NetBay journal
Practical guides for Linux, containers, CI/CD, self-hosting, and running VPS infrastructure from Netbay Engineering.
599 articles · 24 topics
Architect a resilient, latency-optimized multi-region service mesh without hypercloud lock-in using WireGuard tunnels, anycast DNS routing, and Envoy proxies.
Optimize pgvector 0.7+ HNSW indexes on PostgreSQL 16 with memory-tuned maintenance_work_mem, vector quantisation, and calibrated ef_search settings.
Master modern eBPF tracing on Linux 6.12+ kernels to profile syscall latency, block I/O, and CPU schedulers without overhead.
Configure rootless Docker and Linux user namespaces to eliminate root daemon vulnerabilities, restrict container capabilities, and secure your VPS workloads.
A disciplined before/after method for finding what slowed your service down, from capturing a baseline to bisecting the changes that caused it.
A practical walkthrough that turns traffic forecasts into concrete CPU, RAM, and bandwidth numbers you can use to size a small web service now.
Apply namespaces, cgroups, and seccomp to harden containers and isolated processes on Linux, reducing the blast radius of any single workload.
Organize every self-hosted service into one fast landing page using Homer or Dashy, with live health dots, custom widgets, and reverse proxy via Docker.
A step-by-step disaster recovery runbook for VPS: qualify the incident, restore the latest good backup, validate, and cut over with a single checklist.
How interrupt handling, NUMA topology, and the Linux CPU scheduler behave inside a VPS, and exactly when you can realistically improve them.
What CPU credits and burst capacity really mean for throughput, and why sustained-load headroom matters far more than burst bragging rights.
Move a direct-to-internet app behind a reverse proxy in staged steps with minimal downtime, instant rollback, and smoke tests before cutover.