Go

Go (Golang) is a modern, statically typed language designed for simplicity, performance, and scalability. It's a top choice for DevOps and SRE engineers working with AWS, Azure, GCP, Kubernetes, and cloud-native tooling.

Why DevOps & SREs Should Learn Go

• Cloud Native: Go powers Kubernetes, Docker, Prometheus, and many cloud SDKs (AWS, Azure, GCP).

• Performance: Compiled binaries, fast startup, and low memory usage make Go ideal for microservices and CLI tools.

• Concurrency: Goroutines and channels enable efficient parallel processing for automation and monitoring.

• DevOps Ecosystem: Many DevOps tools (Terraform, Helm, Vault, etc.) are written in Go, making it easy to extend or contribute.

• Cross-Platform: Build for Linux, NixOS, WSL, and all major clouds with a single codebase.

Real-Life DevOps & SRE Examples

1. Build & Deploy a Go App on Kubernetes

Dockerfile:

FROM golang:1.21-alpine as builder
WORKDIR /app
COPY . .
RUN go build -o app

FROM alpine:3.18
WORKDIR /app
COPY --from=builder /app/app .
CMD ["./app"]

deployment.yaml:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: myapp
spec:
  replicas: 2
  selector:
    matchLabels:
      app: myapp
  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - name: myapp
        image: myrepo/myapp:v1
        ports:
        - containerPort: 8080

2. Kubernetes Operator with Go (kubebuilder)

// ...existing code for imports...
func (r *MyResourceReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
    // Custom reconciliation logic
    // ...
    return ctrl.Result{}, nil
}

3. Custom Kubernetes Controller

// ...existing code for imports...
func main() {
    // Setup manager, scheme, and controller
    // Watch for resource changes and automate tasks
    // ...
}

4. Prometheus Exporter in Go

package main
import (
    "net/http"
    "github.com/prometheus/client_golang/prometheus"
    "github.com/prometheus/client_golang/prometheus/promhttp"
)
func main() {
    myMetric := prometheus.NewGauge(prometheus.GaugeOpts{
        Name: "my_custom_metric",
        Help: "A custom metric for SREs",
    })
    prometheus.MustRegister(myMetric)
    http.Handle("/metrics", promhttp.Handler())
    http.ListenAndServe(":2112", nil)
}

5. CI/CD Pipeline Step (GitHub Actions)

- name: Build Go binary
  run: go build -o app
- name: Run tests
  run: go test ./...

Best Practices (2025)

• Use Go modules for dependency management

• Write unit and integration tests (use testing and testify)

• Lint and format code with golangci-lint and gofmt

• Use context for cancellation/timeouts in automation

• Pin dependencies in go.mod

• Document code and APIs

Common Pitfalls

• Not handling errors (always check returned errors)

• Hardcoding credentials (use env vars or secret managers)

• Ignoring context cancellation in long-running tasks

• Not using modules (avoid GOPATH for new projects)

References

• Go Official Docs
• Kubernetes Go Client
• Kubebuilder
• Prometheus Go Client

---

Go Joke: Why did the SRE love Go? Because handling errors is a feature, not a bug!