如何使用NodeJ和Bazel为gRPC设置Kubernetes运行状况检查？ grpc-health-probe 1将grpc-health-probe二进制文件添加到Bazel 2实施服务 3使用Bazel构建和部署NodeJs映像 4向Kubernetes添加活跃性和就绪性探针

问题描述

如何使用Node.js和Bazel为gRPC设置GRPC Health Checking？

解决方法

1。将grpc-health-probe二进制文件添加到Bazel

在您的WORKSPACE文件中添加

load("@bazel_tools//tools/build_defs/repo:http.bzl","http_file")

http_file(
    name = "grpc_health_check_bin",downloaded_file_path = "grpc_health_probe",urls = ["https://github.com/grpc-ecosystem/grpc-health-probe/releases/download/v0.3.2/grpc_health_probe-linux-amd64"],)

下载可执行二进制文件。

2。实施服务

选项1：使用grpc-health-check npm模块

• 只需阅读文档即可：）

选项2 自己实施卫生服务

1. 转到this页并复制health-checking.proto文件，该文件当前如下所示：

   syntax = "proto3";
   
   package grpc.health.v1;
   
   message HealthCheckRequest {
     string service = 1;
   }
   
   message HealthCheckResponse {
     enum ServingStatus {
       UNKNOWN = 0;
       SERVING = 1;
       NOT_SERVING = 2;
       SERVICE_UNKNOWN = 3;  // Used only by the Watch method.
     }
     ServingStatus status = 1;
   }
   
   service Health {
     rpc Check(HealthCheckRequest) returns (HealthCheckResponse);
   
     rpc Watch(HealthCheckRequest) returns (stream HealthCheckResponse);
   }
   

2. 在您的BUILD.bazel文件中，创建带有proto文件的文件组，以便我们以后可以将其添加到NodeJs映像中

   filegroup(
       name = "health_checking_proto",srcs = ["health-checking.proto"],)
   

3. 实施卫生服务

   import * as protoLoader from '@grpc/proto-loader'
   import * as grpc from '@grpc/grpc-js'
   
   async function main() {
     const packageDefinition = await protoLoader.load('health-checking.proto')
     const grpcObject = grpc.loadPackageDefinition(packageDefinition)
     const {service} = (grpcObject.grpc as any).health.v1.Health
   
     const server = new grpc.Server()
     const implementation = {
       // status can be on of UNKNOWN,SERVING,NOT_SERVING
       Check: (_call: any,callback: any) => callback(null,{status: 'SERVING'}),}
     server.addService(service,implementation)
   
     server.bindAsync('0.0.0.0:9090',grpc.ServerCredentials.createInsecure(),() => server.start())
   }
   
   main()
   

3。使用Bazel构建和部署NodeJs映像

最终的BUILD文件可能如下所示：

load("@npm//@bazel/typescript:index.bzl","ts_library")
load("@k8s_deploy//:defaults.bzl","k8s_deploy")
load("@io_bazel_rules_docker//nodejs:image.bzl","nodejs_image")

package(default_visibility = ["//visibility:public"])

ts_library(
    name = "lib",srcs = glob(include = ["**/*.ts"]),deps = [
        "@npm//@grpc/grpc-js","@npm//@grpc/proto-loader","@npm//@types/node",],)

filegroup(
    name = "health_checking_proto",)

nodejs_image(
    name = "image",data = [
        # nodejs application
        ":lib",# health-checking.proto file
        ":health_checking_proto",# grpc-health-probe executable binary
        "@grpc_health_check_bin//file",entry_point = ":index.ts",)

k8s_deploy(
    name = "k8s",images = {"k8s:placeholder_name": ":image"},template = ":k8s.yaml",)

4。向Kubernetes添加活跃性和就绪性探针

spec:
  containers:
  - name: server
    image: "[YOUR-DOCKER-IMAGE]"
    ports:
    - containerPort: 9090
    readinessProbe:
      exec:
        command: ["/app/<path to dir with BUILD file>/image.binary.runfiles/grpc_health_check_bin/file/grpc_health_probe","-addr=:9090"]
      initialDelaySeconds: 5
    livenessProbe:
      exec:
        command: ["/app/<path to dir with BUILD file>/image.binary.runfiles/grpc_health_check_bin/file/grpc_health_probe","-addr=:9090"]
      initialDelaySeconds: 10