# DevOps.dev

Devops.dev is a community of DevOps enthusiasts sharing insight, stories, and the latest development in the field. - Medium

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Stop Using Try-Catch Everywhere: Better Exception Handling in .NET

DevFeed: [Stop Using Try-Catch Everywhere: Better Exception Handling in .NET](<https://devfeed.tech/articles/stop-using-try-catch-everywhere-better-exception-handling-in-net-146531.md>)

Original publisher: [Read original article](<https://blog.devops.dev/stop-using-try-catch-everywhere-better-exception-handling-in-net-6c33917bce3a?source=rss----33f8b2d9a328---4>)

Author: Bharat Chandera

Published: 2026-11-20T12:30:30Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Exception handling](<https://devfeed.tech/topics/exception-handling.md>), [Error Handling](<https://devfeed.tech/topics/error-handling.md>), [Exception](<https://devfeed.tech/topics/exception.md>), [.NET](<https://devfeed.tech/topics/net.md>), [Custom exceptions](<https://devfeed.tech/topics/custom-exceptions.md>), [Null-Safety](<https://devfeed.tech/topics/null-safety.md>)

Tags: [asp-net](<https://devfeed.tech/tags/asp-net.md>), [asp-net-core](<https://devfeed.tech/tags/asp-net-core.md>), [aspnetcore](<https://devfeed.tech/tags/aspnetcore.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [business-logic](<https://devfeed.tech/tags/business-logic.md>), [clean-code](<https://devfeed.tech/tags/clean-code.md>), [csharp](<https://devfeed.tech/tags/csharp.md>), [dotnet](<https://devfeed.tech/tags/dotnet.md>), [exception](<https://devfeed.tech/tags/exception.md>), [exception-handling](<https://devfeed.tech/tags/exception-handling.md>), [middleware](<https://devfeed.tech/tags/middleware.md>), [net](<https://devfeed.tech/tags/net.md>)

### AI overview

Catching every exception and returning `BadRequest(ex.Message)` can make a .NET application harder to maintain, debug, and operate. The excerpt illustrates that APIs encounter distinct failures--database connection errors, missing records, payment timeouts, business-rule violations, and unexpected null references--but does not include the article's proposed handling approach.

### Source excerpt

Catching every exception doesn't make your application safer. Learn how to build clean, centralized, and reliable exception handling in... Continue reading on DevOps.dev "

## Java4Shell: What Happened and What to Do Now

DevFeed: [Java4Shell: What Happened and What to Do Now](<https://devfeed.tech/articles/java4shell-what-happened-and-what-to-do-now-146530.md>)

Original publisher: [Read original article](<https://blog.devops.dev/java4shell-what-happened-and-what-to-do-now-6c5721503d99?source=rss----33f8b2d9a328---4>)

Author: Ashen Bhagya

Published: 2026-11-20T12:30:26Z

Content type: tutorial

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Memory Safety](<https://devfeed.tech/topics/memory-safety.md>), [generative-ai-tools](<https://devfeed.tech/topics/generative-ai-tools.md>)

Tags: [base64](<https://devfeed.tech/tags/base64.md>), [bug](<https://devfeed.tech/tags/bug.md>), [calif](<https://devfeed.tech/tags/calif.md>), [confluence](<https://devfeed.tech/tags/confluence.md>), [crash](<https://devfeed.tech/tags/crash.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [devops](<https://devfeed.tech/tags/devops.md>), [java](<https://devfeed.tech/tags/java.md>), [java4shell](<https://devfeed.tech/tags/java4shell.md>), [jenkins](<https://devfeed.tech/tags/jenkins.md>), [jira](<https://devfeed.tech/tags/jira.md>), [keycloak](<https://devfeed.tech/tags/keycloak.md>), [log4shell](<https://devfeed.tech/tags/log4shell.md>), [memory](<https://devfeed.tech/tags/memory.md>), [password](<https://devfeed.tech/tags/password.md>), [programming](<https://devfeed.tech/tags/programming.md>), [secrets](<https://devfeed.tech/tags/secrets.md>), [server](<https://devfeed.tech/tags/server.md>), [servers](<https://devfeed.tech/tags/servers.md>), [software-development](<https://devfeed.tech/tags/software-development.md>)

### AI overview

The excerpt describes CVE-2026-114938 as a flaw in Java's Base64 decoder shortcut on certain x86-64 chips supporting AVX-512 VBMI: malformed input can cause an out-of-bounds write, potentially leaking secrets, crashing an app, or enabling code execution. Exposure requires a compatible chip, an affected Java build, and decoding Base64 from an external source; the excerpt says ARM and plain Java 8 and 11 are unaffected. It recommends checking CPU and Java versions, installing a vendor-patched build, or temporarily starting Java with `-XX:-UseBASE64Intrinsics`, which may slightly slow Base64 processing. It also cautions that apps and containers may bundle their own Java copies, which must be checked separately. The excerpt reports Calif's findings and Oracle's assessment; it does not provide independent verification.

### Source excerpt

A simple guide to the new Java bug, who it hits, and the one-line fix you can use today. What if a short line of random-looking text could take over your server? Not a virus. Not a stolen password. Just about 65 characters of Base64, sent to the wrong place. That's the story of Java4Shell. I'll be honest: when I first saw the name this week, my stomach dropped a little. It sounds like Log4Shell's scary cousin. If you were around in December 2021, you remember that mess. So I sat down and read the reports. Here's what I learned, in plain words. What happened, who should worry, and what you can do today. First, what is Base64? Base64 is a way to turn data into plain text letters. Pictures, files, login tokens. It's everywhere. Think of it like the scissors in your kitchen drawer. You use them all the time and never think about them. Java has a built-in Base64 decoder. Lots of apps use it every second, and nobody notices. So what went wrong? Java likes to be fast. On some newer Intel and AMD chips, it uses a shortcut. These chips have special powers called AVX-512, which let them handle many letters at once. The bug lives inside that shortcut. If someone sends badly formed Base64, Java gets the length math wrong. It writes more data than the space it set aside. Picture pouring a full jug of water into a small glass. The extra water spills onto the table. In a computer, that "table" is memory that belongs to something else. Reports say a string of only 65 characters is enough to trigger it on the right machine. That's shorter than this paragraph. Why is spilled memory such a big deal? When memory spills, three bad things can happen. Secrets can leak. The app can crash. Or, worst of all, an attacker can run their own code on your server. And they don't need to log in first. The security firm Calif found the bug and reported it to Oracle before going public. They say they used it to take over Jira, Confluence, Jenkins and Keycloak servers without a password. Those are too

## 4 Simple Tips to Secure Your SSH Login

DevFeed: [4 Simple Tips to Secure Your SSH Login](<https://devfeed.tech/articles/4-simple-tips-to-secure-your-ssh-login-146528.md>)

Original publisher: [Read original article](<https://blog.devops.dev/4-simple-tips-to-secure-your-ssh-login-ac94a82bc8eb?source=rss----33f8b2d9a328---4>)

Author: Chimin

Published: 2026-11-20T12:30:20Z

Content type: tutorial

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [ssh](<https://devfeed.tech/topics/ssh.md>), [least privilege](<https://devfeed.tech/topics/least-privilege.md>)

Tags: [authentication](<https://devfeed.tech/tags/authentication.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [features](<https://devfeed.tech/tags/features.md>), [linux](<https://devfeed.tech/tags/linux.md>), [login](<https://devfeed.tech/tags/login.md>), [password](<https://devfeed.tech/tags/password.md>), [security](<https://devfeed.tech/tags/security.md>), [server](<https://devfeed.tech/tags/server.md>), [servers](<https://devfeed.tech/tags/servers.md>), [ssh](<https://devfeed.tech/tags/ssh.md>), [ubuntu](<https://devfeed.tech/tags/ubuntu.md>)

### AI overview

The excerpt recommends disabling password and keyboard-interactive SSH authentication in Ubuntu 26.04 by adding settings to a file under `/etc/sshd/ssh_config.d/`, then using SSH keys. It also advises SSH users to use a lower-privileged account instead of root, optionally restricting access with `AllowGroups sshusers`, and restarting the SSH service after configuration changes. A third tip begins with disabling forwarding features, but the excerpt does not provide details.

### Source excerpt

If you have a server, or want to connect to your home computer while out, you probably use SSH frequently, and that's when security really... Continue reading on DevOps.dev "

## Your Team Agrees on the Fix in Teams, Then Retypes It Into GitHub

DevFeed: [Your Team Agrees on the Fix in Teams, Then Retypes It Into GitHub](<https://devfeed.tech/articles/your-team-agrees-on-the-fix-in-teams-then-retypes-it-into-github-136237.md>)

Original publisher: [Read original article](<https://blog.devops.dev/github-app-microsoft-teams-copilot-thread-to-pull-request-606a8ca2150a?source=rss----33f8b2d9a328---4>)

Author: Dave R | Microsoft Azure & AI MVP ☁

Published: 2026-11-16T13:26:12Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Microsoft Teams](<https://devfeed.tech/topics/microsoft-teams.md>), [GitHub](<https://devfeed.tech/topics/github.md>), [GitHub Copilot](<https://devfeed.tech/topics/github-copilot.md>)

Tags: [admin](<https://devfeed.tech/tags/admin.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [cloud-agent](<https://devfeed.tech/tags/cloud-agent.md>), [copilot](<https://devfeed.tech/tags/copilot.md>), [github](<https://devfeed.tech/tags/github.md>), [github-copilot](<https://devfeed.tech/tags/github-copilot.md>), [github-issue](<https://devfeed.tech/tags/github-issue.md>), [microsoft-teams](<https://devfeed.tech/tags/microsoft-teams.md>), [permissions](<https://devfeed.tech/tags/permissions.md>), [pull-request](<https://devfeed.tech/tags/pull-request.md>), [rules](<https://devfeed.tech/tags/rules.md>), [sandbox](<https://devfeed.tech/tags/sandbox.md>), [software-development](<https://devfeed.tech/tags/software-development.md>), [teams](<https://devfeed.tech/tags/teams.md>), [thread](<https://devfeed.tech/tags/thread.md>)

### AI overview

The Microsoft Teams GitHub app is generally available and lets users mention @GitHub to turn thread context into GitHub issues or Copilot pull requests, avoiding manual re-entry. Copilot works in a cloud sandbox, while existing GitHub permissions and approval rules continue to govern the work. Its identity depends on where it is invoked: in a direct message it acts as the user's GitHub account; in a channel or group chat it acts as the app, and pull requests require an additional approval.

### Source excerpt

The GitHub app in Microsoft Teams is now GA. How @GitHub turns a thread into a Copilot pull request, who it acts as, and what to set up... Continue reading on DevOps.dev "

## Kubernetes Is Retiring kube-proxy's IPVS Mode. Canary nftables First

DevFeed: [Kubernetes Is Retiring kube-proxy's IPVS Mode. Canary nftables First](<https://devfeed.tech/articles/kubernetes-is-retiring-kube-proxy-s-ipvs-mode-canary-nftables-first-136239.md>)

Original publisher: [Read original article](<https://blog.devops.dev/kubernetes-is-retiring-kube-proxys-ipvs-mode-canary-nftables-first-5ec3e12c91fb?source=rss----33f8b2d9a328---4>)

Author: Daniel Valev

Published: 2026-11-16T13:26:06Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [canary release](<https://devfeed.tech/topics/canary-release.md>), [upgrade](<https://devfeed.tech/topics/upgrade.md>), [DevOps](<https://devfeed.tech/topics/devops.md>)

Tags: [deprecation](<https://devfeed.tech/tags/deprecation.md>), [devops](<https://devfeed.tech/tags/devops.md>), [dns](<https://devfeed.tech/tags/dns.md>), [exit](<https://devfeed.tech/tags/exit.md>), [feature](<https://devfeed.tech/tags/feature.md>), [gate](<https://devfeed.tech/tags/gate.md>), [interface](<https://devfeed.tech/tags/interface.md>), [ip](<https://devfeed.tech/tags/ip.md>), [iptables](<https://devfeed.tech/tags/iptables.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [kubernetes-clusters](<https://devfeed.tech/tags/kubernetes-clusters.md>), [load-balancer](<https://devfeed.tech/tags/load-balancer.md>), [model](<https://devfeed.tech/tags/model.md>), [nftables](<https://devfeed.tech/tags/nftables.md>), [node](<https://devfeed.tech/tags/node.md>), [override](<https://devfeed.tech/tags/override.md>), [packet](<https://devfeed.tech/tags/packet.md>), [pool](<https://devfeed.tech/tags/pool.md>), [programming](<https://devfeed.tech/tags/programming.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [technology](<https://devfeed.tech/tags/technology.md>), [trace](<https://devfeed.tech/tags/trace.md>), [upgrade](<https://devfeed.tech/tags/upgrade.md>)

### AI overview

The excerpt says kube-proxy IPVS mode has been deprecated since Kubernetes v1.35 and, from v1.40, kube-proxy exits at startup in that mode unless a feature gate is overridden. It identifies a key difference in how ClusterIPs are represented: in IPVS mode, a node's `ip addr` output shows Service ClusterIPs on `kube-ipvs0`. The teaser recommends tracing a packet through iptables, IPVS, and nftables, then canarying nftables on one node pool before upgrading. The excerpt does not provide the packet paths or explain the alternative modes in detail.

### Source excerpt

Trace one packet through iptables, IPVS, and nftables mode, then canary nftables on one node pool before the upgrade decides for you. Continue reading on DevOps.dev "

## 40 Kubernetes Interview Questions That Expose Weak Senior Engineers

DevFeed: [40 Kubernetes Interview Questions That Expose Weak Senior Engineers](<https://devfeed.tech/articles/40-kubernetes-interview-questions-that-expose-weak-senior-engineers-136234.md>)

Original publisher: [Read original article](<https://blog.devops.dev/40-kubernetes-interview-questions-that-expose-weak-senior-engineers-f183baad39fb?source=rss----33f8b2d9a328---4>)

Author: Devrim Ozcay | Founder, ProdRescue

Published: 2026-11-16T13:26:05Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [control-plane](<https://devfeed.tech/topics/control-plane.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [zero downtime deployment](<https://devfeed.tech/topics/zero-downtime-deployment.md>), [health check](<https://devfeed.tech/topics/health-check.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>), [Post Mortem](<https://devfeed.tech/topics/post-mortem.md>)

Tags: [advanced](<https://devfeed.tech/tags/advanced.md>), [affinity](<https://devfeed.tech/tags/affinity.md>), [autoscaling](<https://devfeed.tech/tags/autoscaling.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [devops](<https://devfeed.tech/tags/devops.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [docker](<https://devfeed.tech/tags/docker.md>), [failure](<https://devfeed.tech/tags/failure.md>), [interview](<https://devfeed.tech/tags/interview.md>), [interview-questions](<https://devfeed.tech/tags/interview-questions.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [kubernetes-cluster](<https://devfeed.tech/tags/kubernetes-cluster.md>), [latency](<https://devfeed.tech/tags/latency.md>), [load-balancing](<https://devfeed.tech/tags/load-balancing.md>), [memory](<https://devfeed.tech/tags/memory.md>), [networking](<https://devfeed.tech/tags/networking.md>), [persistent-volume](<https://devfeed.tech/tags/persistent-volume.md>), [production](<https://devfeed.tech/tags/production.md>), [programming](<https://devfeed.tech/tags/programming.md>), [questions](<https://devfeed.tech/tags/questions.md>), [routing](<https://devfeed.tech/tags/routing.md>), [scheduling](<https://devfeed.tech/tags/scheduling.md>), [staff](<https://devfeed.tech/tags/staff.md>)

### AI overview

The article argues that senior Kubernetes troubleshooting requires tracing how scheduling, networking, application lifecycle, resources, and dependencies interact. It explains practical failure patterns and diagnostics: a Pending Pod may lack an eligible node despite spare aggregate CPU; CPU limits can throttle latency-sensitive work; premature readiness can route traffic to unprepared instances; and autoscaling can worsen overload by increasing pressure on a saturated dependency. It recommends checking scheduling events and requests, probe behavior, EndpointSlices, DNS and conntrack metrics, resource throttling, traces, and downstream capacity. Other sections cover safe rollouts and shutdown, database connection budgets, stateful workload recovery, security boundaries, incident response, and tradeoffs in multi-zone, multi-region, and multi-tenant designs. The text presents hypothetical interview scenarios and diagnostic approaches, not evidence of a specific incident or a complete Kubernetes reference.

### Source excerpt

Advanced Kubernetes troubleshooting, distributed systems, networking, autoscaling, and production reliability for Senior and Staff Engineers. A Kubernetes cluster with 180 nodes had plenty of available CPU. Every deployment reported success, and the control plane appeared healthy. Yet customers were experiencing intermittent 503 errors, and the API's p99 latency had climbed from 180 milliseconds to almost nine seconds. The infrastructure team increased the number of replicas. The application team increased connection pool sizes. Someone suggested moving the workloads to larger nodes because the existing cluster was obviously struggling. The situation became worse. The problem was that newly created Pods were passing readiness checks before their applications could serve real traffic. Every additional replica introduced another temporarily unhealthy endpoint into the load-balancing path. Autoscaling was increasing the number of requests routed toward instances that were technically running but operationally useless. This is why Kubernetes interviews become considerably harder at the Staff level. Knowing which command describes a Pod is useful, but it does not demonstrate an understanding of how scheduling, network routing, application lifecycle, resource allocation, and distributed dependencies interact during failure. The following 40 questions focus on those interactions. The scenarios are hypothetical, but the failure mechanisms and diagnostic approaches are grounded in Kubernetes and production engineering behavior. Part I -- Kubernetes Scheduling and Resource Management1. A Pod is Pending even though the cluster has available CPU and memory. Why? A candidate who immediately recommends adding nodes is missing several possibilities. Kubernetes scheduling evaluates requested resources alongside node selectors, affinity rules, taints, tolerations, topology constraints, persistent-volume placement, and other scheduling restrictions. Aggregate cluster capacity is irrel

## Managing Secrets in Dockerized Applications: From GitHub Actions to Production

DevFeed: [Managing Secrets in Dockerized Applications: From GitHub Actions to Production](<https://devfeed.tech/articles/managing-secrets-in-dockerized-applications-from-github-actions-to-production-136240.md>)

Original publisher: [Read original article](<https://blog.devops.dev/managing-secrets-with-docker-from-github-actions-to-the-production-server-99a5bc0cd53c?source=rss----33f8b2d9a328---4>)

Author: Vahid Faraji

Published: 2026-11-16T13:26:01Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Secrets Management](<https://devfeed.tech/topics/secrets-management.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [CI/CD secrets management](<https://devfeed.tech/topics/ci-cd-secrets-management.md>), [secret rotation](<https://devfeed.tech/topics/secret-rotation.md>), [Continuous Delivery (CD)](<https://devfeed.tech/topics/continuous-delivery.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [asp-net](<https://devfeed.tech/tags/asp-net.md>), [asp-net-core](<https://devfeed.tech/tags/asp-net-core.md>), [aspdotnetcore](<https://devfeed.tech/tags/aspdotnetcore.md>), [aws-secrets-manager](<https://devfeed.tech/tags/aws-secrets-manager.md>), [azure](<https://devfeed.tech/tags/azure.md>), [bind](<https://devfeed.tech/tags/bind.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [containers](<https://devfeed.tech/tags/containers.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [devops](<https://devfeed.tech/tags/devops.md>), [docker](<https://devfeed.tech/tags/docker.md>), [docker-image](<https://devfeed.tech/tags/docker-image.md>), [files](<https://devfeed.tech/tags/files.md>), [github-actions](<https://devfeed.tech/tags/github-actions.md>), [hashicorp-vault](<https://devfeed.tech/tags/hashicorp-vault.md>), [information](<https://devfeed.tech/tags/information.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>), [manage](<https://devfeed.tech/tags/manage.md>), [management-secrets](<https://devfeed.tech/tags/management-secrets.md>), [production](<https://devfeed.tech/tags/production.md>), [secrets](<https://devfeed.tech/tags/secrets.md>)

### AI overview

The author keeps production secrets outside Docker images and the deployment workflow by storing each value in a protected file on the VPS, mounting only needed files into containers as read-only, and adding those mounts to GitHub Actions' `docker run` command. In ASP.NET Core, `Program.cs` reads the files at startup into existing `IConfiguration` keys, preserving the application's configuration interface; development uses .NET User Secrets. The approach fits the project's existing multi-container VPS setup and needs only limited workflow and application changes. Its limits include the need to manage file permissions, backups and recovery, access controls, and rotation; because values are read at startup, changed secrets require a container restart. Successful CI or container creation alone does not confirm the application is healthy in production.

### Source excerpt

In one of my projects developed with ASP.NET Core, I was using Docker to run the services and GitHub Actions to manage the CI/CD process. The deployment process was automated: with every push to the main branch, tests were executed first, then a new Docker image was built and transferred to the VPS, and finally the previous container was replaced with the new version. While reviewing this process, I decided to make the management of sensitive production information independent of the build and deployment lifecycle as well. It wasn't just about securing secrets; I wanted each part of the infrastructure to have a clearly defined responsibility. Source code would be stored in Git, the Docker image would be responsible for running the application, GitHub Actions would manage the deployment process, and sensitive information such as API keys, database connection strings, and client secrets would remain in the production environment. My goal was to ensure that each deployment changed only the application version, not the way its sensitive information was managed. I had several options to achieve this, from environment variables and GitHub Actions Secrets to dedicated secret management services such as Azure Key Vault, HashiCorp Vault, and AWS Secrets Manager. However, considering the project's existing architecture, which ran on a VPS with multiple Docker containers, I preferred a simpler and more manageable approach. In the end, I chose protected files on the VPS and read-only Docker bind mounts and implemented this structure with a few limited changes to Program.cs and the GitHub Actions workflow. In the following sections, I'll share my experience with this change, the technical decisions I made, and the lessons I learned during implementation. Step 1: Moving Secrets to the VPS The first thing I did was decide where to store sensitive information on the server. I didn't want to create separate, scattered paths for every secret. I preferred a clear structure that could

## Python 3.15 Lazy Imports Move the Crash From Deploy to First Request

DevFeed: [Python 3.15 Lazy Imports Move the Crash From Deploy to First Request](<https://devfeed.tech/articles/python-3-15-lazy-imports-move-the-crash-from-deploy-to-first-request-136241.md>)

Original publisher: [Read original article](<https://blog.devops.dev/python-3-15-lazy-imports-move-the-crash-from-deploy-to-first-request-5489e98b992c?source=rss----33f8b2d9a328---4>)

Author: Daniel Valev

Published: 2026-11-16T13:25:55Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Python](<https://devfeed.tech/topics/python.md>), [import](<https://devfeed.tech/topics/import.md>)

Tags: [deploy](<https://devfeed.tech/tags/deploy.md>), [devops](<https://devfeed.tech/tags/devops.md>), [interpreter](<https://devfeed.tech/tags/interpreter.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [lazy-import](<https://devfeed.tech/tags/lazy-import.md>), [programming](<https://devfeed.tech/tags/programming.md>), [python](<https://devfeed.tech/tags/python.md>), [python-3-15](<https://devfeed.tech/tags/python-3-15.md>), [request](<https://devfeed.tech/tags/request.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>)

### AI overview

Python 3.15 lazy imports can defer a missing-module failure until the imported name is first used. In the excerpt's example, a service passes readiness checks after adding `lazy` to imports, then `/config` fails on its first request with `ModuleNotFoundError: No module named 'yaml'`. Lazy imports defer when the module is loaded; they do not remove import errors. The excerpt introduces a boot check intended to surface `ImportError` earlier, but cuts off before explaining how that check works.

### Source excerpt

What the interpreter binds, when it reifies, and the boot check that puts the ImportError back where you can see it. Continue reading on DevOps.dev "

## Every AKS Node You Can Patch in Place Is a Node That Can Drift

DevFeed: [Every AKS Node You Can Patch in Place Is a Node That Can Drift](<https://devfeed.tech/articles/every-aks-node-you-can-patch-in-place-is-a-node-that-can-drift-136236.md>)

Original publisher: [Read original article](<https://blog.devops.dev/aks-node-drift-azure-container-linux-immutable-os-90df033e23ac?source=rss----33f8b2d9a328---4>)

Author: Dave R | Microsoft Azure & AI MVP ☁

Published: 2026-11-16T13:25:26Z

Content type: tutorial

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Azure Kubernetes Service](<https://devfeed.tech/topics/azure-kubernetes-service.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [Deployment Strategies](<https://devfeed.tech/topics/deployment-strategies.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [aks](<https://devfeed.tech/tags/aks.md>), [azure](<https://devfeed.tech/tags/azure.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [commands](<https://devfeed.tech/tags/commands.md>), [container](<https://devfeed.tech/tags/container.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [devops](<https://devfeed.tech/tags/devops.md>), [guard](<https://devfeed.tech/tags/guard.md>), [image](<https://devfeed.tech/tags/image.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [layer](<https://devfeed.tech/tags/layer.md>), [linux](<https://devfeed.tech/tags/linux.md>), [node](<https://devfeed.tech/tags/node.md>), [nodepool](<https://devfeed.tech/tags/nodepool.md>), [os](<https://devfeed.tech/tags/os.md>), [patches](<https://devfeed.tech/tags/patches.md>), [patching](<https://devfeed.tech/tags/patching.md>), [pool](<https://devfeed.tech/tags/pool.md>), [secure-boot](<https://devfeed.tech/tags/secure-boot.md>), [ubuntu](<https://devfeed.tech/tags/ubuntu.md>), [update](<https://devfeed.tech/tags/update.md>), [weekly](<https://devfeed.tech/tags/weekly.md>)

### AI overview

Azure Container Linux (ACL) replaces patchable AKS node operating systems with signed, read-only images, reducing the risk of nodes drifting apart; nodes are replaced weekly instead of patched in place. The excerpt says ACL is generally available from Kubernetes 1.34 and is the documented destination for Azure Linux with OS Guard, whose patches end December 10, 2026. Existing Ubuntu or Azure Linux pools can be moved in place with Azure CLI 2.86.0 or later; AKS reimages the pool onto ACL. The excerpt does not provide the promised verification steps or operational limitations.

### Source excerpt

How Azure Container Linux locks down the node OS layer by layer, and how to move off OS Guard before its patches end December 10 Continue reading on DevOps.dev "

## Your Kubernetes Knowledge Is Becoming Cheaper. Your Production Judgment Isn't.

DevFeed: [Your Kubernetes Knowledge Is Becoming Cheaper. Your Production Judgment Isn't.](<https://devfeed.tech/articles/your-kubernetes-knowledge-is-becoming-cheaper-your-production-judgment-isn-t-136242.md>)

Original publisher: [Read original article](<https://blog.devops.dev/your-kubernetes-knowledge-is-becoming-cheaper-your-production-judgment-isnt-cabf2adfc61a?source=rss----33f8b2d9a328---4>)

Author: Arvind Verma

Published: 2026-11-16T13:25:19Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Grafana](<https://devfeed.tech/topics/grafana.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [Prometheus](<https://devfeed.tech/topics/prometheus.md>), [Terraform](<https://devfeed.tech/topics/terraform.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Chaos Engineering](<https://devfeed.tech/topics/chaos-engineering.md>), [AIOps for Kubernetes](<https://devfeed.tech/topics/aiops-for-kubernetes.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [api](<https://devfeed.tech/tags/api.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bash](<https://devfeed.tech/tags/bash.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [ci-cd-pipeline](<https://devfeed.tech/tags/ci-cd-pipeline.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [commands](<https://devfeed.tech/tags/commands.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [decision](<https://devfeed.tech/tags/decision.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [devops](<https://devfeed.tech/tags/devops.md>), [devops-engineer](<https://devfeed.tech/tags/devops-engineer.md>), [docker](<https://devfeed.tech/tags/docker.md>), [english](<https://devfeed.tech/tags/english.md>), [git](<https://devfeed.tech/tags/git.md>), [grafana](<https://devfeed.tech/tags/grafana.md>), [helm](<https://devfeed.tech/tags/helm.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [production](<https://devfeed.tech/tags/production.md>)

### AI overview

The article argues that AI is making configuration such as Kubernetes YAML, Terraform modules, and CI/CD pipelines cheaper to produce, while production judgment remains central to DevOps. It recommends narrowing investigations with evidence, checking the failure's scope and system layer, and treating dashboards as signals rather than proof of user health. Before restarting, rolling back, draining, scaling, or failing over, assess consequences and blast radius; under uncertainty, seek a safe next test. During incidents, communicate impact, hypothesis, evidence being checked, and next action. It also urges engineers to learn failure modes alongside tool mechanics. These are the article's recommendations, not a comparison of specific AI tools.

### Source excerpt

There was a time when knowing how to write Kubernetes YAML felt like a superpower. You knew Deployments. You knew Services. You knew ConfigMaps. You knew Ingress. You knew how to write a Helm chart without accidentally deleting half the cluster. You learned Terraform. You learned Jenkins. You learned Bash. You learned Prometheus queries. You spent years collecting tools, commands, syntax, and implementation patterns. And then AI showed up. Now you can describe what you want in plain English and get a Terraform module. Ask for a Kubernetes Deployment and you get YAML. Ask for a CI/CD pipeline and you get a pipeline. Ask for a Bash script and you get one. Ask for a PromQL query and, assuming the machines haven't collectively decided to punish humanity that day, you get something reasonably useful. 😄 So here's the uncomfortable question: If AI can increasingly handle implementation, what exactly becomes more valuable for a DevOps engineer? I think the answer is simple. Judgment. Not knowing more commands. Not memorizing more YAML. Not collecting another 47 certifications. Knowing what to do when the obvious answer doesn't work. And production has a beautiful way of ensuring that the obvious answer almost never works. 1 The Shift: From Tool Operator to Decision Maker For years, DevOps careers were heavily associated with tools. Learn AWS. Learn Docker. Learn Kubernetes. Learn Terraform. Learn Jenkins. Learn Ansible. Learn Prometheus. Learn Grafana. Learn Git. Learn Linux. And yes, these skills still matter. The tools aren't suddenly useless because AI can generate configuration for them. But something important is changing. The cost of producing configuration is falling. Consider a simple Kubernetes deployment. Previously, an engineer might spend time writing: apiVersion: apps/v1 kind: Deployment metadata: name: payment-api spec: replicas: 3 selector: matchLabels: app: payment-api template: metadata: labels: app: payment-api spec: containers: - name: payment-api image

## 7 Terraform State File Mistakes That Leak Your Cloud Secrets

DevFeed: [7 Terraform State File Mistakes That Leak Your Cloud Secrets](<https://devfeed.tech/articles/7-terraform-state-file-mistakes-that-leak-your-cloud-secrets-128872.md>)

Original publisher: [Read original article](<https://blog.devops.dev/7-terraform-state-file-mistakes-that-leak-your-cloud-secrets-329b420b9d51?source=rss----33f8b2d9a328---4>)

Author: Vortex 404

Published: 2026-11-11T06:16:13Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Terraform](<https://devfeed.tech/topics/terraform.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>)

Tags: [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [devops](<https://devfeed.tech/tags/devops.md>), [file](<https://devfeed.tech/tags/file.md>), [infrastructure-as-code](<https://devfeed.tech/tags/infrastructure-as-code.md>), [leak](<https://devfeed.tech/tags/leak.md>), [mistakes](<https://devfeed.tech/tags/mistakes.md>), [secrets](<https://devfeed.tech/tags/secrets.md>), [state](<https://devfeed.tech/tags/state.md>), [state-management](<https://devfeed.tech/tags/state-management.md>), [terraform](<https://devfeed.tech/tags/terraform.md>), [terraform-state](<https://devfeed.tech/tags/terraform-state.md>)

### AI overview

Terraform state and plan files can store sensitive infrastructure values in plaintext, and marking a variable sensitive does not remove its value from state. The excerpt cites HashiCorp documentation and a 2024 penetration test involving a publicly readable S3 bucket, where an exposed credential was traced to an administrator-level AWS key in another account. It introduces seven state-file mistakes, but the provided text only begins the first: assuming "sensitive = true" hides the value from state. The remaining mistakes and recommended fixes are not included in this excerpt.

### Source excerpt

Terraform stores sensitive values in plaintext by default -- and marking a variable "sensitive" doesn't change that. Here's what actually... Continue reading on DevOps.dev "

## Cloudflare Workers Can Finally Be Profiled in Production -- And It Already Found Real Bugs

DevFeed: [Cloudflare Workers Can Finally Be Profiled in Production -- And It Already Found Real Bugs](<https://devfeed.tech/articles/cloudflare-workers-can-finally-be-profiled-in-production-and-it-already-found-real-bugs-128873.md>)

Original publisher: [Read original article](<https://blog.devops.dev/cloudflare-workers-can-finally-be-profiled-in-production-and-it-already-found-real-bugs-f3f975846c2c?source=rss----33f8b2d9a328---4>)

Author: Vortex 404

Published: 2026-11-11T06:15:15Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Cloudflare Workers](<https://devfeed.tech/topics/cloudflare-workers.md>), [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>)

Tags: [bugs](<https://devfeed.tech/tags/bugs.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [cloudflare-workers](<https://devfeed.tech/tags/cloudflare-workers.md>), [devops](<https://devfeed.tech/tags/devops.md>), [observability](<https://devfeed.tech/tags/observability.md>), [performance](<https://devfeed.tech/tags/performance.md>), [production](<https://devfeed.tech/tags/production.md>), [profiling](<https://devfeed.tech/tags/profiling.md>), [serverless](<https://devfeed.tech/tags/serverless.md>)

### AI overview

Cloudflare added on-demand CPU and memory profiling for production Workers and Durable Objects. It creates interactive flamegraphs from a live isolate without restarting it or routing it elsewhere, addressing a gap left by local Chrome DevTools profiling when problems appear only under production traffic, data, or concurrency. Cloudflare says its initial use found two shipped issues: a function repeatedly traversing a JSON tree and a monitoring library still running despite being intended as disabled. The excerpt does not provide further implementation details.

### Source excerpt

On-demand CPU and memory flamegraphs for live Workers, built around a locking trick that profiles a running isolate without freezing it Continue reading on DevOps.dev "

## Your NestJS API Is Leaking Password Hashes -- And You Probably Don't Know It

DevFeed: [Your NestJS API Is Leaking Password Hashes -- And You Probably Don't Know It](<https://devfeed.tech/articles/your-nestjs-api-is-leaking-password-hashes-and-you-probably-don-t-know-it-127579.md>)

Original publisher: [Read original article](<https://blog.devops.dev/your-nestjs-api-is-leaking-password-hashes-and-you-probably-dont-know-it-dcaea709a987?source=rss----33f8b2d9a328---4>)

Author: Mobin Shaterian

Published: 2026-11-10T21:46:01Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [API](<https://devfeed.tech/topics/api.md>), [Prisma](<https://devfeed.tech/topics/prisma.md>), [Nest](<https://devfeed.tech/topics/nestjs.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [api-security](<https://devfeed.tech/tags/api-security.md>), [nestjs](<https://devfeed.tech/tags/nestjs.md>), [owasp](<https://devfeed.tech/tags/owasp.md>), [password](<https://devfeed.tech/tags/password.md>), [passwords](<https://devfeed.tech/tags/passwords.md>), [prisma](<https://devfeed.tech/tags/prisma.md>), [security](<https://devfeed.tech/tags/security.md>), [typescript](<https://devfeed.tech/tags/typescript.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

The article describes a NestJS API that passed client-supplied Prisma `include` objects directly into queries, allowing authenticated users to retrieve related sensitive fields such as password hashes. It identifies this read-side exposure as Broken Object Property Level Authorization (BOPLA), distinct from checks that only authorize access to an object. The article also describes schema probing through Prisma errors and potential query amplification from deeply nested includes. Its proposed fix is to remove raw `include` input, expose only typed flags for predefined, whitelisted relation selections, and return explicit response DTOs or serialize and map models to omit sensitive fields. Separately, it notes that stored script tags can become dangerous when rendered in other contexts, and recommends sanitizing free-text values and configuring NestJS `ValidationPipe` to whitelist properties.

### Source excerpt

A deep dive into Broken Object Property Level Authorization (BOPLA) -- the silent API vulnerability hiding in your Prisma includes I recently audited a NestJS API and found something that made me pause: a single curl command returned password hashes for every user in a module. No special privileges. No token manipulation. Just a crafted query parameter. The codebase was clean. TypeScript was strict. The developer was clearly experienced. And yet, a classic API vulnerability was sitting wide open. This is the story of that vulnerability, how it works, and how to fix it. The Setup The app had an address book module built with NestJS + Prisma. A GET /module endpoint returned paginated results, and it supported an include query parameter to load related data (owner info, user details, addresses). The intent was flexible -- let the frontend decide what data it needs. The result was a security hole. Here's the request that exposed it: GET /module?include={"owner":{"include":{"accounts":true}}} And here's a snippet of what came back: { "owner": { "accounts": [{ "password": "$v=19$m=65536,t=3,p=4$unxyh6bG...", "accessToken": null, "refreshToken": null }] } } A hashed password. In a public API response. Accessible to any authenticated user. What is BOPLA? Broken Object Property Level Authorization (BOPLA) is listed as API3:2023 in the OWASP API Security Top 10. It occurs when: An API exposes more object properties than the client is authorized to access -- either by failing to filter output or by letting the client define what gets included in the response. It's closely related to the more well-known Mass Assignment vulnerability, but BOPLA focuses on the read side: you can retrieve properties you shouldn't be able to see. The distinction matters. Authorization checks (guards, policies, RBAC) typically answer "can this user access this object?" -- think BOLA (Broken Object Level Authorization). BOPLA asks a different question: "can this user access this field on an object they'r

## Stop Rendering AI Video Drafts on Your Most Expensive Model

DevFeed: [Stop Rendering AI Video Drafts on Your Most Expensive Model](<https://devfeed.tech/articles/stop-rendering-ai-video-drafts-on-your-most-expensive-model-127574.md>)

Original publisher: [Read original article](<https://blog.devops.dev/microsoft-foundry-video-generation-kling-ltx-minimax-cost-158bcbdb26db?source=rss----33f8b2d9a328---4>)

Author: Dave R | Microsoft Azure & AI MVP ☁

Published: 2026-11-10T09:43:11Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [multimodal-ai](<https://devfeed.tech/topics/multimodal-ai.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-video](<https://devfeed.tech/tags/ai-video.md>), [api](<https://devfeed.tech/tags/api.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [async](<https://devfeed.tech/tags/async.md>), [auth](<https://devfeed.tech/tags/auth.md>), [azure](<https://devfeed.tech/tags/azure.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cost](<https://devfeed.tech/tags/cost.md>), [data](<https://devfeed.tech/tags/data.md>), [data-residency](<https://devfeed.tech/tags/data-residency.md>), [draft](<https://devfeed.tech/tags/draft.md>), [entra-id](<https://devfeed.tech/tags/entra-id.md>), [foundry](<https://devfeed.tech/tags/foundry.md>), [microsoft-foundry](<https://devfeed.tech/tags/microsoft-foundry.md>), [model](<https://devfeed.tech/tags/model.md>), [pricing](<https://devfeed.tech/tags/pricing.md>), [programming](<https://devfeed.tech/tags/programming.md>), [python](<https://devfeed.tech/tags/python.md>), [rendering](<https://devfeed.tech/tags/rendering.md>), [software-development](<https://devfeed.tech/tags/software-development.md>), [stage](<https://devfeed.tech/tags/stage.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

Microsoft Foundry's public preview offers five fal video models--Kling, LTX, and MiniMax variants--through one resource and keyless Microsoft Entra ID sign-in. The excerpt says routing drafts, previews, and finals to different models can reduce the published cost for 270 seconds of video from $49.90 to $15.74. It cautions that the pricing sentence says "per minute" while the table header says "per video second"; budget in seconds and verify rates on Azure's pricing page, where Microsoft says they will appear in the coming weeks. The article also covers the async API in Python and data processing location, but this excerpt provides no implementation or residency details.

### Source excerpt

Kling, LTX, and MiniMax on Microsoft Foundry: per-stage pricing, keyless Entra ID auth, the async API in Python, and data residency Continue reading on DevOps.dev "

## I Tried to Build an Operating System With AI. One Year Later, It Runs on 20-Year-Old Computers.

DevFeed: [I Tried to Build an Operating System With AI. One Year Later, It Runs on 20-Year-Old Computers.](<https://devfeed.tech/articles/i-tried-to-build-an-operating-system-with-ai-one-year-later-it-runs-on-20-year-old-computers-127572.md>)

Original publisher: [Read original article](<https://blog.devops.dev/i-tried-to-build-an-operating-system-with-ai-one-year-later-it-runs-on-20-year-old-computers-f108cac56f9d?source=rss----33f8b2d9a328---4>)

Author: SmileyTech.mk

Published: 2026-11-10T09:43:04Z

Content type: opinion

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Kernel](<https://devfeed.tech/topics/kernel.md>), [pc](<https://devfeed.tech/topics/pc.md>), [virtual machines](<https://devfeed.tech/topics/virtual-machines.md>), [Emulator](<https://devfeed.tech/topics/emulator.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [computers](<https://devfeed.tech/tags/computers.md>), [devos](<https://devfeed.tech/tags/devos.md>), [operating-system](<https://devfeed.tech/tags/operating-system.md>), [operating-systems](<https://devfeed.tech/tags/operating-systems.md>), [raspberry-pi](<https://devfeed.tech/tags/raspberry-pi.md>), [rust](<https://devfeed.tech/tags/rust.md>)

### AI overview

SmileyOS is a roughly 20 MB bare-metal operating system written from scratch in Rust, designed for computers with as little as 256 MB of RAM. Its developer says AI coding assistants helped him build it, while hardware debugging required repeated experiments and eventually tools such as network logging. It has booted on several older PCs and a Raspberry Pi 3 Model B, and can run from USB or microSD without replacing an existing laptop OS. The system includes a graphical desktop, networking, a browser, development tools, and more than 30 apps and games; classroom collaboration can work over a local network without internet. Hardware support remains limited, and SBOOK Go and SBOX are still prototypes. A first public commercial release is planned for 2027.

### Source excerpt

How one developer, a lot of enthusiasm, and AI coding tools turned a crazy experiment into SmileyOS -- a tiny operating system built from scratch in Rust. It started with a ridiculous promise. In late 2025, during a discussion with colleagues about artificial intelligence and software development, I made a rather bold claim: With AI, I could build an operating system from scratch in five minutes. We tried. It didn't work. So I made another promise: give me one week, and I'll have an operating system that can at least display "Hello World." One week later, I had a basic terminal, a text editor, a graphical interface and a working Snake game running in a virtual machine. And I couldn't stop. What started as a little experiment turned into something I never seriously expected to build. Today, that experiment is called SmileyOS. SmileyOS + SBOOK Go: a simple computer for every kid | SmileyTech.mk It's an independent, bare-metal operating system written in Rust, with its own kernel, drivers, graphical desktop, applications, games, programming language and development tools. It isn't a Linux distribution or a customized version of an existing operating system. And perhaps the most surprising part? The entire system is around 20 MB and is designed to run on computers with as little as 256 MB of RAM. Computers that most people would probably throw away. SmileyOS desktop -- a real screenshot of the operating system.I'm not a big technology company. I'm one developer with a dream. My name is Nikola Gjorgiev. I'm a software engineer from Macedonia, and I've spent much of my life building software. Nikola Gjorgiev - Abysalto | LinkedIn I've worked on video games, websites, mobile applications and desktop software. But building my own operating system has fascinated me since I was a teenager. Around 2005, I was experimenting with C++, reading about kernels, memory and assembly language, and imagining what it would be like to create an entire operating system myself. Back then, it

## Meta Skipped Its Own Deploy Freeze for the World Cup Final.

DevFeed: [Meta Skipped Its Own Deploy Freeze for the World Cup Final.](<https://devfeed.tech/articles/meta-skipped-its-own-deploy-freeze-for-the-world-cup-final-127573.md>)

Original publisher: [Read original article](<https://blog.devops.dev/meta-skipped-its-own-deploy-freeze-for-the-world-cup-final-4d61566792b2?source=rss----33f8b2d9a328---4>)

Author: Vortex 404

Published: 2026-11-10T09:43:01Z

Content type: news

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Disaster Recovery](<https://devfeed.tech/topics/disaster-recovery.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [argentina](<https://devfeed.tech/tags/argentina.md>), [code](<https://devfeed.tech/tags/code.md>), [cup](<https://devfeed.tech/tags/cup.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [devops](<https://devfeed.tech/tags/devops.md>), [disaster-recovery](<https://devfeed.tech/tags/disaster-recovery.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [event](<https://devfeed.tech/tags/event.md>), [freeze](<https://devfeed.tech/tags/freeze.md>), [history](<https://devfeed.tech/tags/history.md>), [load](<https://devfeed.tech/tags/load.md>), [meta](<https://devfeed.tech/tags/meta.md>), [outages](<https://devfeed.tech/tags/outages.md>), [platform](<https://devfeed.tech/tags/platform.md>), [regions](<https://devfeed.tech/tags/regions.md>), [spain](<https://devfeed.tech/tags/spain.md>), [sre](<https://devfeed.tech/tags/sre.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [whatsapp](<https://devfeed.tech/tags/whatsapp.md>), [world](<https://devfeed.tech/tags/world.md>), [zero](<https://devfeed.tech/tags/zero.md>)

### AI overview

During the July 19, 2026 Argentina-Spain World Cup final, WhatsApp delivered more than 30 million messages per second, 1.23 times its previous all-time peak. The excerpt says Meta kept its usual engineering cadence across 48 teams and 104 matches, without shifting load between regions, imposing a code freeze, or pausing disaster-recovery drills. Meta reported zero outages or degradations. The excerpt does not include the engineering details behind those results.

### Source excerpt

No regional load-shifting, no code freeze, no paused disaster-recovery drills -- during the biggest traffic event in company history, Meta... Continue reading on DevOps.dev "

## How Do You Design a Highly Available Architecture in AWS? A Complete DevOps Engineer's Guide

DevFeed: [How Do You Design a Highly Available Architecture in AWS? A Complete DevOps Engineer's Guide](<https://devfeed.tech/articles/how-do-you-design-a-highly-available-architecture-in-aws-a-complete-devops-engineer-s-guide-127571.md>)

Original publisher: [Read original article](<https://blog.devops.dev/how-do-you-design-a-highly-available-architecture-in-aws-a-complete-devops-engineers-guide-fe49b3cc18d8?source=rss----33f8b2d9a328---4>)

Author: Arvind Verma

Published: 2026-11-10T09:42:59Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Cloud Architecture](<https://devfeed.tech/topics/cloud-architecture.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [reliability-engineering](<https://devfeed.tech/topics/reliability-engineering.md>), [system-design-project](<https://devfeed.tech/topics/system-design-project.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>), [Amazon Route 53](<https://devfeed.tech/topics/amazon-route-53.md>), [Amazon VPC](<https://devfeed.tech/topics/amazon-vpc.md>), [DevOps](<https://devfeed.tech/topics/devops.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [auto-scaling](<https://devfeed.tech/tags/auto-scaling.md>), [availability](<https://devfeed.tech/tags/availability.md>), [aws](<https://devfeed.tech/tags/aws.md>), [cloudfront](<https://devfeed.tech/tags/cloudfront.md>), [database](<https://devfeed.tech/tags/database.md>), [design](<https://devfeed.tech/tags/design.md>), [designing](<https://devfeed.tech/tags/designing.md>), [devops](<https://devfeed.tech/tags/devops.md>), [devops-engineer](<https://devfeed.tech/tags/devops-engineer.md>), [disaster-recovery](<https://devfeed.tech/tags/disaster-recovery.md>), [docker](<https://devfeed.tech/tags/docker.md>), [e-commerce](<https://devfeed.tech/tags/e-commerce.md>), [ec2](<https://devfeed.tech/tags/ec2.md>), [high-availability](<https://devfeed.tech/tags/high-availability.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [implementation](<https://devfeed.tech/tags/implementation.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [interview](<https://devfeed.tech/tags/interview.md>), [interview-questions](<https://devfeed.tech/tags/interview-questions.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [load-balancers](<https://devfeed.tech/tags/load-balancers.md>)

### AI overview

A highly available AWS design spreads application components across multiple Availability Zones so a single component or AZ failure does not take down the service. The guide's example uses Route 53 for DNS, WAF for request filtering, an internet facing ALB to route traffic to healthy EC2 targets in private subnets, and an Auto Scaling Group to replace unhealthy instances and adjust capacity. It pairs RDS Multi-AZ failover with application retry and reconnection logic, while noting that a traditional standby is not for read scaling. S3 and CloudFront serve static assets; CloudWatch alarms and SNS notify operators. The guide distinguishes regional high availability from disaster recovery, recommending backups, recovery objectives, cross-Region plans where needed, and tested recovery. It also flags costs, DNS caching delays, NAT Gateway failure dependencies, and the need to test actual failure behavior.

### Source excerpt

From Route 53 and Application Load Balancers to Multi-AZ Auto Scaling, RDS, CloudFront, and Disaster Recovery. Understand how to design AWS infrastructure that survives failures. Imagine you're working as a DevOps Engineer at an e-commerce company. Your application handles thousands of customer requests every minute. Customers are browsing products, placing orders, making payments, and checking delivery status. Everything is working perfectly. Suddenly, one AWS Availability Zone experiences an infrastructure failure. Your EC2 instances become unavailable. Some application requests start failing. Customers cannot access your website. Your engineering manager asks: "How did our entire application become unavailable because one Availability Zone failed?" This is where understanding High Availability (HA) becomes important. A production-ready architecture should not depend on a single EC2 instance, one Availability Zone, one database instance, or one network component. Instead, infrastructure should be designed to tolerate failures, recover automatically, and maintain application availability. In this article, we'll understand how to design a highly available AWS architecture from scratch, including the architecture, AWS services, implementation decisions, production scenarios, and common interview questions. 📘 SECTION 1: UNDERSTANDING HIGH AVAILABILITY AND AWS ARCHITECTURE1. What Is High Availability? High Availability means designing systems that remain accessible and operational even when individual infrastructure components fail. Let's understand with a simple example. Imagine your application runs on one EC2 instance. If that instance crashes, your application becomes unavailable. This is called a Single Point of Failure (SPOF). Now imagine running six EC2 instances across two Availability Zones. Availability Zone A contains three instances. Availability Zone B contains three instances. If Availability Zone A experiences a failure, instances in Availability Zone B

## What Breaks When Let's Encrypt Moves to 64-Day Certificates

DevFeed: [What Breaks When Let's Encrypt Moves to 64-Day Certificates](<https://devfeed.tech/articles/what-breaks-when-let-s-encrypt-moves-to-64-day-certificates-127578.md>)

Original publisher: [Read original article](<https://blog.devops.dev/what-breaks-when-lets-encrypt-moves-to-64-day-certificates-c22506d8cd2d?source=rss----33f8b2d9a328---4>)

Author: Daniel Valev

Published: 2026-11-10T09:42:37Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [let's encrypt](<https://devfeed.tech/topics/let-s-encrypt.md>), [Secrets Management](<https://devfeed.tech/topics/secrets-management.md>)

Tags: [acme](<https://devfeed.tech/tags/acme.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [calendar](<https://devfeed.tech/tags/calendar.md>), [certbot](<https://devfeed.tech/tags/certbot.md>), [certificates](<https://devfeed.tech/tags/certificates.md>), [devops](<https://devfeed.tech/tags/devops.md>), [dns](<https://devfeed.tech/tags/dns.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [let-s-encrypt](<https://devfeed.tech/tags/let-s-encrypt.md>), [lets-encrypt](<https://devfeed.tech/tags/lets-encrypt.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [staging](<https://devfeed.tech/tags/staging.md>), [tls](<https://devfeed.tech/tags/tls.md>)

### AI overview

Let's Encrypt staging began issuing 64-day certificates on October 14, 2026, with production scheduled to follow on February 10, 2027. The excerpt says ACME clients are likely to handle the change; surrounding automation may not. Calendar-based renewal jobs that assume 90-day lifetimes, alerts fixed at 30 days before expiry, and DNS-01 credentials may need review because renewals will happen more often. It also cautions that the common claim that a 30-day alert threshold halves the safety margin is wrong: under the described renewal behavior, renewal occurs at day 34. The excerpt recommends testing fixes in staging but does not include the detailed calculations or steps.

### Source excerpt

Your ACME client will be fine on Feb 10. Calendar crons, 30-day alerts and DNS-01 setups won't, and you can fix all three on staging now. Continue reading on DevOps.dev "

## The Hidden Cost of "Free" NuGet Packages

DevFeed: [The Hidden Cost of "Free" NuGet Packages](<https://devfeed.tech/articles/the-hidden-cost-of-free-nuget-packages-127577.md>)

Original publisher: [Read original article](<https://blog.devops.dev/the-hidden-cost-of-free-nuget-packages-b094d44c98ce?source=rss----33f8b2d9a328---4>)

Author: Bharat Chandera

Published: 2026-11-10T09:42:23Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [NuGet](<https://devfeed.tech/topics/nuget.md>), [.NET](<https://devfeed.tech/topics/net.md>)

Tags: [audit](<https://devfeed.tech/tags/audit.md>), [change](<https://devfeed.tech/tags/change.md>), [costs](<https://devfeed.tech/tags/costs.md>), [csharp](<https://devfeed.tech/tags/csharp.md>), [dependency](<https://devfeed.tech/tags/dependency.md>), [dotnet](<https://devfeed.tech/tags/dotnet.md>), [licensing](<https://devfeed.tech/tags/licensing.md>), [maintainers](<https://devfeed.tech/tags/maintainers.md>), [net](<https://devfeed.tech/tags/net.md>), [nuget](<https://devfeed.tech/tags/nuget.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [software-development](<https://devfeed.tech/tags/software-development.md>)

### AI overview

NuGet packages being free to download does not make them cost-free to maintain or use: a dependency may change its license in a future major release, lose its sole maintainer, or create security-audit work when its licenses are unknown. The excerpt describes license changes as often following a popular library's maintainer facing unsustainable unpaid work, and cautions against treating maintainers as villains. It introduces a process to catch such costs early, but the supplied text ends before providing the checklist or further steps.

### Source excerpt

A practical checklist for spotting licensing traps, abandoned maintainers, and surprise bills before they land in your .NET project. Continue reading on DevOps.dev "

## Self-Service Namespace Onboarding on Amazon EKS: Automating Access Entries and RBAC with GitHub...

DevFeed: [Self-Service Namespace Onboarding on Amazon EKS: Automating Access Entries and RBAC with GitHub...](<https://devfeed.tech/articles/self-service-namespace-onboarding-on-amazon-eks-automating-access-entries-and-rbac-with-github-127576.md>)

Original publisher: [Read original article](<https://blog.devops.dev/self-service-namespace-onboarding-on-amazon-eks-automating-access-entries-and-rbac-with-github-74f0347e71e2?source=rss----33f8b2d9a328---4>)

Author: Oleh Adam Dubnytskyy

Published: 2026-11-10T09:42:12Z

Content type: tutorial

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [rbac](<https://devfeed.tech/topics/rbac.md>), [Amazon Elastic Kubernetes Service](<https://devfeed.tech/topics/amazon-elastic-kubernetes-service.md>), [Access Control](<https://devfeed.tech/topics/access-control.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [GitHub Actions](<https://devfeed.tech/topics/github-actions.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [AWS Security Token Service](<https://devfeed.tech/topics/aws-security-token-service.md>), [AWS IAM](<https://devfeed.tech/topics/aws-iam.md>), [self-service](<https://devfeed.tech/topics/self-service.md>), [internal developer platform automation](<https://devfeed.tech/topics/internal-developer-platform-automation.md>), [GitHub](<https://devfeed.tech/topics/github.md>)

Tags: [admin](<https://devfeed.tech/tags/admin.md>), [amazon-eks](<https://devfeed.tech/tags/amazon-eks.md>), [amazon-web-services](<https://devfeed.tech/tags/amazon-web-services.md>), [api](<https://devfeed.tech/tags/api.md>), [api-server](<https://devfeed.tech/tags/api-server.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [automation](<https://devfeed.tech/tags/automation.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-eks](<https://devfeed.tech/tags/aws-eks.md>), [ci](<https://devfeed.tech/tags/ci.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [devops](<https://devfeed.tech/tags/devops.md>), [eks](<https://devfeed.tech/tags/eks.md>), [github](<https://devfeed.tech/tags/github.md>), [github-actions](<https://devfeed.tech/tags/github-actions.md>), [iam](<https://devfeed.tech/tags/iam.md>), [job](<https://devfeed.tech/tags/job.md>), [keys](<https://devfeed.tech/tags/keys.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [onboarding](<https://devfeed.tech/tags/onboarding.md>), [rbac](<https://devfeed.tech/tags/rbac.md>), [self-service](<https://devfeed.tech/tags/self-service.md>)

### AI overview

The excerpt describes namespace onboarding on Amazon EKS as two coordinated permission updates: an EKS access policy scope and Kubernetes RBAC. GitHub Actions authenticates through OIDC and temporary IAM role credentials. The EKS authorizer and RBAC authorizer can each allow a request, so effective permissions are their union; a restrictive custom RBAC role cannot narrow permissions already granted by a broader managed policy. The proposed setup uses separate workflows: one applies ClusterRoles and per-namespace RoleBindings, while another updates the access entry's namespace scope from files in a namespaces directory. The scripts validate namespace names, reject empty desired scopes, skip unchanged access-policy updates, and optionally prune stale managed RoleBindings. The excerpt notes that RBAC synchronization requires cluster-scoped rights and that missing namespaces are skipped with a warning before the workflow reports an error. The supplied text ends after describing this implementation and its pull request workflow.

### Source excerpt

Self-Service Namespace Onboarding on Amazon EKS: Automating Access Entries and RBAC with GitHub ActionsTwo authorization layers on Amazon EKS Every call my pipeline makes to the cluster's Kubernetes API server passes through two authorizers: Kubernetes RBAC and EKS access policies. Authentication: the access entry says who you are The pipeline never holds long-lived AWS keys. A GitHub Actions job gets an OIDC token, exchanges it for temporary credentials of an IAM role, and calls the cluster with them. More details on how it works can be seen here. Layer 1: Kubernetes RBAC RBAC is the standard Kubernetes model: Roles and ClusterRoles define rules, and RoleBindings attach them to a user or group. These objects live inside the cluster. You can write any rule you need, including rules for custom resources such as KEDA ScaledObject. Layer 2: EKS access policies Access policies are permission sets that AWS manages and that you attach to an access entry through the EKS API. They mirror the Kubernetes user-facing roles: AmazonEKSEditPolicy corresponds to edit, AmazonEKSAdminPolicy to admin, and so on. Each association has a scope: the whole cluster (type=cluster) or a list of namespaces (type=namespace). You can't modify an access policy or create your own. When the managed policies don't fit, more control can be configured by leveraging Kubernetes RBAC objects instead. How the two layers combine The API server chains its authorizers. The RBAC authorizer runs first and returns immediately if it allows the request. If RBAC has no opinion, the request goes to the EKS authorizer, which checks the access policies. If neither allows it, the request is denied. The access entry only establishes who the caller is; RBAC and access policies each get one chance to allow the request. Both authorizers can only allow or pass, so the effective permissions of ci-github-actions in a namespace are the union of both layers. That has two consequences for the automation: Onboarding a namespace

## CloudFormation Got Faster Because Agents Deploy Now. Your Agent Still Reads COMPLETE as Ready.

DevFeed: [CloudFormation Got Faster Because Agents Deploy Now. Your Agent Still Reads COMPLETE as Ready.](<https://devfeed.tech/articles/cloudformation-got-faster-because-agents-deploy-now-your-agent-still-reads-complete-as-ready-127570.md>)

Original publisher: [Read original article](<https://blog.devops.dev/cloudformation-got-faster-because-agents-deploy-now-your-agent-still-reads-complete-as-ready-d41a8e156978?source=rss----33f8b2d9a328---4>)

Author: Vitor Hansen

Published: 2026-11-10T09:42:04Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [AWS CloudFormation](<https://devfeed.tech/topics/aws-cloudformation.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [amazon-vpc](<https://devfeed.tech/tags/amazon-vpc.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-cdk](<https://devfeed.tech/tags/aws-cdk.md>), [aws-cloudformation](<https://devfeed.tech/tags/aws-cloudformation.md>), [aws-lambda](<https://devfeed.tech/tags/aws-lambda.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cdk](<https://devfeed.tech/tags/cdk.md>), [ci](<https://devfeed.tech/tags/ci.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [cloudformation](<https://devfeed.tech/tags/cloudformation.md>), [coding](<https://devfeed.tech/tags/coding.md>), [complete](<https://devfeed.tech/tags/complete.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [devops](<https://devfeed.tech/tags/devops.md>), [express](<https://devfeed.tech/tags/express.md>), [faster](<https://devfeed.tech/tags/faster.md>), [iam](<https://devfeed.tech/tags/iam.md>), [infrastructure-as-code](<https://devfeed.tech/tags/infrastructure-as-code.md>)

### AI overview

CloudFormation Express mode can finish a stack operation before resources are ready, so an agent that treats `CREATE_COMPLETE` as a readiness signal may call an endpoint too soon, misdiagnose the failure, and redeploy. The author proposes a small, credential-free script that inspects a synthesized CDK template, recommends deployment mode and rollback settings, and prints readiness checks for resources referenced by outputs or explicitly named with `--wait-for`. Agents or pipelines run those checks after deployment and before using outputs; gate failures should be treated as infrastructure readiness failures. The excerpt reports up to 4x faster Express deployments, with rollback off by default, and up to 2x faster default-mode deployments that retain stabilization semantics. Gates can give back some Express speedup. The script covers only 13 resource types, does not follow nested stacks, and may misclassify unlisted resources; the author's own timing runs remain pending.

### Source excerpt

The first time I let a coding agent iterate on a CDK stack for a customer demo, it did what agents do. It ran cdk deploy, waited for the green line, read the stack outputs, and immediately called the URL it found there. The call failed. The agent read the failure as a bug in the code it had just written, changed the code, and deployed again. Nothing was wrong with the code. The resource behind the output simply wasn't serving yet. That was before express mode. This year CloudFormation shipped three changes that make that loop much faster, and AWS says plainly who they are for. The June express mode launch talks about developers and AI tools iterating on infrastructure. The October speedup to the default mode lists developers, CI/CD pipelines and AI agents. The force rollback announcement names AI agents among the workflows it supports. I think all three are good changes, and I recommend express mode to customers. My argument here is narrower. Express mode gets most of its speed by changing what COMPLETE means, and an agent is the caller most likely to keep reading it the old way. So the speedup is safe in an agent loop only if something sits between COMPLETE and the agent's next tool call. I wrote that something as a small script, and this post walks through it. What actually shipped in 2026 Three dated, public changes, all in CloudFormation and CDK: Express mode, June 30. A stack operation completes as soon as CloudFormation applies each resource's configuration, without waiting for the resource to stabilize. AWS says deployments get up to 4x faster, based on internal benchmarks. It works with existing templates, change sets and nested stacks, at no extra cost, in all commercial Regions (São Paulo included). Rollback is disabled by default in this mode. Default mode up to 2x faster, October 1. No opt-in. AWS reports the biggest gains on IAM, Amazon S3, AWS Lambda and Amazon VPC resources, again from internal benchmarks. The default mode still runs its stabilization

## Moving Away from PATs in Azure DevOps: A Practical Guide to Cross-Organization Package Consumption

DevFeed: [Moving Away from PATs in Azure DevOps: A Practical Guide to Cross-Organization Package Consumption](<https://devfeed.tech/articles/moving-away-from-pats-in-azure-devops-a-practical-guide-to-cross-organization-package-consumption-127575.md>)

Original publisher: [Read original article](<https://blog.devops.dev/moving-away-from-pats-in-azure-devops-a-practical-guide-to-cross-organization-package-consumption-2c090dc308a9?source=rss----33f8b2d9a328---4>)

Author: Bharani

Published: 2026-11-10T09:41:59Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [Azure DevOps](<https://devfeed.tech/topics/azure-devops.md>), [Entra ID](<https://devfeed.tech/topics/entra-id.md>), [pip](<https://devfeed.tech/topics/pip.md>), [Azure](<https://devfeed.tech/topics/azure.md>)

Tags: [access-tokens](<https://devfeed.tech/tags/access-tokens.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [azure](<https://devfeed.tech/tags/azure.md>), [azure-artifacts](<https://devfeed.tech/tags/azure-artifacts.md>), [azure-databricks](<https://devfeed.tech/tags/azure-databricks.md>), [azure-devops](<https://devfeed.tech/tags/azure-devops.md>), [azure-pipelines](<https://devfeed.tech/tags/azure-pipelines.md>), [entra-id](<https://devfeed.tech/tags/entra-id.md>), [identity](<https://devfeed.tech/tags/identity.md>), [managed-identity](<https://devfeed.tech/tags/managed-identity.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-entra](<https://devfeed.tech/tags/microsoft-entra.md>), [microsoft-entra-id](<https://devfeed.tech/tags/microsoft-entra-id.md>), [package-management](<https://devfeed.tech/tags/package-management.md>), [packages](<https://devfeed.tech/tags/packages.md>), [personal-access-tokens](<https://devfeed.tech/tags/personal-access-tokens.md>), [python](<https://devfeed.tech/tags/python.md>), [python-packages](<https://devfeed.tech/tags/python-packages.md>), [upstream](<https://devfeed.tech/tags/upstream.md>), [workload-identity](<https://devfeed.tech/tags/workload-identity.md>), [workload-identity-federation](<https://devfeed.tech/tags/workload-identity-federation.md>)

### AI overview

The excerpt frames cross-organization Azure Artifacts consumption as a challenge when security policies prohibit PATs: one team publishes internal Python packages, and another needs them through tools such as pip, npm, Maven, or NuGet. Although Service Principal or Workload Identity Federation authentication may seem suitable, package feeds behave differently from Azure resources such as Storage Accounts, Key Vaults, and Container Registries. The article says it will explore consuming packages without exposing PATs directly to pipelines and using Azure Artifacts upstream sources to simplify distribution; this excerpt does not provide the configuration steps.

### Source excerpt

Organizations are increasingly retiring Personal Access Tokens (PATs) in favor of modern identity-based authentication mechanisms such as... Continue reading on DevOps.dev "

## The 95% Confident Agent That Was Right 62% of the Time

DevFeed: [The 95% Confident Agent That Was Right 62% of the Time](<https://devfeed.tech/articles/the-95-confident-agent-that-was-right-62-of-the-time-124914.md>)

Original publisher: [Read original article](<https://blog.devops.dev/the-95-confident-agent-that-was-right-62-of-the-time-f638b16909d1?source=rss----33f8b2d9a328---4>)

Author: ObsNSods

Published: 2026-11-09T20:46:01Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [trust](<https://devfeed.tech/topics/trust.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [about](<https://devfeed.tech/tags/about.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic-ai-pipeline](<https://devfeed.tech/tags/agentic-ai-pipeline.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent-development](<https://devfeed.tech/tags/ai-agent-development.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alert-triage](<https://devfeed.tech/tags/alert-triage.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [article](<https://devfeed.tech/tags/article.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [build](<https://devfeed.tech/tags/build.md>), [building](<https://devfeed.tech/tags/building.md>), [but](<https://devfeed.tech/tags/but.md>), [call](<https://devfeed.tech/tags/call.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [confidence-score](<https://devfeed.tech/tags/confidence-score.md>), [core](<https://devfeed.tech/tags/core.md>), [devops](<https://devfeed.tech/tags/devops.md>), [right](<https://devfeed.tech/tags/right.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

VectorStep describes a runtime gate that decides whether each agent step may proceed using a trust vector: self-reported confidence, a second-agent verifier, evidence for claims in the run trace, deterministic checks, and the agent's historical calibration. Signals combine conservatively: a verifier can lower trust but not raise it, enforced grounding caps trust, and any failed deterministic check sets trust to zero. Calibration can adjust confidence based on human-labeled outcomes, with versioned buckets keyed to the prompt and agent configuration; it is advisory by default and requires enough samples before use. The gate compares combined trust with a configured threshold and can escalate, abort, or proceed, with recorded signals intended to make decisions auditable. The article also describes evidence-based promotion readiness and replaying labeled runs against candidate changes. The author notes limits: verifiers and evidence can be wrong or incomplete, deterministic checks do not establish overall correctness, and calibration depends on accurate human labels and past outcomes. VectorStep is described as beta.

### Source excerpt

An agent finishes a task and tells you: "I'm 95% confident." Should you believe it? VectorStep LiveTail pipeline output If you've run LLM agents against anything that matters, you probably know the answer. We've all done the eye-roll when a model sounds absolutely sure and turns out the next response is a u-turn. Reported confidence is a style of writing at least as much as it is a measurement. Yet in most agent tooling that single number is the thing deciding whether the agent's output gets acted on, and that's even if there's any sort of check. Let's be honest most agents in the wild are just prompted and then left to get on with it. I've spent twenty-plus years in observability, where data matters and proving a problem in production needs to be backed up with data. When I started building autonomous AI pipelines (originally for alert triage), I kept hitting the same problem. The tooling had plenty of ways to build agents and almost no way to decide, per decision, whether one had proved it was trustworthy with its track record. That gap became VectorStep, and the core of it is an idea I call the trust vector. In this article I'll explain how it works, walk through a real gating decision with numbers, and make the case for why I think the design is genuinely different. I'll also be upfront about what it doesn't do. The problem: averages and vibes There are generally two common ways teams decide whether to trust an agent today. 1. Take the agent's word for it. The model emits a confidence score, or you ask it to rate its own answer, and you threshold on that. The weakness is obvious, it's the agent grading its own work with no outside check. Self-reported confidence correlates with how an answer reads, not with whether it's right. 2. Evaluate it once, before launch. You build an eval set, measure accuracy, and ship if the number looks good. This is necessary, but it answers a different question. Evals tell you what an agent did on average, once, against a fixed set

## FinOps starts before the bill: visibility and ownership in small, chaotic teams

DevFeed: [FinOps starts before the bill: visibility and ownership in small, chaotic teams](<https://devfeed.tech/articles/finops-starts-before-the-bill-visibility-and-ownership-in-small-chaotic-teams-122300.md>)

Original publisher: [Read original article](<https://blog.devops.dev/finops-starts-before-the-bill-visibility-and-ownership-in-small-chaotic-teams-5fa0a3ed17b7?source=rss----33f8b2d9a328---4>)

Author: Soufiane Chahid

Published: 2026-11-07T20:46:01Z

Content type: article

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [finops](<https://devfeed.tech/topics/finops.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Finance](<https://devfeed.tech/topics/finance.md>), [Cloud Networking](<https://devfeed.tech/topics/cloud-networking.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [bill](<https://devfeed.tech/tags/bill.md>), [billing](<https://devfeed.tech/tags/billing.md>), [builds](<https://devfeed.tech/tags/builds.md>), [business](<https://devfeed.tech/tags/business.md>), [business-value](<https://devfeed.tech/tags/business-value.md>), [but](<https://devfeed.tech/tags/but.md>), [certificate](<https://devfeed.tech/tags/certificate.md>), [change](<https://devfeed.tech/tags/change.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [company](<https://devfeed.tech/tags/company.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [devops](<https://devfeed.tech/tags/devops.md>), [devsecops](<https://devfeed.tech/tags/devsecops.md>), [finops](<https://devfeed.tech/tags/finops.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [ownership](<https://devfeed.tech/tags/ownership.md>), [small](<https://devfeed.tech/tags/small.md>), [teams](<https://devfeed.tech/tags/teams.md>), [visibility](<https://devfeed.tech/tags/visibility.md>)

### AI overview

In small teams sharing infrastructure across projects, FinOps often stalls before optimization because the people managing resources cannot see the bill or identify who owns each resource. The article recommends starting with visibility and ownership: give infrastructure operators read-only billing access, map running resources to projects, and use showback to make project costs visible without charging teams. It also treats unmanaged changes as operational and financial risks, and says to make infrastructure changes visible through a shared channel. For sizing, measure usage, weigh extra capacity costs against the cost of failure, and account for how quickly a machine can be resized. Security monitoring can also reveal cost anomalies such as compromised workloads consuming compute. Its 30-day sequence is to see resources and costs, assign owner labels, alert on certificate expiry and sustained high CPU, then review project costs and sizing. The author argues this requires someone with authority over both infrastructure and budget.

### Source excerpt

A production site went down for four hours because a certificate expired. Nobody had touched the certificate. A firewall rule had been changed at the cloud provider's edge, outside the usual runbook process, and it quietly blocked the renewal. There was no business impact at the time. But there could have been, and nobody saw it coming, because nobody had the full picture. At first glance, this isn't a FinOps story; it isn't about money. But it has the same root as most cost problems in small teams: things change where no one is looking. I've spent the past year as the only infrastructure and DevSecOps engineer for several products in production, so I've seen this from the inside. Most FinOps writing assumes a big company: a finance team, an infrastructure bill split by tags and ownership for each team, even maybe a dedicated FinOps practitioner. Even the State of FinOps 2026 survey reflects this: 80% of respondents work at organizations with 1,000+ employees, and those managing $100M+ in annual cloud spend average 8 to 10 FinOps practitioners. A five-person startup where the CTO holds the cloud account usually sees its bill just fine. The trouble starts when one small team runs shared infrastructure for many projects, as in a venture studio, an incubator, an agency, or a research lab, and the money sits somewhere else. One person builds and runs the infrastructure. Someone else holds the billing accounts. Another one pays them. Resources appear without anyone telling the person in charge. In that world, FinOps does not fail at optimization; it fails before that, at seeing and informing. FinOps in one paragraph FinOps isn't about cutting costs. It's about getting the most business value from every dollar spent on technology. Sometimes the right answer is to spend more, because it's worth it. The FinOps Foundation describes it as a loop of three phases: Inform: see where the money goes and who uses it. Optimize: use less, and pay less for what you use. Operate: turn

[Next page](<https://devfeed.tech/sources/devops-dev.md?cursor=WyIyMDI2LTExLTA3VDIwOjQ2OjAxKzAwOjAwIiwgIjIwZDliMDA5LWE2ZDEtNGEyNS04Nzk5LTM0MDE4OTE2ODExNCJd>)