Learn from production

• Tracing large job failures to serial console bottlenecks from oom events

  Jack Lindamood

  This case study shows how OOM Killer terminated a critical network daemon on Kubernetes nodes, causing a network outage.

  It covers debugging via serial console and implementing memory reservations to prevent system-critical process termination.

• Three Weeks in the Trenches: Hunting a 4GB Native Memory Leak That .NET Couldn’t See

  Kalyan Josyula

  This case study shows how a team traced repeated pod OOM kills in ASP.NET Core to native memory growth from zombie SignalR connections, glibc fragmentation, and kernel socket buffers.

• Autoscaling Hid Our LLM Cost Regression (85% → 4% Cache Hit Rate)

  Nick Roan

  This case study shows how a single RAG chunk size change collapsed vLLM prefix-cache hit rate from 85% to 4%, triggering an 80% GPU replica increase while latency stayed flat.

  It also includes the fix: adding a two-phase cache replay gate in CI.

• How My Client Hit Linux Kernel Network Limits on AWS EKS

  Dat Ton

  This case study explains how cURL 65 errors and DNS resolution failures on AWS EKS were caused by Linux kernel network limits being exceeded, resolved by increasing netdev_budget, netdev_budget_usecs, and netdev_max_backlog parameters.

More case studies →

Matching jobs

1. • DevOps Engineer with MaintainX

   • Salary: $117K to $302.17K a year

   • Location: remote from

   • Tech stack: Kubernetes

2. • Developer Advocate with MariaDB plc

   • Salary: $45K to $382.8K a year

   • Location: remote from

   • Tech stack: Kubernetes, SQL, Java, Python, Redis, Kafka, Datadog, Grafana, Prometheus, Splunk

3. • Engineering Manager with MaintainX

   • Salary: $58.5K to $473K a year

   • Location: based in the office in San Francisco, CA, USA

   • Tech stack: Kubernetes

4. • Software Engineer with Manychat

   • Salary: US$9 to US$533.5K a year

   • Location: based in the office (and remote from home) in Amsterdam, NL

   • Tech stack: Kubernetes, Docker, Python, SQL, Redis, PostgreSQL, Grafana, Prometheus

5. • Support Engineer with MariaDB plc

   • Salary: $45K to $224.4K a year

   • Location: remote from

   • Tech stack: Kubernetes, Docker, Java, Redis, Cassandra, Kafka

Discover more Kubernetes jobs on Kube Careers →

Build something

• Build a Production-Ready vLLM Inference Server on Kubernetes with AMD Instinct GPUs

  Ojas Singh

  This tutorial teaches how to run a production-ready vLLM inference server on Kubernetes with AMD Instinct GPUs using containerd, the AMD GPU Operator, persistent storage, and MetalLB.

• Designing for Failure with CloudNativePG

  Dylan Da Costa

  This tutorial explains how to design CloudNativePG for production failure by using plugin-based backups, WAL archiving, point-in-time recovery, snapshots, and PgBouncer so recovery is treated as the real operational priority.

• Cilium Network Policies: L7 HTTP Filtering with eBPF on Kubernetes

  Dilan Dashintha

  This tutorial shows how to use Cilium and Hubble to enforce HTTP path based network policies in Kubernetes with eBPF, so you can allow or block specific endpoints without sidecars.

• The Complete OpenSSL & TLS Debugging Guide: From Root CA to Kubernetes

  Amolpratap Singh

  This tutorial explains TLS and certificate debugging from root CA basics to Kubernetes secrets, with OpenSSL and curl commands for inspecting certs, validating handshakes, and fixing common production errors.

More tutorials →

Call for Papers closing soon

1. 3

   days

   DevOpsDays Istanbul 2026

   The Call For Paper is open until 31 May 2026 at GMT-4. More info →

   • Location: Istanbul, TR

   • In-person conference organized by DevOps Turkey.

   • The conference starts on the 29 September 2026.

   • Apply here

2. 3

   days

   DevOps Midwest 2026

   The Call For Paper is open until 31 May 2026 at GMT-4. More info →

   • Location: St. Louis, MO, USA

   • In-person conference organized by Sketch.

   • The conference starts on the 22 August 2026.

   • Apply here

3. 4

   days

   KubeCon + CloudNativeCon North America 2026

   The Call For Paper is open until 1 June 2026 at GMT-4. More info →

   • Location: Los Angeles, CA, USA

   • In-person conference organized by Linux Foundation.

   • The conference starts on the 26 October 2026.

   • Apply here

4. 4

   days

   Heapcon 2026

   The Call For Paper is open until 1 June 2026 at GMT-4. More info →

   • Location: Belgrade, RS

   • In-person conference organized by heapspace.

   • The conference starts on the 6 November 2026.

   • Apply here

5. 4

   days

   Cloud Native Days Norway

   The Call For Paper is open until 1 June 2026 at GMT-4. More info →

   • Location: Bergen, NO

   • In-person conference organized by CND Norway.

   • The conference starts on the 27 October 2026.

   • Apply here

6. 4

   days

   Devopsdays Istanbul

   The Call For Paper is open until 1 June 2026 at GMT-4. More info →

   • Location: Istanbul, TR

   • In-person conference organized by Devopsdays.

   • The conference starts on the 24 October 2026.

   • Apply here

7. 4

   days

   Kubernetes Community Days Melbourne 2026

   The Call For Paper is open until 1 June 2026 at GMT-4. More info →

   • Location: Melbourne, AU

   • In-person conference organized by KCD Australia.

   • The conference starts on the 4 August 2026.

   • Apply here

More articles

• Why Kubernetes Has No Login — And How We Solved It for AuditRadar

  hybrid2k3

  This article explains why vanilla Kubernetes has no real login event and shows a practical session-tracking workaround using credential-id fingerprints from audit logs, with a side-by-side comparison against OpenShift OAuth behavior.

• Building eBPF-Based Bandwidth Limiting in AWS Network Policy Agent — Why Vibe Coding Isn’t Enough

  Jayanth Varavani

  This article walks you through building EDT-based eBPF bandwidth limiting in the AWS Network Policy Agent, showing where AI-generated code silently broke and how domain knowledge caught each bug.

• Architecting GPUaaS for Enterprise AI On-Prem

  Joe Sasson

  This article explains how a single-node OpenShift cluster can be turned into a multi-tenant on-premises GPU platform with reservation-based scheduling, MIG partitioning, time slicing, isolated namespaces, and controller-driven self-healing.

• Why I built ctx_: the context switcher that actually gets DevOps work

  Vedran Lebo

  This article introduces ctx_, a CLI tool that switches an entire DevOps working context at once, including Kubernetes context, cloud credentials, environment variables, VPN, SSH tunnels, secrets, and browser profile.

Even more articles →