通过kuberay部署的Rayservice如何添加readiness和iveness

show · 2023 年4 月 20 日 02:41

通过kuberay部署的Rayservice，head和worker如何添加readiness和iveness，保证服务的健康检查
kuberay rayservice
https://ray-project.github.io/kuberay/guidance/rayservice/

Qstar · 2023 年4 月 25 日 12:52

Rayservice Yaml 中包含有 rayClusterConfig，其中 template 部分可以根据需求设置 readiness/liveness probe：以如下官方示例查看

github.com

ray-project/kuberay/blob/master/ray-operator/config/samples/ray_v1alpha1_rayservice.yaml

# Make sure to increase resource requests and limits before using this example in production.
# For examples with more realistic resource configuration, see
# ray-cluster.complete.large.yaml and
# ray-cluster.autoscaler.large.yaml.
apiVersion: ray.io/v1alpha1
kind: RayService
metadata:
  name: rayservice-sample
spec:
  serviceUnhealthySecondThreshold: 300 # Config for the health check threshold for service. Default value is 60.
  deploymentUnhealthySecondThreshold: 300 # Config for the health check threshold for deployments. Default value is 60.
  serveConfig:
    importPath: fruit.deployment_graph
    runtimeEnv: |
      working_dir: "https://github.com/ray-project/test_dag/archive/41d09119cbdf8450599f993f51318e9e27c59098.zip"
    deployments:
      - name: MangoStand
        numReplicas: 1
        userConfig: |
          price: 3

This file has been truncated. show original

在官方提供的示例中，https://github.com/ray-project/kuberay/blob/master/ray-operator/controllers/ray/common/pod.go#L242 部分以代码形式描述了检测手段，开启的方式是通过配置 Annotation(ray.io/ft-enabled): kuberay/pod.go at master · ray-project/kuberay · GitHub

检测的方式是通过检查 raylet 的端口服务，相关代码部分：kuberay/constant.go at master · ray-project/kuberay · GitHub

感谢关注 Ray 中文社区~~~

show · 2023 年4 月 26 日 06:45

感觉说了很多，官方示例/代码形式检测手段/配置annotation/检查raylet端口服务，但我好像还是不太明白在RayService的yaml里面应该怎么配置，是serviceUnhealthySecondThreshold/deploymentUnhealthySecondThreshold这两个值吗
谢谢