I was replicating youtube video demo to create Hello World workload. On my personal free GitHub account I had limitation to allocate machine type to codespace only upto 4 cores while requirement was 8 cores. I asked GitHub Copilot to fix the repo files for a 4 core machine type.
Error in Micrsoft Fabric Experience:
Development workload startup failure.
The development workload startup has failed due to an inactive local server or a configuration issue.
Solution:
Fixed: OOM Errors on 4-Core Codespaces
The Problem
The development server crashes with Out-of-Memory (OOM) errors when running the Fabric Extensibility Toolkit on standard 4-core Codespace machines.
Root Cause
Webpack bundling for this toolkit is memory-intensive. On 4-core machines (which typically only have 8GB RAM), the default Node.js heap limit and source map generation exceed the available system memory.
Solution Applied
1. Increased Node Heap Size
Updated the start script in package.json to allocate more memory to the Node process.
// Before
"start:codespace": "node --max-old-space-size=4096 ..."
// After
"start:codespace": "node --max-old-space-size=6144 ..."
2. Conditional Source Maps
Disabled source maps specifically when running inside a Codespace environment to reduce the memory footprint by approximately 2GB.
// Before
devtool: 'source-map'
// After
devtool: process.env.CODESPACES ? false : 'source-map'
3. Environment Cleanup
Performed a manual cleanup of leftover build assets and cache from previous failed setup attempts to ensure a clean slate.
Result
The Dev server and gateway now start successfully on 4-core machines without memory crashes.
[!TIP] Recommendation: While these fixes stabilize the 4-core experience, upgrading to an 8-core machine is still recommended for optimal build speed and performance.
