在 Sloth Prometheus 中设置自定义 SLO 周期窗口 (7d) 时出错

Question

我正在尝试按照文档 https://sloth.dev/usage/slo-period-windows/ 将 SLO 周期窗口设置为 7d

图片：ghcr.io/slok/sloth:v0.11.0

在启用和不启用 customSloconfig 的情况下使用 helm (values.yaml) 时：

sloth:
  defaultSloPeriod: 7d  # The slo period used by sloth (e.g. 30d).

customSloConfig:
  enabled: true
  path: /windows
  data: {}

它出现错误： 未找到 AlertWindows.sloth.slok.dev“”

直接与清单::

apiVersion: sloth.slok.dev/v1
kind: AlertWindows
spec:
  sloPeriod: 7d
  page:
    quick:
      errorBudgetPercent: 8
      shortWindow: 5m
      longWindow: 1h
    slow:
      errorBudgetPercent: 12.5
      shortWindow: 30m
      longWindow: 6h
  ticket:
    quick:
      errorBudgetPercent: 20
      shortWindow: 2h
      longWindow: 1d
    slow:
      errorBudgetPercent: 42
      shortWindow: 6h
      longWindow: 3d

kubectl apply -f window.yaml

它出现错误：

错误：找不到名称的资源映射：“sloth-period-window”命名空间：“”来自“window.yaml”：版本“sloth.slok.dev/v1”中的种类“AlertWindows”没有匹配项 确保首先安装 CRD

我尝试为 AlertWindow (alertwindows.sloth.slok.dev) 创建 CRD，但还没有成功。任何人都有正确的 CRD，如果这是解决它的最佳方法或其他帮助，请问？

预计 pod 与日志一起工作：

INFO[0001] Hot-reload triggered from http webhook        version=v0.11.0 window=7d

Answer 1

已经解决了，如果有人需要的话。

使用掌舵： 在 values.yaml 中如下设置这些变量

labelSelector: slo-window=7d
defaultSloPeriod: 7d

和

customSloConfig:
  enabled: true
  path: /windows
  data: 
    apiVersion: sloth.slok.dev/v1
    kind: AlertWindows
    spec:
      sloPeriod: 7d
      page:
        quick:
          errorBudgetPercent: 1
          shortWindow: 2m
          longWindow: 30m
        slow:
          errorBudgetPercent: 2
          shortWindow: 15m
          longWindow: 3h
      ticket:
        quick:
          errorBudgetPercent: 5
          shortWindow: 1h
          longWindow: 12h
        slow:
          errorBudgetPercent: 5
          shortWindow: 3h
          longWindow: 36h

我创建了 CRD，使用它：

---
apiVersion: apiextensions.k8s.io/v1
kind: CustomResourceDefinition
metadata:
  annotations:
    controller-gen.kubebuilder.io/version: (devel)
  creationTimestamp: null
  name: alertwindows.sloth.slok.dev
spec:
  group: sloth.slok.dev
  names:
    categories:
    - saw
    kind: AlertWindows
    listKind: AlertWindowsList
    plural: alertwindows
    shortNames:
    - saw
    singular: alertwindow
  scope: Namespaced
  versions:
  - additionalPrinterColumns:
    - jsonPath: .spec.sloPeriod
      name: SLOPERIOD
      type: string
    - jsonPath: .metadata.creationTimestamp
      name: AGE
      type: date
    name: v1
    schema:
      openAPIV3Schema:
        description: AlertWindows....
        properties:
          apiVersion:
            description: 'APIVersion defines the versioned schema of this representation
              of an object. Servers should convert recognized schemas to the latest
              internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources'
            type: string
          kind:
            description: 'Kind is a string value representing the REST resource this
              object represents. Servers may infer this from the endpoint the client
              submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds'
            type: string
          metadata:
            type: object
          spec:
            description: spec for Alert Windows.
            type: object
            properties:
              labels:
                additionalProperties:
                  type: string
                description: Labels are....
                type: object
              sloPeriod: 
                description: period.
                type: string
              page:
                description: xxx
                type: object
                properties:  
                  quick:
                    description: xxx
                    type: object
                    properties:
                      errorBudgetPercent:
                        type: integer
                      shortWindow:
                        type: string
                      longWindow:
                        type: string 
                  slow:
                    description: xxx
                    type: object
                    properties:
                      errorBudgetPercent:
                        type: number
                      shortWindow:
                        type: string
                      longWindow:
                        type: string                              
              ticket:
                description: xxx
                type: object
                properties:  
                  quick:
                    description: xxx
                    type: object
                    properties:
                      errorBudgetPercent:
                        type: integer
                      shortWindow:
                        type: string
                      longWindow:
                        type: string 
                  slow:
                    description: xxx
                    type: object
                    properties:
                      errorBudgetPercent:
                        type: integer
                      shortWindow:
                        type: string
                      longWindow:
                        type: string     
          status:
            properties:
              lastPromOpRulesSuccessfulGenerated:
                description: LastPromOpRulesGeneration tells the last atemp made for
                  a successful SLO rules generate.
                format: date-time
                type: string
              observedGeneration:
                description: ObservedGeneration tells the generation was acted on,
                  normally this is required to stop an infinite loop when the status
                  is updated because it sends a watch updated event to the watchers
                  of the K8s object.
                format: int64
                type: integer
              processedSLOs:
                description: ProcessedSLOs tells how many SLOs haven been processed
                  for Prometheus operator.
                type: integer
              promOpRulesGenerated:
                description: PromOpRulesGenerated tells if the rules for prometheus
                  operator CRD have been generated.
                type: boolean
              promOpRulesGeneratedSLOs:
                description: PromOpRulesGeneratedSLOs tells how many SLOs have been
                  processed and generated for Prometheus operator successfully.
                type: integer
            required:
            type: object
        type: object
    served: true
    storage: true
    subresources:
      status: {}

在 PrometheusServiceLevel 文件 (SLO) 中，有必要添加此标签：

  labels:
    slo-window: 7d 

希望它能帮助别人。