C++ Production Deployment: Docker, systemd, Kubernetes, Monitoring [#50-5]

Introduction: “Works locally, dies on the server”

Typical production issues:

• glibc mismatch between build and run hosts
• OOM in containers without proper limits
• No restart policy on crash
• Logs only on stdout—lost on restart
• 502 during deploy without readiness

Topics: multi-stage Docker, systemd on VMs, Kubernetes Deployments/Services, GitHub Actions, rolling / blue-green, Prometheus, JSON logs (e.g. spdlog).

---

Scenarios

Issue	Mitigation
GLIBC_x.y not found	Build in same base as runtime, or static libstdc++/musl where appropriate
Pod OOMKilled	Set memory limits; fix leaks (ASan/Valgrind)
Brief 502 on deploy	readinessProbe, maxUnavailable: 0, graceful shutdown
Lost logs	Centralize (Loki/ELK), volume mounts
Manual deploy mistakes	CI/CD with image tags = git SHA
No reboot autostart	systemd enable

---

Docker

• Multi-stage: builder with compiler, slim runtime with only .so deps + binary.
• non-root user, HEALTHCHECK, minimal apt/apk.
• docker-compose for local app + Prometheus + Grafana.

---

systemd

• [Service]: ExecStart, Restart=on-failure, TimeoutStopSec for graceful exit, journald logging, Security hardening options.

---

Kubernetes

• Deployment with RollingUpdate, resources, liveness + readiness HTTP checks.
• ConfigMaps / Secrets for env.

---

CI/CD

• Build image in CI, push to registry, deploy with pinned digests/tags.

---

Monitoring & logging

• /metrics for Prometheus; Grafana dashboards.
• Structured JSON logs with correlation IDs.

---

Related posts

• Debugging techniques
• Monitoring dashboard

---

Summary

Align build and runtime bases, enforce resource limits, automate releases, and observe with metrics + logs—C++ needs the same SRE discipline as any other stack.