DevOps&SRE Library

Say the Thing You Want

You’re in a 1:1 with your manager, and things are going just fine. You talk about the project and that other thing. Toward the end, she asks: “Anything else?” And there is something else. You want to lead that new initiative. Or move to a different team. Or you’ve been thinking about what stands in the way of your promotion. The thought is right there, sitting in the back of your throat. You’re going to say it, and then… “Nope, all good.” You get out of the call feeling a specific kind of regret. You rationalize it somehow and then tell yourself you’ll bring it up next time (you won’t).

https://terriblesoftware.org/2026/04/01/say-the-thing-you-want

Why Teamwork Makes (Or Breaks) Your Incident Response

High-severity incidents expose how a team really works together, usually within the first ten minutes.

https://uptimelabs.io/articles/teamwork-incident-response

Why reviewing AI-generated code is devilishly hard

When working on code with GenAI assistance you need a better understanding of the system than when working without.

https://www.spinellis.gr/blog/20260523

Humans aren't fast enough for 4 9's

When thinking about Service Level Objectives (SLOs) and contractual Service Level Agreements (SLAs) for availability, I always like to put the percentages into concrete numbers.

https://incident.io/blog/humans-arent-fast-enough-for-4-nines

On benchmarking

Benchmarking is hard. There are many ways to do it wrong and few to do it right. But zooming out from any single system or harness, there are broad principles that should be applied to all benchmarking. Using these correctly makes it difficult to produce biased results. Am I the world's best benchmarker? Certainly not. I invented the language balls, after all. But correctness and precision are important parts of PlanetScale's culture. We've spent considerable time learning the art of benchmarking, and are here to share best-practices. Here, we're focusing primarily on benchmarking databases, but these principles apply to many domains.

https://planetscale.com/blog/on-benchmarking

You Shipped It Fast. But Did You Ship It Right?

AI tools have genuinely changed how fast teams can produce code, but they haven't changed how fast a codebase can safely absorb that code.

https://stackoverflow.blog/2026/05/12/you-shipped-it-fast-but-did-you-ship-it-right

The Problem with AI-Generated Post-Incident Reviews

The real learning comes from analyzing the incident while writing the document, not reading it; the document at the end is the residue of the learning.

https://greatcircle.com/blog/2026/05/05/problem-with-ai-generated-post-incident-reviews

agentgram

A single front door for all your AI agents and MCPs

https://github.com/dfradehubs/agentgram

Что общего у SRE и рыбаков? «GitOps = реальность» — это миф? Не создаёт ли Chaos Engineering ещё больше хаоса?.. Звучит как те самые внезапные вопросы перед сном в будний день 👀 И, кстати, ответ на все три у нас имеется! Правда, не здесь, а в подкасте «В SREду на кухне» — его ведут опытные инженеры из Авито. Они обсуждают наболевшее, приглашают внешних гостей и коллег, а также делятся дополнительными инсайтами, статьями по теме и анонсами встреч в своём канале. Советуем подписаться и сохранить на будущее пару выпусков 🧠

SlimFaas

SlimFaas is a lightweight, plug-and-play Function-as-a-Service (FaaS) platform for Kubernetes (and Docker-Compose / Podman-Compose).

https://github.com/SlimPlanet/SlimFaas

Building a Production-Grade HA Kubernetes Cluster on a Homelab with $0 in Cloud Costs

How I turned four Proxmox nodes, some enterprise surplus drives, and an afternoon into a fully automated HA k3s cluster with Rancher, Traefik, and Ansible — all running on hardware that draws less power than a gaming PC.

https://thiago-marsal.medium.com/homelab-k3s-ha-cluster-a-complete-architecture-guide-6a60005b6e99

Designing for Failure with CloudNativePG

This post focuses on three areas that separate a demo from production systems: backups, recovery and connection pooling.

https://dylanmarkdacosta.medium.com/designing-for-failure-with-cloudnativepg-2c3987605a39

Durable Workflows Beyond Vercel: Version-Safe Orchestration for Kubernetes

Workflow DevKit lets you write durable, long-running workflows directly in your Next.js and Node.js apps. You define steps with ’use step’, and the SDK handles persistence, retries, and replay automatically. Workflows survive server restarts, can sleep for days, and resume exactly where they left off. On Vercel, all of this works out of the box — the platform handles deployment versioning and queue routing behind the scenes. But what happens when you deploy to your own Kubernetes cluster? Version mismatch. And it’s subtle enough to corrupt data before you notice. We built Platformatic World to fix this. It’s a drop-in World implementation that brings the same deployment safety to any Kubernetes cluster. Every workflow run is pinned to the code version that created it. Queue messages are routed to the correct versioned pods. Old versions stay alive until all their in-flight runs are complete.

https://blog.platformatic.dev/durable-workflows-kubernetes-version-safe

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

When we set out to build the Logins page for AuditRadar — a real-time audit log explorer for OpenShift and Kubernetes — we hit a wall that forced us to deeply understand how authentication actually works on each platform.

https://blog.audit-radar.com/why-kubernetes-has-no-login-and-how-we-solved-it-for-auditradar

A one-line Kubernetes fix that saved 600 hours a year

Every time we restarted Atlantis, the tool we use to plan and apply Terraform changes, we’d be stuck for 30 minutes waiting for it to come back up.

https://blog.cloudflare.com/one-line-kubernetes-fix-saved-600-hours-a-year

InfraDev Community приглашает на InfraDev Meetup #4: про AI и не только Прямо сейчас мы наблюдаем как AI в SDLC меняет процесс разработки — об удачных примерах и кейсах поговорим в этот раз. Обсудим, как разрабатывать инфраструктурные сервисы с помощью AI и как построить MLOps-платформу для обучения моделей. И не только: пока AI не перестроил DevOps-цикл, классические вызовы сборки образов для виртуальных машин остаются актуальными. Спикеры ▫️Кирилл Фролов, эксперт-разработчик в отделе разработки базовых сервисов, VK Cloud, VK Tech ▫️Павел Шипилов, Старший разработчик ML Платформы в Avito ▫️Александр Александров, системный архитектор в направлении разработки и управления инфраструктурой, VK Cloud, VK Tech Подробнее о докладах читайте на странице мероприятия. Когда: 10 июня, с 18:00 до 23:59 Где: Москва, Ленинградский пр., 70, офис VK Tech, БЦ «Алкон» (количество мест ограничено). Приходите на встречу или участвуйте онлайн. Зарегистрироваться.

Kubernetes production readiness checklist https://assets.learnkube.com/kubernetes-production-readiness-worksheet.v1.pdf

Why Your KServe InferenceService Won't Become Ready: Four Production Failures and Fixes

A practitioner's account of the errors the KServe getting-started documentation doesn't tell you about — with exact terminal output, root causes, and working Kustomize patches.

https://sodiq-jimoh.hashnode.dev/why-your-kserve-inferenceservice-won-t-become-ready-four-production-failures-and-fixes

Incident Report: May 19, 2026 - GCP Account Suspension

Railway experienced a platform-wide service disruption due to Google Cloud incorrectly placing our account in a suspended status.

https://blog.railway.com/p/incident-report-may-19-2026-gcp-account-outage

You've Got (Too Much) Mail: Behind the Scenes of the 3/25/26 Voice Outage

As part of a routine infrastructure change, a configuration update accidentally caused a large portion of Discord's session management servers to shut down simultaneously.

https://discord.com/blog/behind-the-scenes-of-the-3-25-26-voice-outage