Integration / Native / Connector Integration
NativeExternalIntegration Pattern

Integration Native Connectors (External)

David TirabassiUpdated
Enterprise/Solution Architect · Almost 30 years' experience, specialising in integration · TOGAF 9.1 certifiedLinkedInGitHub

Problem

An internal workload in a private subnet must call a system that an external party owns, synchronously over the public internet, using that system's own SDK or connector. Wired up ad hoc, those calls open insecure connections that bypass network isolation and egress controls.

Solution

Keep the native component, but route every external call through a Managed NAT or Egress Gateway in the Public Subnet (Perimeter). The gateway allows only listed FQDNs and IPs, presents a static egress IP the provider can allow-list, and logs every attempt.

Cloud Paradigm

  • Application-Native Integration
  • Secure Egress Connectivity
  • Workload-Centric Security
  • Zero Trust Architecture (for outbound)

Solution Flow

Outbound Flow (Internal Egress via Native Integration):

  1. Backend Workload (Native Integration): The internal microservice uses its native integration component (e.g., a specific library, SDK or built-in connector) to send a secure HTTPS request to an external system.
  2. Private Subnet (Workloads): The workload runs in a Private Subnet (Workloads) with no direct Public Internet access, so the request goes to the designated egress path.
  3. Managed NAT / Egress Gateway: Traffic passes through a Managed NAT or Egress Gateway in a Public Subnet (Perimeter), which enforces FQDN/IP allow-listing, performs network address translation (NAT) and logs every egress attempt.
  4. Edge Protection (Optional Egress WAF): If regulation requires it, an optional egress WAF applies deep packet inspection, DLP or advanced threat protection before traffic reaches the Public Internet.
  5. External Target: The external system receives the request from a static, allow-listed egress IP address.

When to Use

  • The workload already ships a mature, application-specific SDK, library or connector for the external system, making custom integration code redundant.
  • The caller must block on a synchronous result (e.g., real-time validation, pricing or lookup calls).
  • The external provider needs static, allow-listed egress IPs to admit your traffic at their perimeter.
  • Outbound traffic must be governed centrally with FQDN/IP allow-listing, NAT and full egress logging for audit and compliance.
  • The external endpoint is stable and fast enough that synchronous coupling won't cascade failures into your workload.

When NOT to Use

  • The exchange tolerates eventual consistency. An event-driven or message-queue pattern decouples the systems and absorbs external outages better.
  • No native connector exists and bespoke protocol handling would be brittle. An API gateway or integration platform (iPaaS) fits better.
  • The external system is slow or unreliable, so synchronous blocking risks thread exhaustion and cascading latency.
  • High-volume batch movement is needed, which bulk file transfer or ETL pipelines handle better than per-request calls.
  • Several internal workloads need the same external integration. Use a shared broker or facade instead of duplicating native clients.

Trade-offs

  • Fast delivery by reusing vendor-supplied components vs tight coupling to that library's release cadence, versioning and deprecation risk.
  • Simple, direct synchronous flow vs weaker fault isolation: external latency or downtime directly degrades the caller.
  • Centralised egress via Managed NAT/Gateway with static IPs and allow-listing vs a shared chokepoint that adds a hop and must be scaled and monitored.
  • Strong compliance through egress logging, WAF and DLP vs extra perimeter cost, inspection latency and operational overhead.
  • Secure config and secrets injection vs the lifecycle burden of external credentials and endpoint rotation.

Illustrative Scenario

A hypothetical scenario showing how the pattern applies. Not a client case study.

Suppose a pharmaceutical distributor's order-fulfilment microservice has to check a pharmacy's controlled-substance licence status against a national regulatory registry before each shipment is released. The service runs in a private workloads subnet with no public route to the internet. It uses the registry's official Python SDK to send a synchronous HTTPS request carrying the pharmacy's registration identifier. The request goes to a Managed NAT gateway in the perimeter subnet, which translates it to a static egress IP the registry has allow-listed, filters it by FQDN and logs it. Because dispensing data is regulated, the traffic also passes through an egress WAF for DLP inspection before it crosses the public internet. The SDK applies a circuit breaker and exponential backoff, so a registry outage holds dispatch cleanly instead of cascading through the service. OpenTelemetry traces record latency and error rates, and endpoint credentials arrive as injected secrets, never hardcoded in the code.

Additional Details

  • Protocol & Format: Use secure synchronous protocols (e.g., HTTPS, gRPC over TLS) and standard formats (e.g., JSON, XML, Protobuf).
  • Error Handling & Retry Mechanisms: Build error handling, circuit breakers and exponential backoff retries into the native component to cope with transient network issues or external outages.
  • Configuration Management: Manage endpoints, credentials and parameters securely as code and inject them through environment variables or a secrets management service. Never hardcode them.
  • Observability: Collect logs, metrics (latency, error rates, request volume) and distributed traces (e.g., OpenTelemetry) for the native component and the Egress Gateway, for troubleshooting and audit.
  • Compliance & Auditing: Outbound connections and data exchanges must meet the relevant compliance frameworks. Capture detailed audit logs and store them securely.

Security Controls

  • Transport Security: Enforce Transport Layer Security (TLS 1.2 or higher) on all outbound connections from the native integration component.
  • Authentication & Authorisation: For system-to-system integrations, authenticate with OAuth 2.0 (Client Credentials Grant), Mutual TLS (mTLS) or secure API Keys (held in a secrets management service), as the external target requires.
  • Outbound Controls: Host the workload in a Private Subnet (Workloads) with no direct Public Internet access. Route all outbound traffic through a Managed NAT or Egress Gateway in a Public Subnet (Perimeter) that allows connections only to explicitly allow-listed external IP addresses or Fully Qualified Domain Names (FQDNs). Inspect outbound traffic for Data Loss Prevention (DLP) if compliance requires it.
  • Workload Security & Isolation: Deploy the workload in a Private Subnet (Workloads), follow secure coding practices, and scan and patch it regularly. Access to it follows Zero Trust principles.
  • Secrets Management: Store all credentials, API keys and certificates for external authentication in a managed secrets management service.

Related Patterns