increase wait time for DNS cert

[paigewilliams]

increasing wait time for prod deploy

[Copilot]

Pull request overview

This PR increases the DNS propagation wait window in the EC2 user-data bootstrap script so production deploys have more time for the domain’s A record to point at the instance before requesting a Let’s Encrypt certificate.

Changes:

• Increase DNS resolution polling loop iterations from 20 to 2880 (with 30s sleeps, up to 24 hours total).

Comments suppressed due to low confidence (1)

infra/user_data.tftpl:109

• The DNS wait loop now allows up to 24h (2880 * 30s) but still proceeds to certbot even if DNS never resolves to the instance IP. That can cause long/hung deployments and confusing certificate failures. Consider making the timeout explicit, using a resolver available by default (avoid dig dependency), and aborting with a non-zero exit code if the expected IP is not observed within the timeout.

for i in $(seq 1 2880); do
  RESOLVED_IP=$(dig +short ${domain_name} | tail -1)
  if [ "$RESOLVED_IP" = "$EXPECTED_IP" ]; then
    echo "DNS resolved correctly to $EXPECTED_IP"
    break
  fi
  echo "Waiting for DNS... (got $RESOLVED_IP, want $EXPECTED_IP)"
  sleep 30
done

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.