21.4. 运行性能分析查询

流程

1. 创建名为 nodeobservabilityrun.yaml 的 NodeObservabilityRun 资源文件，其中包含以下文本：

   apiVersion: nodeobservability.olm.openshift.io/v1alpha2
   kind: NodeObservabilityRun
   metadata:
     name: nodeobservabilityrun
   spec:
     nodeObservabilityRef:
       name: cluster

   Copy to Clipboard Toggle word wrap

2. 运行 NodeObservabilityRun 资源来触发性能分析查询：

   $ oc apply -f nodeobservabilityrun.yaml

   Copy to Clipboard Toggle word wrap

3. 运行以下命令，检查 NodeObservabilityRun 的状态：

   $ oc get nodeobservabilityrun nodeobservabilityrun -o yaml  | yq '.status.conditions'

   Copy to Clipboard Toggle word wrap

   输出示例

   conditions:
   - lastTransitionTime: "2022-07-07T14:57:34Z"
     message: Ready to start profiling
     reason: Ready
     status: "True"
     type: Ready
   - lastTransitionTime: "2022-07-07T14:58:10Z"
     message: Profiling query done
     reason: Finished
     status: "True"
     type: Finished

   Copy to Clipboard Toggle word wrap

   分析查询在状态变为 True 后完成，类型为 Finished。

4. 通过运行以下 bash 脚本，从容器的 /run/node-observability 路径中检索配置集数据：

   for a in $(oc get nodeobservabilityrun nodeobservabilityrun -o yaml | yq .status.agents[].name); do
     echo "agent ${a}"
     mkdir -p "/tmp/${a}"
     for p in $(oc exec "${a}" -c node-observability-agent -- bash -c "ls /run/node-observability/*.pprof"); do
       f="$(basename ${p})"
       echo "copying ${f} to /tmp/${a}/${f}"
       oc exec "${a}" -c node-observability-agent -- cat "${p}" > "/tmp/${a}/${f}"
     done
   done

   Copy to Clipboard Toggle word wrap