TECHNICAL NEWS & REPORTS

Intlight Tech Insights & Engineering Reports

HOMENews & Blog

A curated collection of deep-dive technical articles on Nuxt 4 architecture, Multi-Region Kubernetes, Zero-Trust WAF, Microservices, and Autonomous AI Agents.

KubernetesTECHNICAL NEWS & REPORTS

I got tired of SSHing into 10 VMs a day, so I built a live map of my whole infrastructure

"Every day at work looked the same. Something breaks, or I need to push a new image, and I'm SSHing..."
B
ByteStrixAuthor:
17/8/2026 6 phút
I got tired of SSHing into 10 VMs a day, so I built a live map of my whole infrastructure

Every day at work looked the same. Something breaks, or I need to push a new image, and I'm SSHing into three different VMs, running docker ps on one, kubectl get pods on another, piecing together in my head what's actually connected to what. The official Kubernetes dashboard never stuck for me either, and I never bothered installing Lens everywhere I needed it.

At some point I just wanted one thing: open a browser tab and see everything. Not a list. An actual map, where I can see a VM, the containers on it, a Kubernetes namespace next to it, and how they talk to each other.

So I built it. It's called InfraCanvas, and I want to show the two ways you can connect something to it, because I think the "how easy is this actually" question matters more than any feature list.

Way 1: you already have a kubeconfig? You're done.

If kubectl get pods works for you right now, InfraCanvas works for you right now. No agent, nothing installed on your cluster, nothing to configure. Open the dashboard, click "Add cluster," drop your kubeconfig file in, pick a context if it has more than one. That's it — within a few seconds your nodes, namespaces, deployments, pods and services show up on the canvas, color coded by health. It just talks to your cluster's API server the same way kubectl does. The kubeconfig gets encrypted before it's stored, never logged anywhere.

One honest caveat, because I'd rather tell you now than have you hit it yourself: if your kubeconfig came from aws eks update-kubeconfig or gcloud container clusters get-credentials, it authenticates through a CLI plugin, and that can't run from a backend that isn't your own machine. Self-hosting InfraCanvas on your own box? Not an issue, the plugin runs right there. Using the hosted version? There's a small relay pod you deploy instead, one kubectl apply, connects outbound from inside your cluster. The UI tells you upfront which kubeconfigs need this instead of just failing silently on you.

Way 2: plain VMs, one command. Not everything is Kubernetes — a lot of real infrastructure is a VM running Docker, or systemd services, or a Node process someone started with PM2 in 2022 and never touched again. One install command from the dashboard, a small Go binary as a systemd service, streams everything up over a single outbound WebSocket. No inbound ports.

What you get once it's connected: a real terminal into a pod or host, live logs, restart a container or service, scale a deployment, edit a manifest and apply it, all without leaving the browser. None of this is groundbreaking alone — what I haven't found anywhere else is all of it on one map instead of scattered across a terminal, a cloud console, and a dashboard that only understands one of the two.

It's open source (AGPL) — code's on GitHub: https://github.com/bytestrix/InfraCanvas. Hosted version if you'd rather not run anything: https://infracanvas.app

I'm still early, small project, built mostly solo — if you try it and something's rough, I'd genuinely like to hear about it.


🔗 Nguồn bài viết gốc: ByteStrix

Discussions & Comments14

Leave a Technical Comment

Share your architectural thoughts or ask technical questions...

H
Hannah SchmidtDevOps & CI/CD Lead
1 day ago

Can confirm: automated canary deployments with Argo Rollouts and Prometheus metrics analysis prevented several outages for our payment gateways.

E
Elena RostovaLead SRE & Platform Architect
45 mins ago

The KEDA autoscaling setup with custom Prometheus metrics is production-grade. We observed a 60% compute cost reduction after switching to event-driven pod scaling.

L
Liam O'ConnorFrontend Performance Specialist
3 hours ago

Nuxt 4 with selective hydration and zero-JS interactive islands delivers mind-blowing speed. Sub-100ms INP and 99+ Core Web Vitals out of the box.

L
Lucas MoreauCloud Native Developer
12 hours ago

Kafka event streaming with schema registry ensures backward compatibility even as payload models evolve across microservice boundaries.

A
Alexander WrightPrincipal Systems Architect @ Stripe
20 mins ago

Superb architectural breakdown! The hybrid L1 in-memory + L2 distributed Redis cache pattern is crucial for mitigating high-concurrency thundering herd issues.

D
David ChenStaff Infrastructure Engineer
12 mins ago

Totally agree, Alexander. Pairing that with singleflight request deduplication on the backend virtually eliminates DB spikes.

SPOTLIGHT & LATEST NEWS

Hot Trending Topics

View All →
I Thought I'd Lost the Plot. I Was Writing It. 🔥 HOT SPOTLIGHT
AI Agents6 min read

I Thought I'd Lost the Plot. I Was Writing It.

I Thought I'd Lost the Plot. I Was Writing It. I set out to build autonomous...

Explore
What Is the Circuit Breaker Pattern? A Practical Guide 🔥 HOT SPOTLIGHT
Microservices6 min read

What Is the Circuit Breaker Pattern? A Practical Guide

What Is the Circuit Breaker Pattern? A Practical Guide for Developers Imagine your...

Explore
I attacked my own npm package before launching it. It let the proposer approve their own writes 🔥 HOT SPOTLIGHT
Security6 min read

I attacked my own npm package before launching it. It let the proposer approve their own writes

My library exists so a human approves an LLM's UPDATE before it runs. It never checked that the approver was somebody other than the proposer — and wrote \"approved\" into the audit trail anyway.

Explore
Build an MCP Server in Go (Part 1): Designing a diagnostic-grade Kubernetes client 🔥 HOT SPOTLIGHT
Kubernetes6 min read

Build an MCP Server in Go (Part 1): Designing a diagnostic-grade Kubernetes client

This post designs the Kubernetes client. The next post wraps it as an MCP server and wires it to an...

Explore
The Write Policy Is the Hard Part: Promotion Pipelines for Agent Memory 🔥 HOT SPOTLIGHT
AI Agents6 min read

The Write Policy Is the Hard Part: Promotion Pipelines for Agent Memory

Storing agent memory is easy. Deciding what earns a permanent write, and keeping the write-path alive through RBAC and network policy, is the real work.

Explore
I Changed How I Think About AI Memory 🔥 HOT SPOTLIGHT
AI Agents6 phút

I Changed How I Think About AI Memory

I Changed How I Think About AI Memory When I first built Lean AI Memory, I focused too...

Explore
Real-Life Refactoring Example: ~3x Less Code to Read 🔥 HOT SPOTLIGHT
Microservices6 phút

Real-Life Refactoring Example: ~3x Less Code to Read

There is a popular idea that refactoring is making code shorter. It is not entirely wrong....

Explore
The Tragedy of the Clean-Handed Auditor 🔥 HOT SPOTLIGHT
Security6 phút

The Tragedy of the Clean-Handed Auditor

\"I could save them if they'd only listen...\" Hey, you. Yeah, you: the compliance or governance...

Explore
Related Articles5 articles
View all Kubernetes →
Build an MCP Server in Go (Part 1): Designing a diagnostic-grade Kubernetes client6 min read
Kubernetes8/17/2026

Build an MCP Server in Go (Part 1): Designing a diagnostic-grade Kubernetes client

This post designs the Kubernetes client. The next post wraps it as an MCP server and wires it to an...

Author: Fer RiosRead Article
Building Sluice: QoS-Aware Capacity Governance for Self-Hosted LLM Inference6 phút
Kubernetes8/17/2026

Building Sluice: QoS-Aware Capacity Governance for Self-Hosted LLM Inference

📦 Project: https://github.com/VampiricCyborg/sluice 1. The Problem: When Capacity Becomes...

Author: Madhav M SRead Article
One GPU, four ways to share it: ten scenarios, and the headline finding I had to retract6 phút
Kubernetes8/17/2026

One GPU, four ways to share it: ten scenarios, and the headline finding I had to retract

I measured every GPU sharing mode across ten scenarios, published a headline finding that duplicate models are nearly free on unified memory, then failed to replicate it and retracted it. Here is what the controlled replication showed and how the original measurement fooled me.

Author: Christopher MaherRead Article
The Backup Awakens: A Star Wars Story6 phút
Kubernetes17/8/2026

The Backup Awakens: A Star Wars Story

The Quest Begins (The \"Why\") Honestly, I used to think backups were the boring chores you...

Author: TimevoltRead Article
I Automated My Entire GitOps Security Stack. The First Thing It Blocked Was My Own Salary.6 phút
Kubernetes17/8/2026

I Automated My Entire GitOps Security Stack. The First Thing It Blocked Was My Own Salary.

I Automated My Entire GitOps Security Stack. The First Thing It Blocked Was My Own...

Author: Le BeltagyRead Article