164384067058563,
-0.08080712705850601,
-0.1485174149274826,
0.054929669946432114,
0.05749886482954025,
-0.017489485442638397,
0.021274834871292114,
0.0019791473168879747,
-0.07134338468313217,
0.050297509878873825,
-0.07... | 0.16397 |
`code`, `method`, `protocol`, `router`, `service` | The total size of HTTP responses in bytes handled by a router. | === "Prometheus" | Metric | Type | [Labels](#labels) | Description | |-----------------------|-----------|---------------------------------------------------|---------------------------------------------... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.07929059863090515,
0.08112861961126328,
-0.0954422652721405,
0.012404474429786205,
-0.057539161294698715,
-0.1244443729519844,
0.010330619290471077,
0.016089897602796555,
-0.007353269960731268,
0.010990036651492119,
0.006007220596075058,
-0.028955910354852676,
-0.026201454922556877,
-0.... | 0.142598 |
Description | |-----------------------|-----------|------------|------------| | [`traefik\_service\_requests\_total`](#opt-traefik-service-requests-total "#opt-traefik-service-requests-total") | Count | `code`, `method`, `protocol`, `service` | The total count of HTTP requests processed on a service. | | [`traefik\_ser... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.07927872240543365,
0.07865148782730103,
-0.09845974296331406,
-0.0026043683756142855,
-0.032465860247612,
-0.15100035071372986,
0.07734843343496323,
-0.039039917290210724,
0.02130073867738247,
0.005662866868078709,
0.046697307378053665,
-0.05521106719970703,
0.07584366947412491,
-0.0361... | 0.102506 |
`service` | Request processing duration histogram on a service. | | [`traefik.service.retries.total`](#opt-traefik-service-retries-total-3 "#opt-traefik-service-retries-total-3") | Count | `service` | The count of requests retries on a service. | | [`traefik.service.server.up`](#opt-traefik-service-server-up-3 "#opt-tr... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/metrics.md | master | traefik | [
-0.07783646881580353,
0.03609006851911545,
-0.10816251486539841,
0.011447490192949772,
-0.06675595790147781,
-0.1753639280796051,
0.09360295534133911,
0.011108675971627235,
0.0034600154031068087,
0.02867158129811287,
-0.0010000340407714248,
-0.003528364934027195,
-0.001165903639048338,
-0.... | 0.176399 |
# CLI & Ping Checking the Health of your Traefik Instances {: .subtitle } ## CLI The CLI can be used to make a request to the `/ping` endpoint to check the health of Traefik. Its exit status is `0` if Traefik is healthy and `1` otherwise. This can be used with [HEALTHCHECK](https://docs.docker.com/engine/reference/buil... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/healthcheck.md | master | traefik | [
0.0035820570774376392,
0.10222460329532623,
-0.1004442498087883,
-0.05335264652967453,
0.021424690261483192,
-0.13732638955116272,
0.08234476298093796,
0.012572934851050377,
0.014555185101926327,
0.027378233149647713,
-0.014037731103599072,
-0.08733057975769043,
0.00746564706787467,
-0.059... | 0.16246 |
## Logs Logs concern everything that happens to Traefik itself (startup, configuration, events, shutdown, and so on). ### Configuration Example ```yaml tab="File (YAML)" log: filePath: "/path/to/log-file.log" format: json level: INFO ``` ```toml tab="File (TOML)" [log] filePath = "/path/to/log-file.log" format = "json"... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.03257652744650841,
0.04397932067513466,
-0.04578595981001854,
0.015167523175477982,
0.03291485086083412,
-0.12336607277393341,
0.024823494255542755,
0.04858636111021042,
0.04476432502269745,
0.017101891338825226,
0.031777285039424896,
-0.030025539919734,
-0.03804386407136917,
0.02222904... | 0.090378 |
| | [`log.otlp.http.tls.ca`](#opt-log-otlp-http-tls-ca "#opt-log-otlp-http-tls-ca") | The path to the certificate authority used for the secure connection to the OpenTelemetry Collector, it defaults to the system bundle. | | No | | [`log.otlp.http.tls.cert`](#opt-log-otlp-http-tls-cert "#opt-log-otlp-http-tls-cert") | ... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.03258664906024933,
0.09115118533372879,
-0.0677938461303711,
0.004379404243081808,
0.010229016654193401,
-0.12622907757759094,
0.004456989001482725,
-0.012717648409307003,
0.0008250821847468615,
-0.029657145962119102,
0.07955297082662582,
-0.06708946079015732,
0.04587491601705551,
0.018... | 0.036076 |
= "drop" ``` ```sh tab="CLI" --accesslog=true --accesslog.dualoutput=true --accesslog.format=json --accesslog.filters.statuscodes=200,300-302 --accesslog.filters.retryattempts --accesslog.filters.minduration=10ms --accesslog.fields.defaultmode=keep --accesslog.fields.names.ClientUsername=drop --accesslog.fields.headers... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.005568759515881538,
0.0747414380311966,
-0.04800118878483772,
0.07297992706298828,
-0.029783856123685837,
-0.05568133667111397,
0.02294781431555748,
-0.007883347570896149,
0.012948023155331612,
0.0066362712532281876,
0.004703330807387829,
-0.03727392107248306,
0.026191115379333496,
0.00... | 0.042795 |
to the collector. See [resourceAttributes](#resourceattributes\_1) for details. | [] | No | | [`accesslog.otlp.http`](#opt-accesslog-otlp-http "#opt-accesslog-otlp-http") | This instructs the exporter to send access logs to the OpenTelemetry Collector using HTTP. | | No | | [`accesslog.otlp.http.endpoint`](#opt-accessl... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.0004384970525279641,
0.12159059196710587,
-0.06604797393083572,
0.07097738981246948,
0.04248466715216637,
-0.11457175016403198,
-0.008155344054102898,
-0.042365819215774536,
0.013809015974402428,
0.036350853741168976,
0.011118029244244099,
-0.08504088968038559,
-0.010264059528708458,
-0... | 0.059331 |
[`StartUTC`](#opt-StartUTC "#opt-StartUTC") | The time at which request processing started. | | [`StartLocal`](#opt-StartLocal "#opt-StartLocal") | The local time at which request processing started. | | [`Duration`](#opt-Duration "#opt-Duration") | The total time taken (in nanoseconds) by processing the response, incl... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.03762168064713478,
0.05016785487532616,
-0.11310479044914246,
0.0007843263447284698,
0.0157704446464777,
-0.16374649107456207,
-0.024268386885523796,
0.008714606054127216,
0.031098786741495132,
0.05833209678530693,
0.011481581255793571,
-0.030634209513664246,
-0.056249767541885376,
-0.0... | 0.092003 |
| The name of the Kubernetes Ingress resource the router handles. Only available with the Kubernetes Ingress and Kubernetes Ingress Nginx providers. | | [`KubernetesServiceName`](#opt-KubernetesServiceName "#opt-KubernetesServiceName") | The name of the Kubernetes Service associated with the Ingress the router handles.... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/observability/logs-and-accesslogs.md | master | traefik | [
-0.08040682971477509,
0.014286392368376255,
-0.04033923149108887,
-0.023109130561351776,
0.01297070924192667,
-0.06639015674591064,
0.04034210368990898,
-0.03623969480395317,
0.08565207570791245,
0.07812102138996124,
0.03274950385093689,
0.03540382534265518,
-0.028173675760626793,
-0.02350... | 0.201097 |
# OCSP Check certificate status and perform OCSP stapling. {: .subtitle } ## Overview ### OCSP Stapling When OCSP is enabled, Traefik checks the status of every certificate in the store that provides an OCSP responder URL, including the default certificate, and staples the OCSP response to the TLS handshake. The OCSP c... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/ocsp.md | master | traefik | [
-0.09085472673177719,
0.059041064232587814,
-0.06396742165088654,
0.04181782156229019,
0.04228181391954422,
-0.19899307191371918,
0.016377054154872894,
0.0024561688769608736,
0.08840183168649673,
-0.06185964494943619,
0.0257940124720335,
0.01899268664419651,
-0.040108971297740936,
-0.04014... | 0.065255 |
# SPIFFE Secure the backend connection with SPIFFE. {: .subtitle } [SPIFFE](https://spiffe.io/docs/latest/spiffe-about/overview/) (Secure Production Identity Framework For Everyone), provides a secure identity in the form of a specially crafted X.509 certificate, to every workload in an environment. Traefik is able to ... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/spiffe.md | master | traefik | [
-0.0815366804599762,
0.010854125022888184,
-0.12170494347810745,
0.005094804335385561,
-0.009045298211276531,
-0.1088598296046257,
-0.01169572863727808,
-0.02709304541349411,
-0.026721974834799767,
-0.036602601408958435,
0.026934100314974785,
-0.060252681374549866,
0.008347940631210804,
-0... | 0.045776 |
# Tailscale Provision TLS certificates for your internal Tailscale services. {: .subtitle } To protect a service with TLS, a certificate from a public Certificate Authority is needed. In addition to its vpn role, Tailscale can also [provide certificates](https://tailscale.com/kb/1153/enabling-https/) for the machines i... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/tailscale.md | master | traefik | [
-0.04347376525402069,
0.05685238912701607,
0.04434477910399437,
-0.03466483950614929,
-0.01779997907578945,
-0.044582050293684006,
0.02047831006348133,
0.03497782349586487,
0.04702160879969597,
-0.014304602518677711,
-0.015182431787252426,
-0.05396685004234314,
0.06087478995323181,
0.03758... | -0.054698 |
## Configuration Example Below is an example of a basic configuration for ACME in Traefik. ```yaml tab="File (YAML)" entryPoints: web: address: ":80" websecure: address: ":443" certificatesResolvers: myresolver: acme: email: your-email@example.com storage: acme.json httpChallenge: # used during the challenge entryPoint... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/acme.md | master | traefik | [
-0.08639201521873474,
0.04705224558711052,
-0.11143123358488083,
0.013350747525691986,
0.008404701016843319,
-0.07806307077407837,
0.046569135040044785,
0.00835611205548048,
0.0058046928606927395,
-0.003979166503995657,
0.03333443030714989,
-0.10182579606771469,
0.07914410531520844,
-0.003... | 0.066726 |
"#opt-acme-dnsChallenge-resolvers") | DNS servers to resolve the FQDN authority. | [] | No | | [`acme.dnsChallenge.propagation.delayBeforeChecks`](#opt-acme-dnsChallenge-propagation-delayBeforeChecks "#opt-acme-dnsChallenge-propagation-delayBeforeChecks") | By default, the provider will verify the TXT DNS challenge rec... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/acme.md | master | traefik | [
-0.12785862386226654,
0.001263848040252924,
-0.010275307111442089,
0.07004045695066452,
-0.027294939383864403,
-0.05109812691807747,
-0.03272685408592224,
-0.02142402157187462,
-0.0210829209536314,
0.002084860112518072,
0.03653465211391449,
-0.052713096141815186,
0.00019582612731028348,
0.... | 0.018665 |
`example.org` (account foo) and `example.com` (account bar) you can create a CNAME on `example.org` called `\_acme-challenge.example.org` pointing to `challenge.example.com`. This way, you can obtain certificates for `example.org` with the bar account. ??? info "`delayBeforeCheck`" By default, the `provider` verifies t... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/acme.md | master | traefik | [
-0.12640978395938873,
0.002644528169184923,
-0.02988344244658947,
0.030614444985985756,
-0.015576479025185108,
-0.07184385508298874,
-0.002602167194709182,
-0.004294693004339933,
0.04973287880420685,
-0.046043511480093,
-0.0004891962744295597,
-0.10160431265830994,
0.07246338576078415,
0.0... | 0.01441 |
URL Encoding without padding format ```yaml tab="File (YAML)" certificatesResolvers: myresolver: acme: # ... eab: kid: abc-keyID-xyz hmacEncoded: abc-hmac-xyz ``` ```toml tab="File (TOML)" [certificatesResolvers.myresolver.acme] # ... [certificatesResolvers.myresolver.acme.eab] kid = "abc-keyID-xyz" hmacEncoded = "abc-... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/acme.md | master | traefik | [
-0.014192946255207062,
0.033931612968444824,
0.02668628841638565,
-0.07632042467594147,
-0.033333756029605865,
-0.01917017064988613,
-0.0000465968441858422,
-0.04266532510519028,
0.09996598213911057,
-0.028794147074222565,
-0.004888458177447319,
-0.06794992834329605,
0.08130130171775818,
0... | -0.09353 |
In Traefik, TLS Certificates can be generated using Certificates Resolvers. In Traefik, two certificate resolvers exist: - [`acme`](./acme.md): It allows generating ACME certificates stored in a file (not distributed). - [`tailscale`](./tailscale.md): It allows provisioning TLS certificates for internal Tailscale servi... | https://github.com/traefik/traefik/blob/master/docs/content/reference/install-configuration/tls/certificate-resolvers/overview.md | master | traefik | [
-0.09591119736433029,
0.058768026530742645,
-0.037701357156038284,
0.021965812891721725,
-0.02319801039993763,
-0.12215561419725418,
-0.02138197049498558,
0.04067140445113182,
0.04659373685717583,
-0.04090135917067528,
-0.0064792148768901825,
-0.017409978434443474,
0.014358429238200188,
-0... | 0.047667 |
# Disruption Controls By Reason ## User Stories 1. Users need the capability to schedule upgrades only during business hours or within more restricted time windows. Additionally, they require a system that doesn't compromise the cost savings from consolidation when upgrades are blocked due to drift. 2. Users want to mi... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/disruption-controls-by-reason.md | main | karpenter | [
-0.033467456698417664,
0.03430568799376488,
0.08257271349430084,
0.07507724314928055,
0.01302582398056984,
-0.03960568457841873,
-0.0010138021316379309,
-0.03613700345158577,
0.03258912265300751,
0.0472450889647007,
-0.010252278298139572,
-0.022759266197681427,
0.00011243912013014778,
-0.0... | 0.172243 |
// a 0s at the end. // +kubebuilder:validation:Pattern=`^([0-9]+(m|h)+(0s)?)$` // +kubebuilder:validation:Type="string" // +optional Duration \*metav1.Duration `json:"duration,omitempty" hash:"ignore"` } ``` #### Example ```yaml apiVersion: karpenter.sh/v1beta1 kind: NodePool metadata: name: default spec: # This is not... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/disruption-controls-by-reason.md | main | karpenter | [
-0.01659795455634594,
0.06311236321926117,
0.030467944219708443,
0.009831750765442848,
0.01575247198343277,
-0.01678529381752014,
-0.02046475186944008,
-0.022434765473008156,
0.04140565171837807,
0.017567390576004982,
-0.03248656913638115,
-0.12521107494831085,
0.0034804523456841707,
-0.02... | 0.15723 |
👍 Simplicity and Clarity: Offers a straightforward and clear mapping of budget to disruption reason. \* 👎 Increased Configuration Overhead: Requires duplicating settings for multiple reasons, increasing setup complexity. \* 👎 Less Flexibility: Lacks the flexibility to share a budget across multiple reasons. #### Pro... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/disruption-controls-by-reason.md | main | karpenter | [
-0.05422457307577133,
0.00037008666549809277,
-0.007452890742570162,
-0.0008205089834518731,
0.10910461843013763,
0.015149497427046299,
0.03184865787625313,
0.07347628474235535,
0.06571140885353088,
0.08944709599018097,
0.04479644075036049,
-0.11084385216236115,
-0.0071379439905285835,
-0.... | 0.18152 |
\* 👎 Complexity in Default Reason Handling: The model does not facilitate easy defaulting for unspecified ("default") disruption reasons. \* 👎 Increased Budget Complexity: The use of budgets becomes more complex as they are now nested within another field, adding a layer of intricacy to their application. ### API Des... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/disruption-controls-by-reason.md | main | karpenter | [
-0.03678280860185623,
0.011663267388939857,
-0.0028820873703807592,
0.01693667471408844,
0.09986312687397003,
0.012070896103978157,
0.014383804984390736,
0.09522689878940582,
0.035964205861091614,
0.08767029643058777,
-0.009310327470302582,
-0.09630545973777771,
-0.013540724292397499,
0.01... | 0.199757 |
Nodes) are allowed ``` We could also allow a semantic like this ```yaml apiVersion: karpenter.sh/v1beta1 kind: NodePool metadata: name: worker-pool spec: # This is not a complete NodePool Spec. disruption: consolidationPolicy: WhenEmpty expireAfter: 7d budgets: - nodes: "100%" # Always allow Expiration and Emptiness to... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/disruption-controls-by-reason.md | main | karpenter | [
-0.03459838405251503,
0.04720573499798775,
-0.003222409402951598,
0.05604089796543121,
0.03946467116475105,
-0.007045698817819357,
0.03130464628338814,
-0.0035117659717798233,
0.03512469306588173,
0.08717075735330582,
0.05391813814640045,
-0.08134396374225616,
0.04679534211754799,
0.028333... | 0.160511 |
# Spot Consolidation ## Problem Statement Customers face increased costs due to underutilized spot nodes. A customer with 4 x 24cpu pods would launch a 24xl (96cpu) instance type. If three of the four pods complete, the node will indefinitely run at 25% reservation until the pod completes, is able to fit elsewhere in t... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/spot-consolidation.md | main | karpenter | [
0.007136038038879633,
-0.01868375577032566,
0.03914501145482063,
0.052225273102521896,
0.013467454351484776,
-0.02947940304875374,
-0.003797267097979784,
0.013656659051775932,
0.04793144762516022,
0.048256613314151764,
-0.010725481435656548,
0.005254815798252821,
0.036470577120780945,
-0.0... | 0.169635 |
specified in their NodePool. As a result, their initial launch would succeed but would get no consolidations after their initial launch. Conceptually, this approach is straightforward to explain to customers and directly aligned with Spot best practices. Some customers may be disappointed by the flexibility requirement... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/spot-consolidation.md | main | karpenter | [
-0.05363820493221283,
-0.04837249591946602,
0.030853882431983948,
0.03532324731349945,
0.0219259150326252,
0.01822446472942829,
-0.004478491377085447,
0.004997572395950556,
0.005042382515966892,
0.026458226144313812,
0.04206477105617523,
0.037811242043972015,
0.009342447854578495,
-0.02381... | 0.123324 |
best practices, and the customer should expect increased interruptions. See appendix for examples. \* 👍👍 Limits consolidation from regressing to a lowest price allocation strategy \* 👍 Extensible to new methods of capacity allocation, since Karpenter core is now completely agnostic to spot \* 👍 Prevents consolidati... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/spot-consolidation.md | main | karpenter | [
-0.03620895743370056,
-0.023650042712688446,
0.01687493547797203,
0.030177444219589233,
-0.016016760841012,
-0.014817170798778534,
0.021488286554813385,
0.03958575055003166,
0.025099260732531548,
0.022463252767920494,
-0.009441652335226536,
0.02905147150158882,
0.004337824881076813,
-0.031... | 0.10719 |
much longer than necessary \* 👎 Increased complexity of additional controls that interact with other disruption behaviors Note: Minimum node lifetime has other use cases, such minimizing disruption due to “consolidation delete“ actions for recently scheduled pods. We may need to explore similar solutions to this probl... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/spot-consolidation.md | main | karpenter | [
-0.008009438402950764,
-0.03901371732354164,
0.07329943776130676,
0.07013790309429169,
0.06076417118310928,
0.03501179441809654,
-0.001037599635310471,
0.019633950665593147,
0.017270663753151894,
0.048598166555166245,
-0.02482183836400509,
0.01233789324760437,
-0.002466937294229865,
-0.019... | 0.0533 |
# Do-Not-Disrupt Grace Period ## Summary This feature adds a grace period option to the `karpenter.sh/do-not-disrupt` pod annotation, allowing users to temporarily protect pods from disruption while ensuring that nodes can eventually be disrupted for cluster maintenance and updates. ## Motivation ### Problem Statement ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/do-not-disrupt-grace-period.md | main | karpenter | [
-0.03520550951361656,
0.05464202165603638,
0.08447591960430145,
0.01623174175620079,
0.061882104724645615,
-0.026528093963861465,
0.021055828779935837,
-0.045726291835308075,
0.02699919231235981,
0.0682002380490303,
-0.0007541446248069406,
0.00212016305886209,
-0.009987536817789078,
-0.045... | 0.285297 |
Protection against user error or application bugs - \*\*Scenario\*\*: A workload owner forgets to remove the annotation after job completion, or the application hangs - \*\*Grace Period\*\*: Declares "this should never run longer than X hours" - automatic cleanup without manual intervention - \*\*Why\*\*: Platform team... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/do-not-disrupt-grace-period.md | main | karpenter | [
-0.07953882962465286,
0.07828492671251297,
-0.013747071847319603,
0.018674468621611595,
0.04729918763041496,
-0.06533581018447876,
0.032966043800115585,
-0.042044833302497864,
-0.08737045526504517,
0.06420614570379257,
-0.03463861718773842,
-0.008125548250973225,
0.01587486080825329,
-0.05... | 0.227416 |
New function `IsDoNotDisruptActive(pod, clock)` - checks if protection is still active - Update `IsDisruptable(pod)` to use clock-aware checking 2. \*\*`pkg/controllers/state/statenode.go`\*\*: - Update pod disruptability validation to use clock parameter for testing 3. \*\*`pkg/controllers/disruption/types.go`\*\*: - ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/do-not-disrupt-grace-period.md | main | karpenter | [
-0.006991797126829624,
0.024824323132634163,
0.011038178578019142,
0.0180479995906353,
0.052531518042087555,
-0.049319077283144,
0.007841135375201702,
-0.031766634434461594,
-0.011378157883882523,
0.0665336325764656,
0.013577607460319996,
-0.007559634745121002,
-0.033279016613960266,
-0.04... | 0.201608 |
two conditions are met: - The pod's `do-not-disrupt` period elapses - The node's `terminationGracePeriod` elapses \*\*Example Scenario\*\*: ```yaml # NodePool with 72-hour eventual disruption spec: disruption: consolidationPolicy: WhenEmpty template: spec: terminationGracePeriod: 72h --- # Pod with 4-hour protection me... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/do-not-disrupt-grace-period.md | main | karpenter | [
-0.024478541687130928,
0.08731494098901749,
0.0728018507361412,
0.02126740850508213,
0.06021850183606148,
-0.027528777718544006,
0.024671554565429688,
-0.004195734858512878,
0.03663712739944458,
0.08705063909292221,
0.046646539121866226,
-0.026648782193660736,
-0.0005089001497253776,
-0.05... | 0.225481 |
scope\*\*: PodDisruptionBudget controls disruption policy but doesn't provide time-based expiration - \*\*Pragmatic leverage point\*\*: Karpenter is the natural place for node lifecycle controls to prevent indefinite resource blocking - As reviewer noted: "I understand the competing layers and how Karpenter ends up bei... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/do-not-disrupt-grace-period.md | main | karpenter | [
-0.044001832604408264,
0.01683412864804268,
0.08143151551485062,
-0.027991777285933495,
0.011226989328861237,
0.004633336793631315,
-0.05215765908360481,
-0.030351489782333374,
0.04020432382822037,
0.07790866494178772,
-0.00696600554510951,
-0.03667882829904556,
-0.011251335963606834,
-0.0... | 0.145116 |
# RFC: NodeRegistrationHealthy Status Condition on NodePool ## Motivation Karpenter may initiate the creation of nodes based on a NodePool configuration, but these nodes might fail to join the cluster due to unforeseen registration issues that Karpenter cannot anticipate or prevent. An example illustrating this issue i... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/noderegistrationhealthy-status-condition.md | main | karpenter | [
-0.007283337414264679,
-0.013443983159959316,
0.037264153361320496,
0.06506653130054474,
0.03277486935257912,
-0.0019865871872752905,
0.0333668552339077,
-0.04766891151666641,
-0.039044756442308426,
0.034537553787231445,
-0.034852489829063416,
-0.08194877207279205,
0.06351043283939362,
0.0... | 0.112397 |
(i.e. can remove the NodeRegistrationHealthy status condition from a NodePool if an external fix allows Nodes to register). ### Option 1: Track if a node was successfully launched using a NodePool - Recommended This option sets `NodeRegistrationHealthy` status condition on the nodePool by checking if the node launched ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/noderegistrationhealthy-status-condition.md | main | karpenter | [
-0.03019026480615139,
0.022922951728105545,
-0.008337484672665596,
0.07514144480228424,
0.10612093657255173,
-0.04293450713157654,
0.0394456647336483,
-0.09355931729078293,
-0.06471478939056396,
-0.0070160068571567535,
0.04709501564502716,
-0.04274347424507141,
0.05886932089924812,
-0.0131... | 0.027735 |
👎 This approach tends to get more complex as we need to scale with the cluster size. 4. 👎 Managing buffer entry expiration adds another layer of complexity. We need to ensure that a NodePool's unhealthy status doesn't persist indefinitely when no new launch attempts have occurred within a specified timeframe. ### How... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/noderegistrationhealthy-status-condition.md | main | karpenter | [
-0.001084672287106514,
0.010685349814593792,
0.002524343319237232,
0.05547233670949936,
0.06064903363585472,
0.023686103522777557,
0.02330704592168331,
-0.01206178031861782,
-0.029038332402706146,
0.04038294032216072,
-0.02854987047612667,
0.019292892888188362,
0.02683098241686821,
0.02291... | 0.13238 |
# Karpenter Drift Hash Versioning with CRD ## Summary of Static Drift Hashing Mechanism Drift utilizes a hashing mechanism to identify when a `NodeClaim` has been drifted from its owning `NodePool`. Karpenter will hash fields that are within the `NodePool.Spec.Template`, excluding [special cases of drift](https://karpe... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/drift-hash-versioning.md | main | karpenter | [
-0.0630992203950882,
0.01673872210085392,
0.0699809268116951,
-0.005841243080794811,
0.13960184156894684,
-0.02913535200059414,
0.0431491881608963,
-0.0032909747678786516,
0.07244367897510529,
0.034271933138370514,
0.013701696880161762,
-0.011082686483860016,
0.047997161746025085,
-0.04178... | 0.150879 |
already considered drifted will remain drifted if the `karpenter.sh/nodepool-hash-version` doesn't match Pros/Cons - 👍👍 Simple implementation effort - 👍 Covers general cases of updates to the `NodePool` CRD or the hashing mechanism - 👎 Nodes that were considered drifted before the Karpenter upgrade are not able to ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/drift-hash-versioning.md | main | karpenter | [
-0.05499626323580742,
-0.011323519982397556,
0.09785600006580353,
-0.01971661113202572,
0.11844772100448608,
-0.04256279394030571,
0.02110712043941021,
-0.01570817269384861,
0.03388352319598198,
0.03862302750349045,
0.017038444057106972,
-0.041260067373514175,
0.03883003443479538,
-0.04366... | 0.068503 |
propagate on to the NodeClaim to adopt this method for static drift checking - 👎👎 A high maintenance effort required in supporting individual field evaluation ## Recommendation The main goal we are trying to achieve for drift hash versioning is stability on node churn for clusters during a Karpenter upgrade with upda... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/drift-hash-versioning.md | main | karpenter | [
-0.0844392254948616,
0.03394409269094467,
0.06589775532484055,
-0.00712770689278841,
0.10237829387187958,
-0.028186095878481865,
0.026515211910009384,
-0.01534972619265318,
0.0035325137432664633,
0.04957849904894829,
-0.010893047787249088,
-0.03306609019637108,
0.04010063782334328,
-0.0202... | 0.114822 |
# Karpenter v1 API ## Overview In October 2023, the Karpenter maintainer team [released the beta version](https://aws.amazon.com/blogs/containers/karpenter-graduates-to-beta/) of the NodePool, NodeClaim, and EC2NodeClass APIs, promoting them on a trajectory toward stable (v1). After multiple months in beta, we have suf... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/v1-api.md | main | karpenter | [
-0.06710239499807358,
-0.001408895943313837,
0.10508263111114502,
-0.006623329594731331,
0.08341407775878906,
-0.011102925054728985,
-0.1047440692782402,
-0.03375571221113205,
-0.03884672746062279,
0.04245193675160408,
-0.05119830369949341,
-0.014585236087441444,
-0.027700740844011307,
-0.... | 0.130829 |
spec: template: metadata: labels: billing-team: my-team annotations: example.com/owner: "my-team" spec: nodeClassRef: group: karpenter.k8s.aws # Updated since only a single version will be served kind: EC2NodeClass name: default taints: - key: example.com/special-taint effect: NoSchedule startupTaints: - key: example.c... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/v1-api.md | main | karpenter | [
0.008112190291285515,
0.08333806693553925,
-0.010840930975973606,
0.028658924624323845,
0.06911122798919678,
0.004479129333049059,
-0.040388911962509155,
-0.026042820885777473,
0.07192596048116684,
0.005139537621289492,
-0.005240036640316248,
-0.12161753326654434,
-0.020743858069181442,
-0... | 0.1389 |
a max drain timeout](https://github.com/kubernetes-sigs/karpenter/issues/743) for their nodes. This avoids nodes staying stuck in a draining state on the cluster due to fully blocking PDBs or `karpenter.sh/do-not-disrupt` pods. Starting in v1, we will support a feature that will allow you to configure this max time (ca... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/v1-api.md | main | karpenter | [
-0.005068824160844088,
0.05897758528590202,
0.06790994852781296,
0.047554124146699905,
0.04825059324502945,
-0.024307534098625183,
-0.01777752675116062,
0.012579242698848248,
0.03194262459874153,
0.08683889359235764,
-0.0557866171002388,
-0.06228067725896835,
-0.03986106440424919,
-0.03813... | 0.175349 |
lastTransitionTime: "2024-02-02T19:54:34Z" reason: NodeClaimNotLaunched message: "NodeClaim hasn't succeeded launch" ``` #### Add `consolidateAfter` for `consolidationPolicy: WhenUnderutilized` \*\*Category:\*\* Stability The [Karpenter v1 Roadmap](./v1-roadmap.md) currently proposes the addition of the `consolidateAft... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/v1-api.md | main | karpenter | [
-0.04908464848995209,
-0.04635103791952133,
0.029265327379107475,
0.03750045597553253,
0.011317478492856026,
0.006573917344212532,
-0.03612465038895607,
0.0024774735793471336,
0.007755317725241184,
0.035727448761463165,
0.014431831426918507,
-0.017089491710066795,
0.006658589001744986,
-0.... | 0.153392 |
name: default taints: - key: example.com/special-taint effect: NoSchedule startupTaints: - key: example.com/another-taint effect: NoSchedule requirements: - key: "karpenter.k8s.aws/instance-category" operator: In values: ["c", "m", "r"] minValues: 2 resources: requests: cpu: "20" memory: "8192Mi" expireAfter: 720h | Ne... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/v1-api.md | main | karpenter | [
0.042228925973176956,
0.009429297409951687,
-0.021627826616168022,
-0.003188473405316472,
0.03941166028380394,
-0.017756029963493347,
-0.04439904913306236,
-0.004554274957627058,
0.08359836786985397,
0.05190722644329071,
0.013214482925832272,
-0.0815957710146904,
-0.039477381855249405,
-0.... | 0.148722 |
# Karpenter - Integration Testing RFC ## Summary Historically, the [`aws/karpenter-provider-aws`](https://github.com/aws/karpenter-provider-aws) repository has served as the primary testing ground for Karpenter's core functionality, with comprehensive integration tests covering provisioning, scheduling, disruption hand... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-integration-testing.md | main | karpenter | [
-0.06690863519906998,
-0.004508221056312323,
0.07303774356842041,
0.019414417445659637,
0.016484906896948814,
-0.0717439204454422,
-0.05502050742506981,
-0.045635826885700226,
0.030833644792437553,
0.036151058971881866,
-0.006478765979409218,
-0.11686348915100098,
0.021051175892353058,
-0.... | 0.155239 |
\* Environment: Kind clusters with KWOK Karpenter Controller for simulated node provisioning \* Test Coverage: \* NodePool Configuration \* Standalone NodeClaim Configuration \* Provisioning \* Scheduling \* Disruption Management \* Consolidation \* Drift (including NodePool changes) \* Expiration \* Emptiness \* Node ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-integration-testing.md | main | karpenter | [
-0.038499943912029266,
-0.01419724989682436,
0.06567221879959106,
0.058743517845869064,
0.07367699593305588,
-0.027379650622606277,
-0.019985783845186234,
-0.03651152178645134,
0.004900646395981312,
0.09559132903814316,
0.01574602909386158,
-0.07057590037584305,
0.033465176820755005,
-0.06... | 0.174093 |
Cases: \* should expire the node after the expiration is reached \* should replace expired node with a single node and schedule all pods ##### Integration \* High Level Description: Multiple functionality are tested. This is the catch all suite for testing karpenter \* Source: https://github.com/aws/karpenter-provider-... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-integration-testing.md | main | karpenter | [
-0.016235213726758957,
0.04655186086893082,
0.031453657895326614,
0.005769731011241674,
0.034703560173511505,
0.0028685571160167456,
0.0266990065574646,
-0.001175416517071426,
0.0013020663755014539,
0.06497058272361755,
0.042109690606594086,
-0.09869731962680817,
0.024305865168571472,
-0.0... | 0.216827 |
a pod has InitContainers (cpu) \* sidecar requirements + later init requirements do exceed container requirements \* sidecar requirements + later init requirements do not exceed container requirements \* init container requirements exceed all later requests \* should provision a right-sized node when a pod has InitCont... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-integration-testing.md | main | karpenter | [
-0.04653513431549072,
0.036406002938747406,
0.04439783841371536,
0.06235651671886444,
0.02354809083044529,
-0.06904257088899612,
-0.05116727948188782,
-0.039525095373392105,
-0.03649281710386276,
0.0634005144238472,
-0.012392951175570488,
-0.012639057822525501,
0.0004270664067007601,
-0.03... | 0.158831 |
# Capacity Reservations \_Note: This RFC pulls various excerpts from the [`aws/karpenter-provider-aws` On-Demand Capacity Reservation RFC](https://github.com/aws/karpenter-provider-aws/blob/main/designs/odcr.md).\_ ## Overview Cloud providers including GCP, Azure, and AWS allow you to pre-reserve VM (or bare-metal serv... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/capacity-reservations.md | main | karpenter | [
0.03206377476453781,
-0.03287310153245926,
0.017561839893460274,
0.02808937057852745,
-0.038294579833745956,
0.028017690405249596,
-0.03013501688838005,
-0.05271206796169281,
0.04717502370476723,
0.10491139441728592,
-0.02255156822502613,
-0.060479775071144104,
0.013784454204142094,
0.0264... | -0.008394 |
launch all capacity types by default requirements: [] ``` ### Reserved capacity type scheduling prioritization \_Note: Some excerpts taken from [`aws/karpenter-provider-aws` RFC](https://github.com/aws/karpenter-provider-aws/blob/main/designs/odcr.md#consider-odcr-first-during-scheduling).\_ Karpenter's current schedul... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/capacity-reservations.md | main | karpenter | [
-0.00651914207264781,
-0.010847612284123898,
0.047009795904159546,
0.041661955416202545,
0.03472738713026047,
-0.03747646510601044,
-0.0016415365971624851,
-0.04791417717933655,
0.04415469244122505,
0.09443527460098267,
-0.031402185559272766,
-0.04249868541955948,
-0.0018302190583199263,
0... | 0.073187 |
\_Note: Some excerpts taken from [`aws/karpenter-provider-aws` RFC](https://github.com/aws/karpenter-provider-aws/blob/main/designs/odcr.md#consolidation).\_ #### Consolidating into reserved capacity nodes Karpenter would need to update its consolidation algorithm to ensure that consolidating between a `spot` and/or `o... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/capacity-reservations.md | main | karpenter | [
-0.022816063836216927,
0.00263677048496902,
0.003050778293982148,
0.05276632308959961,
0.021904492750763893,
-0.023615945130586624,
-0.0011796619510278106,
-0.020593371242284775,
0.04050068184733391,
0.07080313563346863,
-0.03333340212702751,
-0.022971464321017265,
0.033535268157720566,
-0... | 0.089683 |
# KWOK DRA Driver Design ## Summary The upstream kubernetes/perf-tests repository includes a [DRA KWOK Driver](https://github.com/kubernetes/perf-tests/pull/3491/files), but it's designed for \*\*ClusterLoader2 scale testing\*\* with pre-created static nodes that cannot be used for Karpenter testing. This design introd... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-dra-kwok-driver.md | main | karpenter | [
-0.11216799914836884,
0.06054943427443504,
0.00686378451064229,
0.07126457989215851,
-0.0035020324867218733,
-0.05699104443192482,
-0.03428923338651657,
-0.007855074480175972,
0.06593040376901627,
0.00645920867100358,
-0.044578805565834045,
-0.08642774075269699,
0.004155005794018507,
-0.03... | 0.140747 |
6. \*\*Scheduler sees configured devices\*\*: ResourceSlices with fake devices become available for DRA pod scheduling 7. \*\*Test validation\*\*: Validates that the driver correctly provides DRA resources and enables successful pod scheduling ### Case 2: CRD-Based Fallback Configuration Tests \*\*DRA resource provisio... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/karpenter-dra-kwok-driver.md | main | karpenter | [
-0.07031606882810593,
0.01279684528708458,
0.071417436003685,
0.06599932163953781,
0.029248608276247978,
-0.05487397313117981,
-0.03965390846133232,
-0.040053460747003555,
0.004297296982258558,
0.015764938667416573,
-0.008366749621927738,
-0.01485655177384615,
-0.015454202890396118,
-0.010... | 0.144569 |
# Karpenter KwoK Cloud Provider The kubernetes-sigs/karpenter repo uses a simulated kubernetes environment to test code changes. Users must use a cloud provider of their choice to test their changes, which can be costly and creates an unreasonable barrier to entry for contributors. Related issue: https://github.com/kub... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/kwok-provider.md | main | karpenter | [
-0.056529443711042404,
-0.00047712374362163246,
0.09028098732233047,
0.017953259870409966,
-0.0000017624452084419318,
-0.017467427998781204,
0.017243599519133568,
-0.0249300729483366,
0.07205571979284286,
0.07344165444374084,
-0.014759520068764687,
-0.14006491005420685,
-0.001986258197575807... | 0.102133 |
= (#vCPU \* 0.25) + (#GiB \* 0.01) # Roughly calculated based on guessing pricing functions from multiple cloud providers. 2. on\_demand\_price = size \* price\_of\_base\_size() 3. spot\_price = on\_demand\_price \* .7 # Some discount factor to reflect savings somewhere in the middle of the referenced CPs. ``` | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/kwok-provider.md | main | karpenter | [
-0.010938270017504692,
0.023209091275930405,
-0.030743379145860672,
0.05537256971001625,
0.045471031218767166,
-0.007080092094838619,
0.052168745547533035,
0.08087415248155594,
0.08685269951820374,
0.07234547287225723,
0.035617582499980927,
-0.09943666309118271,
0.04444355145096779,
-0.053... | 0.078977 |
# Balanced Consolidation: Scoring Moves by Savings and Disruption ## Motivation Today, Karpenter's consolidation is all-or-nothing. `WhenEmptyOrUnderutilized` consolidates any node where pods can be repacked more cheaply, regardless of how little is saved or how many pods are disrupted. `WhenEmpty` consolidates only no... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
-0.0029254381079226732,
0.0030758578795939684,
0.042382821440696716,
0.06211010366678238,
0.010831593535840511,
-0.0386602021753788,
-0.004572201985865831,
-0.009058281779289246,
0.02107754908502102,
0.08485903590917587,
0.038905199617147446,
0.03071410395205021,
-0.0009798503015190363,
-0... | 0.146256 |
#2562](https://github.com/kubernetes-sigs/karpenter/pull/2562): `ConsolidationPriceImprovementFactor` field (0.0-1.0) with operator-level default and NodePool override. Cost Improvement Factor with a different UX. - [PR #2893](https://github.com/kubernetes-sigs/karpenter/pull/2893): Decision ratio with a configurable `... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
-0.006015781778842211,
-0.016253875568509102,
0.03661878779530525,
0.03551671281456947,
0.027176955714821815,
-0.022832786664366722,
0.027095435187220573,
0.05909699201583862,
0.04934992641210556,
0.1106841191649437,
-0.027712348848581314,
-0.06532788276672363,
0.02779870294034481,
-0.0030... | 0.178815 |
score normalizes savings and disruption against NodePool totals. ``` nodepool\_cost = sum(node.price for node in nodepool.nodes) nodepool\_total\_disruption\_cost = sum(node.disruption\_cost for node in nodepool.nodes) ``` Each node's disruption cost is 1.0 (per-node cost) plus `sum(max(0, EvictionCost(pod)))` for its ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
0.0495772585272789,
-0.035836249589920044,
0.028645377606153488,
0.05376439914107323,
0.11117760837078094,
-0.04588223993778229,
-0.0012038374552503228,
0.01984057016670704,
0.020021023228764534,
0.054439976811409,
-0.05698129162192345,
-0.06535506993532181,
0.0604194737970829,
-0.01291968... | 0.14165 |
show REPLACE and DELETE moves from a simulated cluster: 5000 pods with log-normal CPU/memory requests, packed across c7i/m7i/r7i instances, after 10 rounds of workload churn. Each curve shows cumulative savings vs. cumulative disruption under a different ranking strategy. At every disruption level, score-based ranking ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
0.031838905066251755,
-0.043105870485305786,
0.04966963827610016,
0.0683937817811966,
0.05846565216779709,
-0.03687683120369911,
-0.03329770267009735,
0.019817138090729713,
0.03405310586094856,
0.05734365060925484,
-0.0221937857568264,
-0.008205023594200611,
0.03972775116562843,
-0.0341950... | 0.161748 |
a secondary damping effect between rounds but is not sufficient on its own to prevent cycling. `consolidateAfter` determines candidacy: a node whose last pod event is within the `consolidateAfter` window is not a candidate. Non-candidate nodes still contribute to the denominators, preventing the post-replacement cycle ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
0.028484944254159927,
-0.017074765637516975,
0.07568269222974777,
0.08014227449893951,
0.06831043213605881,
-0.04128250852227211,
-0.02870144136250019,
0.009597030468285084,
0.027672335505485535,
0.00935242511332035,
-0.002807109383866191,
0.005700158886611462,
0.037632085382938385,
-0.094... | 0.154743 |
0.50 >= 0.5 --> approved ``` The replacement costs exactly half the original. In a uniform pool, this is the boundary at k=2: a replace that saves less than half the source node's cost is rejected. In a heterogeneous pool, the threshold depends on pool-level fractions, not per-node percentage. At k=1, the score simplif... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
0.039530202746391296,
0.047275058925151825,
0.05402841418981552,
0.048734791576862335,
0.11132380366325378,
-0.06142600625753403,
-0.014108900912106037,
0.004759841598570347,
0.005017316900193691,
0.06084216386079788,
-0.07691320776939392,
-0.06579134613275528,
0.0801859050989151,
-0.02858... | 0.045461 |
exhaustive enumeration. ### State Space We enumerate configurations in a bounded space using on-demand prices from three instance families in us-east-1: c7i (compute-optimized), m7i (general-purpose), and r7i (memory-optimized), medium through 4xlarge (15 price points). 1 to 6 nodes per pool, 0 to 4 pods per node, per-... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
-0.01978646218776703,
-0.011313234455883503,
0.0038400194607675076,
0.019453423097729683,
0.0752897784113884,
0.03697153180837631,
0.02663145586848259,
0.022741762921214104,
-0.004585973918437958,
0.037883076816797256,
-0.06895262002944946,
-0.07764744758605957,
0.05188513919711113,
-0.013... | 0.167462 |
viable, and it captures all cross-family pairs where the replacement costs less than half the original. The 8 additional cross-family pairs at k=3 are available to operators who set `consolidationPolicy: 3` (see [`balanced-consolidation-properties.py`](scripts/balanced-consolidation-properties.py)). ## API Choices ### ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
-0.11111199110746384,
-0.02754884772002697,
-0.06471677869558334,
-0.04030166193842888,
-0.10341085493564606,
-0.047891031950712204,
0.02451401762664318,
0.007899980992078781,
-0.0242777019739151,
-0.04748872295022011,
0.006833059713244438,
0.046826642006635666,
0.023597655817866325,
-0.03... | 0.110217 |
the per-cycle timeout problem. ## Frequently Asked Questions ### What happens in a uniformly inefficient cluster where no single REPLACE clears the threshold? Every REPLACE scores low (e.g., 0.2 for a one-tier downsize) and is rejected. DELETEs still work: a DELETE of a node whose pods fit elsewhere scores 1.0 in a uni... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/balanced-consolidation.md | main | karpenter | [
0.013918671756982803,
0.006404984276741743,
0.06135953217744827,
0.021157370880246162,
0.01663460209965706,
0.0011142583098262548,
-0.0738450363278389,
-0.050303928554058075,
0.06529621034860611,
0.011444656178355217,
-0.06279786676168442,
0.011686600744724274,
0.05150436982512474,
-0.0413... | 0.105501 |
# Karpenter - Node Overlay ## Background Node launch decisions are the output a scheduling simulation algorithm that involves satisfying pod scheduling constraints with available instance type offerings. Cloud provider implementations are responsible for surfacing these offerings to the Karpenter scheduler through the ... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/node-overlay.md | main | karpenter | [
-0.027048593387007713,
0.03158750757575035,
0.0746164619922638,
0.035088710486888885,
0.0629214346408844,
-0.055495407432317734,
-0.03956986218690872,
0.010316873900592327,
0.09743435680866241,
0.10025777667760849,
-0.07017428427934647,
-0.05823352187871933,
0.0086439223960042,
-0.07070638... | 0.10909 |
signed float representing the price override ``` kind: NodeOverlay metadata: name: price-override spec: requirements: - key: karpenter.sh/capacity-type operator: In values: ["on-demand"] # Examples of price override: price: "5.00" # Set price to $5.00 ``` \*\*Price Adjustment:\*\* Define a pricing adjustment for instan... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/node-overlay.md | main | karpenter | [
-0.06098978593945503,
0.03982303664088249,
-0.011623035185039043,
0.01940334029495716,
-0.03268031030893326,
-0.0337052159011364,
-0.01622062362730503,
0.058357588946819305,
0.035301342606544495,
0.04956415668129921,
-0.021704357117414474,
-0.088068887591362,
-0.011679382063448429,
-0.0240... | 0.123639 |
understand and reason about, particularly when NodeOverlays have intersecting requirements for the same scheduling dimensions. ## Node Overlay Field Behavior and Consolidation Integration Node Overlay introduces three key fields that affect Karpenter's behavior: `priceAdjustment`, `price` and `capacity`. The `priceAdju... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/node-overlay.md | main | karpenter | [
-0.012399381026625633,
-0.009131355211138725,
0.0674731433391571,
0.05331350490450859,
0.018401870504021645,
-0.0638437569141388,
-0.011401072144508362,
-0.012495321221649647,
0.04067368432879448,
0.03099096193909645,
-0.029260654002428055,
-0.003196685342118144,
-0.0004874605219811201,
-0... | 0.166642 |
different tradeoffs for system reliability and operator experience. \*\*Fail Open Approach (recommended):\*\* \* Using alphabetical ordering to resolve conflicts when weights are equal \* Setting overlay Ready status condition to false when conflicts occur \* Adding status conditions to inform users that the overlays a... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/node-overlay.md | main | karpenter | [
-0.0028648243751376867,
0.017102235928177834,
0.06486574560403824,
0.05849340930581093,
-0.004860156215727329,
-0.06525402516126633,
-0.03691728785634041,
0.00006227255653357133,
0.019945470616221428,
0.04319031909108162,
0.005769162438809872,
0.021526042371988297,
0.01764482446014881,
-0.... | 0.132167 |
of gradual rollouts. In contrast, the separate NodeOverlay CRD provides better separation of concerns, more flexible configuration options through overlay stacking, and a more maintainable system that can better adapt to future needs while keeping the NodePool API focused on its core provisioning responsibilities. ## O... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/node-overlay.md | main | karpenter | [
0.010227967984974384,
0.023457106202840805,
0.03478916734457016,
0.026592714712023735,
0.07843527942895889,
-0.04017804563045502,
-0.04531410336494446,
0.03299126401543617,
-0.016678061336278915,
0.029002780094742775,
-0.04981622472405434,
-0.03017318807542324,
-0.0021655219607055187,
-0.0... | 0.102161 |
## Designs The designs in this folder are merged RFCs. They are historical artifacts that demonstrate the designs of features when they were implemented. They are not kept up to date as changes are made. ## Documentation These documents are also not the source truth for the karpenter.sh website. The source for that liv... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/README.md | main | karpenter | [
-0.07565862685441971,
0.08604888617992401,
-0.0002302482316736132,
-0.004175938665866852,
0.03328773379325867,
0.020630519837141037,
-0.041724324226379395,
0.002221619477495551,
0.03882059082388878,
0.04584822431206703,
-0.037466004490852356,
-0.02240535244345665,
0.0037436680868268013,
-0... | 0.184114 |
# NodePool Requirement Flexibility \*\*NodeClaim\*\*: Karpenter’s representation of a Node request. A NodeClaim declaratively describes how the node should look, including instance type requirements that constrain the options that the Cloud Provider can launch with. For the AWS Cloud Provider, the instance requirements... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/nodepool-requirement-flexibility.md | main | karpenter | [
-0.025332745164632797,
0.005994230043143034,
0.07706290483474731,
0.009526800364255905,
0.03877009078860283,
-0.010177651420235634,
0.013492333702743053,
-0.04491127282381058,
0.041384562849998474,
0.08079005032777786,
-0.05155966058373451,
-0.07296991348266602,
0.030554506927728653,
0.002... | 0.125409 |
the future, we may have to break our `minValues` API ### Solution 2: `minValues` in NodeClass API ``` apiVersion: karpenter.sh/v1beta1 kind: EC2NodeClass spec: minValues: node.kubernetes.io/instance-type: 10 karpenter.kwok.sh/instance-family: 2 --- apiVersion: karpenter.sh/v1beta1 kind: NodePool spec: template: spec: r... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/nodepool-requirement-flexibility.md | main | karpenter | [
-0.06007510796189308,
0.07541273534297943,
0.048103224486112595,
0.019036035984754562,
0.007654796354472637,
0.05232534557580948,
-0.009579773060977459,
-0.020828155800700188,
0.08019109815359116,
0.022241367027163506,
-0.012452800758183002,
-0.1203666403889656,
0.014070327393710613,
-0.03... | 0.077832 |
Imagine conflicting requirements such as: ``` requirements: - key: karpenter.kwok.sh/instance-generation operator: Exists flexibility: 1 - key: karpenter.kwok.sh/instance-category operator: Exists flexibility: 2 ``` If our instance type options remaining are : [“c4.large”, “r5.large”], it’s not possible to remove an in... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/nodepool-requirement-flexibility.md | main | karpenter | [
-0.06771716475486755,
0.06876432150602341,
0.06473664939403534,
0.02781880460679531,
0.02294650487601757,
-0.008501064032316208,
0.018113557249307632,
0.017735488712787628,
-0.028003312647342682,
-0.04580710083246231,
-0.0008766434038989246,
-0.08631162345409393,
-0.011129669845104218,
0.0... | 0.028518 |
consolidated down to one of the bottom 10 instance types, you would no longer be able to continue consolidating. ### How does `minValues` interact with spot-to-spot consolidation? [Spot-to-spot consolidation already requires a spot instance flexibility of 15](https://github.com/kubernetes-sigs/karpenter/blob/main/desig... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/nodepool-requirement-flexibility.md | main | karpenter | [
0.023863395676016808,
-0.055913470685482025,
0.0671384409070015,
0.0011954159708693624,
-0.045189518481492996,
0.0015282952226698399,
0.013034860603511333,
-0.007841563783586025,
0.0053605809807777405,
-0.024979956448078156,
-0.04532334953546524,
-0.028944609686732292,
0.023164881393313408,
... | 0.074531 |
# Karpenter - Static Capacity ## Background Karpenter currently operates as a dynamic cluster autoscaler, automatically adjusting node counts based on pending pod demand. However, several important use cases require maintaining a fixed set of nodes such as: 1. Performance-critical applications where just-in-time provis... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/static-capacity.md | main | karpenter | [
0.008607663214206696,
-0.010106903500854969,
0.004294568207114935,
0.06497152894735336,
-0.0064241159707307816,
-0.044326409697532654,
-0.020226377993822098,
-0.015087599866092205,
-0.03078354336321354,
0.07352389395236969,
-0.052018508315086365,
-0.07773586362600327,
-0.006853214465081692,
... | 0.235891 |
attempt to create a fixed number of NodeClaims (replicas). Limits would start to dictate instance shape, not just aggregate capacity. This creates a conflict between honoring replicas vs. satisfying hard limits. However, we are introducing a new field, limits.nodes, to explicitly cap the maximum number of nodes which c... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/static-capacity.md | main | karpenter | [
-0.05799851939082146,
-0.014744930900633335,
0.0678071528673172,
0.021791094914078712,
0.04238402843475342,
-0.022410893812775612,
0.0068626743741333485,
-0.0005756787140853703,
0.033690255135297775,
0.06601102650165558,
-0.04595442861318588,
-0.054583679884672165,
0.03174880892038345,
0.0... | 0.158911 |
These are evaluated once per NodeClaim at creation, meaning the selection is based solely on what the template allows. As a result, even though all NodeClaims come from the same static NodePool, they may still result in different instance types (shapes/flavors), depending on availability, since that decision happens du... | https://github.com/kubernetes-sigs/karpenter/blob/main/designs/static-capacity.md | main | karpenter | [
-0.03847244754433632,
-0.0005903649143874645,
0.06157613545656204,
0.01439395360648632,
0.07737677544355392,
-0.0004779493610840291,
-0.009507374837994576,
-0.04807078093290329,
0.06709051877260208,
0.039284415543079376,
-0.03319777548313141,
-0.08383039385080338,
0.01847650855779648,
0.03... | 0.131921 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.