Application Real-Time Monitoring Service:Monitor a Go application deployed in an ACK cluster

How it works

Monitoring a Go application in ACK involves three components:

1. ack-onepilot -- An ACK cluster component that connects Java or Go applications deployed in ACK clusters to the ARMS backend.

2. instgo -- An ARMS build tool that instruments your Go binary at compile time. Go compiles to a static binary, so instrumentation must happen during the build (unlike Java, which supports runtime injection).

3. Pod labels -- Kubernetes labels on your Deployment that tell ack-onepilot which pods to monitor and what name to display in the ARMS console.

Data flow:

Go binary (compiled with instgo) --> ack-onepilot (collects traces) --> ARMS backend --> ARMS console

Setup overview:

Prerequisites

ACK cluster:

• An ACK managed cluster, ACK dedicated cluster, or registered cluster. For details, see Create an ACK managed cluster

• A Go application deployed or ready to deploy. For supported frameworks, see Go libraries supported by Application Monitoring

ARMS:

• Application Monitoring activated with the pay-by-observable-data billing mode enabled

• Application Monitoring includes a free monthly quota of 50 GB. Unused quota does not carry over to the next month. If usage exceeds the free quota, standard rates apply. For pricing details, see Billing

Sample application (optional)

If you do not have a Go application ready, deploy the following sample to follow along. It creates two Deployments (otel-go-client and otel-go-server) and one Service (otel-server) in the default namespace. See Create a stateless application by using a Deployment.

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: otel-go-client
  name: otel-go-client
  namespace: default
spec:
  progressDeadlineSeconds: 600
  replicas: 1
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: otel-go-client
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      labels:
        app: otel-go-client
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/private-mesh/hellob:otel-client-arms-go-test-0.0.1
          imagePullPolicy: Always
          name: client
          resources:
            requests:
              cpu: 250m
              memory: 300Mi
          terminationMessagePath: /dev/termination-log
          terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30

---

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: otel-go-server
  name: otel-go-server
  namespace: default
spec:
  progressDeadlineSeconds: 600
  replicas: 1
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: otel-go-server
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      labels:
        app: otel-go-server
    spec:
      containers:
        - image: registry.cn-hangzhou.aliyuncs.com/private-mesh/hellob:otel-server-arms-go-test-0.0.1
          imagePullPolicy: Always
          name: server
          resources:
            requests:
              cpu: 250m
              memory: 300Mi
          terminationMessagePath: /dev/termination-log
          terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30

---

apiVersion: v1
kind: Service
metadata:
  labels:
    app: otel-go-server
  name: otel-server
  namespace: default
spec:
  internalTrafficPolicy: Cluster
  ipFamilies:
    - IPv4
  ipFamilyPolicy: SingleStack
  ports:
    - name: http
      port: 9000
      protocol: TCP
      targetPort: 9000
  selector:
    app: otel-go-server
  sessionAffinity: None
  type: ClusterIP

Step 1: Install ack-onepilot

1. Log on to the ACK console. On the Clusters page, click the name of your cluster.

2. In the left-side navigation pane, click Add-ons. On the Add-ons page, search for ack-onepilot.

3. In the ack-onepilot section, click Install. In the dialog box that appears, click OK.

Step 2: Compile the application with instgo

instgo is the ARMS build tool for Go. It wraps go build and injects instrumentation into your binary at compile time. Store instgo in a writable directory, because it triggers automatic updates during compilation.

# Public endpoint
http://arms-apm-<region-id>.oss-<region-id>.aliyuncs.com/instgo/instgo-<os>-<arch>

# VPC endpoint
http://arms-apm-<region-id>.oss-<region-id>-internal.aliyuncs.com/instgo/instgo-<os>-<arch>

wget "http://arms-apm-cn-hangzhou.oss-cn-hangzhou.aliyuncs.com/instgo/instgo-linux-amd64" -O instgo

chmod +x instgo

instgo go build {arg1} {arg2} {arg3}

Step 3: Add pod labels to your Deployment

Add three labels under spec.template.metadata.labels to enable ARMS monitoring for a Deployment.

labels:
  aliyun.com/app-language: golang    # Identifies this as a Go application
  armsPilotAutoEnable: 'on'          # Enables ARMS monitoring
  armsPilotCreateAppName: "<your-deployment-name>"  # Display name in the ARMS console

To add these labels through the ACK console:

1. In the left-side navigation pane, choose Workloads > Deployments. Select the namespace where your Deployment resides.

2. Find your Deployment and choose > Edit YAML in the Actions column.

3. In the Edit YAML dialog box, add the three labels under spec.template.metadata.labels and click Update.

Example: The following screenshots show the labels added to the otel-go-client and otel-go-server sample Deployments.

otel-go-client:

Verify that monitoring is working

After updating the Deployment YAML, the pods redeploy automatically. Wait 1--2 minutes for trace data to appear.

1. In the ARMS console, navigate to Application Monitoring > Application List.

2. Click the name of your application to view its monitoring data.

View monitoring data

For a full walkthrough, see View monitoring details.

Trace Explorer

Trace Explorer lets you filter and aggregate trace data in real time. Use it to identify failed or slow calls and drill into individual traces.

Click a trace to view its call chain, including timing, status codes, and span details.

Metrics

The application overview page shows key performance indicators such as request count, response time, and error rate.

The topology view maps dependencies between services, showing request flow and latency at each hop.

Configure alerting

Set up alert rules to get notified when performance degrades or errors spike. Notifications can be sent to contacts or DingTalk group chats. For details, see Alert rules.

Clean up resources

After completing this guide:

• To keep monitoring active: Make sure your Alibaba Cloud account has no overdue payments.

• To stop monitoring: Uninstall the ARMS agent from your application. For instructions, see Uninstall the Golang agent.

See also

• Associate trace IDs with business logs -- Link trace IDs to application logs for faster root cause analysis when errors occur.

• Continuous profiling -- Use flame graphs to diagnose high CPU utilization, memory usage, or slow calls in production.

• Use Managed Service for Prometheus to query application monitoring data -- Visualize ARMS metrics in Prometheus dashboards and build custom views.