5.5. 从分布式追踪平台(Jaeger)迁移到分布式追踪数据收集

流程

1. 将 OpenTelemetry Collector 配置为 sidecar。

   apiVersion: opentelemetry.io/v1alpha1
   kind: OpenTelemetryCollector
   metadata:
     name: otel
     namespace: <otel-collector-namespace>
   spec:
     mode: sidecar
     config: |
       receivers:
         jaeger:
           protocols:
             grpc:
             thrift_binary:
             thrift_compact:
             thrift_http:
       processors:
         batch:
         memory_limiter:
           check_interval: 1s
           limit_percentage: 50
           spike_limit_percentage: 30
         resourcedetection:
           detectors: [openshift]
           timeout: 2s
       exporters:
         otlp:
           endpoint: "tempo-<example>-gateway:8090" 1
           tls:
             insecure: true
       service:
         pipelines:
           traces:
             receivers: [jaeger]
             processors: [memory_limiter, resourcedetection, batch]
             exporters: [otlp]

   1

   此端点指向使用 Tempo Operator 部署的 <example> TempoStack 实例的网关。

2. 创建用于运行应用程序的服务帐户。

   apiVersion: v1
   kind: ServiceAccount
   metadata:
     name: otel-collector-sidecar

3. 为某些处理器所需的权限创建集群角色。

   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRole
   metadata:
     name: otel-collector-sidecar
   rules:
     1
   - apiGroups: ["config.openshift.io"]
     resources: ["infrastructures", "infrastructures/status"]
     verbs: ["get", "watch", "list"]

   1

   resourcedetectionprocessor 需要基础架构和基础架构/状态的权限。

4. 创建 ClusterRoleBinding 来为服务帐户设置权限。

   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:
     name: otel-collector-sidecar
   subjects:
   - kind: ServiceAccount
     name: otel-collector-deployment
     namespace: otel-collector-example
   roleRef:
     kind: ClusterRole
     name: otel-collector
     apiGroup: rbac.authorization.k8s.io

5. 将 OpenTelemetry Collector 部署为 sidecar。
6. 通过从 Deployment 对象中删除 "sidecar.jaegertracing.io/inject": "true" 注解，从应用程序中删除注入的 Jaeger Agent。
7. 通过将 sidecar.opentelemetry.io/inject: "true" 注解添加到 Deployment 对象的 .spec.template.metadata.annotations 字段来启用 OpenTelemetry sidecar 自动注入。
8. 使用为应用程序部署创建的服务帐户，以允许处理器获取正确的信息并将其添加到您的追踪中。

流程

1. 配置 OpenTelemetry Collector 部署。

2. 创建部署 OpenTelemetry Collector 的项目。

   apiVersion: project.openshift.io/v1
   kind: Project
   metadata:
     name: observability

3. 创建用于运行 OpenTelemetry Collector 实例的服务帐户。

   apiVersion: v1
   kind: ServiceAccount
   metadata:
     name: otel-collector-deployment
     namespace: observability

4. 创建集群角色，以设置处理器所需的权限。

   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRole
   metadata:
     name: otel-collector
   rules:
     1
     2
   - apiGroups: ["", "config.openshift.io"]
     resources: ["pods", "namespaces", "infrastructures", "infrastructures/status"]
     verbs: ["get", "watch", "list"]

   1

   k8sattributesprocessor 需要 pods 和 namespaces 资源的权限。

   2

   resourcedetectionprocessor 需要 infrastructures 和 infrastructures/status 的权限。

5. 创建 ClusterRoleBinding 来为服务帐户设置权限。

   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:
     name: otel-collector
   subjects:
   - kind: ServiceAccount
     name: otel-collector-deployment
     namespace: observability
   roleRef:
     kind: ClusterRole
     name: otel-collector
     apiGroup: rbac.authorization.k8s.io

6. 创建 OpenTelemetry Collector 实例。

   注意

   此收集器会将 trace 导出至 TempoStack 实例。您必须使用 Red Hat Tempo Operator 创建 TempoStack 实例，并放在正确的端点中。

   apiVersion: opentelemetry.io/v1alpha1
   kind: OpenTelemetryCollector
   metadata:
     name: otel
     namespace: observability
   spec:
     mode: deployment
     serviceAccount: otel-collector-deployment
     config: |
       receivers:
         jaeger:
           protocols:
             grpc:
             thrift_binary:
             thrift_compact:
             thrift_http:
       processors:
         batch:
         k8sattributes:
         memory_limiter:
           check_interval: 1s
           limit_percentage: 50
           spike_limit_percentage: 30
         resourcedetection:
           detectors: [openshift]
       exporters:
         otlp:
           endpoint: "tempo-example-gateway:8090"
           tls:
             insecure: true
       service:
         pipelines:
           traces:
             receivers: [jaeger]
             processors: [memory_limiter, k8sattributes, resourcedetection, batch]
             exporters: [otlp]

7. 将追踪端点指向 OpenTelemetry Operator。

8. 如果您要将 trace 直接从应用程序导出到 Jaeger，请将 API 端点从 Jaeger 端点改为 OpenTelemetry Collector 端点。

   使用带有 Golang 的 jaegerexporter 导出 trace 的示例

   exp, err := jaeger.New(jaeger.WithCollectorEndpoint(jaeger.WithEndpoint(url))) 1

   1

   URL 指向 OpenTelemetry Collector API 端点。