Imported from jeremylongshore/tons-of-skills-marketplace (skills/.curated/replit-migration-deep-dive/SKILL.md). Install upstream with npx skills add jeremylongshore/tons-of-skills-marketplace --skill replit-migration-deep-dive. Copyright stays with the author (MIT).

Replit Migration Deep Dive

Current State

!cat .replit 2>/dev/null | head -10 || echo 'No .replit found' !cat Procfile 2>/dev/null || echo 'No Procfile (not Heroku)' !cat Dockerfile 2>/dev/null | head -10 || echo 'No Dockerfile' !cat railway.json 2>/dev/null || echo 'No railway.json'

Overview

Comprehensive guide for migrating existing applications to Replit from Heroku, Railway, Vercel, Render, or local development. Covers converting configuration files, migrating databases, adapting to Replit's Nix-based environment, and setting up Replit-native features.

Prerequisites

• Source application with working deployment
• Access to current database for export
• Git repository with application code
• Replit Core or Teams plan

Migration Paths

Instructions

Step 1: Import from GitHub

1. Go to replit.com > Create Repl > Import from GitHub
2. Paste your repository URL
3. Replit auto-detects language and creates default config
4. Review and adjust .replit and replit.nix

Step 2: Convert from Heroku

Procfile to .replit:

# Heroku Procfile
web: npm start
worker: node worker.js
release: node migrate.js

# Equivalent .replit
run = "npm start"
entrypoint = "index.js"

[deployment]
run = ["sh", "-c", "node migrate.js && npm start"]
build = ["sh", "-c", "npm ci --production"]
deploymentTarget = "autoscale"

[env]
NODE_ENV = "production"

Heroku addons to Replit services:

Environment variables:

# Export from Heroku
heroku config -s > heroku-env.txt

# Import to Replit: copy each line into Secrets tab
# Or use Replit Secrets tool (lock icon in sidebar)

Step 3: Convert from Railway

railway.json to .replit:

// railway.json
{
  "build": { "builder": "NIXPACKS" },
  "deploy": {
    "startCommand": "npm start",
    "healthcheckPath": "/health"
  }
}

# Equivalent .replit
run = "npm start"

[deployment]
run = ["sh", "-c", "npm start"]
build = ["sh", "-c", "npm ci"]
deploymentTarget = "autoscale"

Step 4: Convert from Docker

Dockerfile to replit.nix:

# Dockerfile
FROM node:20-slim
RUN apt-get update && apt-get install -y python3 postgresql-client
WORKDIR /app
COPY package*.json ./
RUN npm ci --production
COPY . .
CMD ["node", "dist/index.js"]

# Equivalent replit.nix
{ pkgs }: {
  deps = [
    pkgs.nodejs-20_x
    pkgs.python311
    pkgs.postgresql
  ];
}

# .replit
run = "node dist/index.js"

[deployment]
run = ["sh", "-c", "node dist/index.js"]
build = ["sh", "-c", "npm ci --production && npm run build"]
deploymentTarget = "autoscale"

Step 5: Database Migration

Export from source:

# From Heroku Postgres
heroku pg:backups:capture
heroku pg:backups:download
pg_restore -d LOCAL_DB latest.dump

# From Railway
railway run pg_dump > backup.sql

# From any PostgreSQL
pg_dump --format=custom DATABASE_URL > backup.dump

Import to Replit PostgreSQL:

# In Replit Shell, after provisioning PostgreSQL in Database pane:

# Option 1: SQL file
psql "$DATABASE_URL" < backup.sql

# Option 2: Custom format dump
pg_restore -d "$DATABASE_URL" backup.dump

# Option 3: Schema only (recreate, then migrate data app-level)
psql "$DATABASE_URL" < schema.sql
node scripts/migrate-data.js

Migrate from non-PostgreSQL (MongoDB, etc.):

// scripts/migrate-from-mongo.ts
import { MongoClient } from 'mongodb';
import { Pool } from 'pg';

const mongo = new MongoClient(process.env.MONGO_URL!);
const pg = new Pool({ connectionString: process.env.DATABASE_URL });

async function migrate() {
  await mongo.connect();
  const users = await mongo.db('app').collection('users').find().toArray();

  await pg.query(`
    CREATE TABLE IF NOT EXISTS users (
      id TEXT PRIMARY KEY,
      email TEXT UNIQUE,
      name TEXT,
      data JSONB,
      created_at TIMESTAMPTZ
    )
  `);

  for (const user of users) {
    await pg.query(
      'INSERT INTO users (id, email, name, data, created_at) VALUES ($1, $2, $3, $4, $5)',
      [user._id.toString(), user.email, user.name, JSON.stringify(user), user.createdAt]
    );
  }

  console.log(`Migrated ${users.length} users`);
  await mongo.close();
  await pg.end();
}

migrate();

Step 6: Post-Migration Checklist

## After Migration

### Configuration
- [ ] .replit configured with correct run and build commands
- [ ] replit.nix includes all system dependencies
- [ ] All env vars moved to Replit Secrets
- [ ] PORT reads from environment variable
- [ ] App listens on 0.0.0.0 (not localhost)

### Database
- [ ] PostgreSQL provisioned in Database pane
- [ ] Data imported and verified
- [ ] Connection string uses DATABASE_URL env var
- [ ] SSL configured: { rejectUnauthorized: false }

### Testing
- [ ] App runs successfully in Workspace ("Run")
- [ ] Health endpoint works: /health returns 200
- [ ] All API endpoints functional
- [ ] Auth flow works (if using Replit Auth)
- [ ] File uploads work (if using Object Storage)

### Deployment
- [ ] Deployed successfully (Autoscale or Reserved VM)
- [ ] Custom domain configured (if applicable)
- [ ] SSL certificate provisioned
- [ ] Post-deploy health check passes

### Cleanup
- [ ] Old platform deprovisioned after verification period
- [ ] DNS records updated (if custom domain)
- [ ] CI/CD updated to point to Replit
- [ ] Team notified of new deployment URL

Error Handling

Resources

• Import from GitHub
• Import from Bolt
• Import from Lovable
• Nix on Replit
• Replit Deployments

Next Steps

For advanced troubleshooting after migration, see replit-advanced-troubleshooting.