Terraform Advanced — Bulk Import Strategies

Strategies for importing hundreds of existing resources into Terraform.

---

Strategy 1: Script-Generated Import Blocks

#!/bin/bash
# generate-imports.sh — Generate import blocks from AWS CLI

# List all S3 buckets
aws s3api list-buckets --query 'Buckets[].Name' --output text | tr '\t' '\n' | while read bucket; do
  cat <<EOF
import {
  to = aws_s3_bucket.this["${bucket}"]
  id = "${bucket}"
}
EOF
done > imports.tf

---

Strategy 2: Terraformer (Third-Party)

# Install
$ brew install terraformer

# Import all S3 buckets
$ terraformer import aws --resources=s3 --regions=us-east-1

# Import multiple resource types
$ terraformer import aws \
    --resources=vpc,subnet,security_group \
    --regions=us-east-1

---

Strategy 3: Incremental Import

Week 1: Import networking (VPC, subnets, route tables)
Week 2: Import compute (EC2, ASGs, ALBs)
Week 3: Import data (RDS, ElastiCache, S3)
Week 4: Import IAM (roles, policies)
Week 5: Import DNS (Route53)

💡 Production Tip: Import in dependency order — networking first, then compute, then application-level resources. This matches the natural terraform apply order.

---

What’s Next?

Gracefully remove resources from state.

← generate-config-out The removed Block →