HUAWEI CLOUD ASM¶

The link tracking data from HUAWEI CLOUD ASM is exported to Guance for viewing and analysis.

Config¶

Preparatory work¶

The prerequisite for using ASM is to have purchased a CCE (Cloud Container Engine) cluster.Deploy the datakit daemonset.

Creating ASM¶

Service Mesh -> Purchasing a Mesh -> Basic Edition

Grid name," select the "Istio" version, choose the cluster version, and the control nodes for "Istio", then submit directly.

After approximately 1-3 minutes, the installed page will displays the running status of ASM, billing methods, and the number and status of clusters, instances, and grayscale tasks.

After the installation is completed, create the "bookinfo" application.

Before installation, you need to prepare a load balancer. Here, I have already prepared it. You can select it, choose port 80 for external access, and select the Docker image repository address as docker.io/istio. Then, proceed with the installation.

After that, it will view the created services and gateways in chart of 'service management' and 'gateway management'.

Accessing the external access address (such as http://external-access-address-ip/productpage) to check if the service is functioning properly.

Sending trace data to the Guance¶

Enable the OpenTelemetry collector¶

ReferenceOpenTelemetry Documentation for Integrating the Collector

ConfigMap Add

opentelemetry.conf: |
      [[inputs.opentelemetry]]
          [inputs.opentelemetry.http]
            enable = true
            http_status_ok = 200
            trace_api = "/otel/v1/trace"
            metric_api = "/otel/v1/metric"
          [inputs.opentelemetry.grpc]
            trace_enable = true
            metric_enable = true
            addr = "0.0.0.0:4317"

Mount opentelemetry.conf

- mountPath: /usr/local/datakit/conf.d/opentelemetry/opentelemetry.conf
    name: datakit-conf
    subPath: opentelemetry.conf

Redeploy datakit

kubectl apply -f datakit.yaml

After deployment, check in the monitor to see if "opentelemetry" has been enabled.

# kubectl exec -it -n datakit pods/datakit-lfb95 datakit monitor

Modify the ASM output address¶

To facilitate testing, first modify the sampling rate of ASM

# kubectl edit -n istio-system cm istio-1-15-5-r4
...
sampling: 100   #This parameter has been changed from 1 to 100.
...

Modify ASM output to the Guance

# kubectl edit -n monitoring cm otel-collector-conf
...
exporters:
      apm:
        address: "100.125.0.78:8923"
        project_id: f5f4c067d68b4b3aad86e173b18367bf
        cluster_id: 5f642ce9-4aca-11ee-9dbd-0255ac10024d
      otlp:   # Add OTLP output.
        endpoint: "http://datakit-service.datakit:4317"
        tls:
          insecure: true
        compression: none 
...
traces/apm:
          receivers: [ zipkin ]
          processors: [ memory_limiter, batch ]
          exporters: [ otlp ]   #Change the output to OTLP.

After configuring the changes, let's access the address http://124.70.68.49/productpage several times. Then, go to the Guance workspace, into the 'APM' -> 'Traces', select the host ip, and enter 'All Traces', you can check the traces.

Click on a trace, and point the 'Service Invocation Relation' to view detailed information.