Implementation: How should Railway deployments be triggered from the GitHub ...

Documentation for Sessions

⚙️ Implementation: How should Railway deployments be triggered from the GitHub ...

Date: 2026-01-24
Time: 14:39:38 UTC
Session: editor2-64.5
Agent: documentation
Confidence: 95%
Status: ✅ Documented

📋 Decision Context

Question:

How should Railway deployments be triggered from the GitHub monorepo?

Task: Railway deployment strategy

🎯 Decision Made

Chosen:

Dedicated deploy branches per service (deploy/api, deploy/ai-api, deploy/docs-api)

Options Considered:

Railway CLI from CI
Auto-deploy on main branch
Dedicated deploy branches per service
GitHub Actions with Railway API

💭 Reasoning

Deploy branches solve monorepo challenges: 1) Each service deploys independently when its branch is updated, 2) Merge from main only when ready to deploy, 3) Each branch contains its own railway.toml pointing to the correct Dockerfile, 4) Easy rollback by reverting branch, 5) Clear separation of concerns. Railway watches specific branches, not file paths, making this the most reliable approach for monorepos.

📁 Files Involved

docs/english/Deployment/Internal/Railway/README.md
docs/english/Deployment/RAILWAY-GITHUB-DEPLOYMENT.md

🔗 Related

Decision ID: dev_1769265578662_j355aq0
Session: 64
Workspace: editor2

Auto-generated by MCP Smart Consolidation