Service virtualization replaces a hand-written in-process mock when you need an integration test to send real protocol requests—such as HTTP—to a simulated dependency. Keep in-process fakes and mocks for narrow unit tests; use a virtual service to exercise the application’s client, serialization, and request/response behavior without calling the real upstream.
What is the difference between a mock and a virtual service?
A hand-written mock or fake typically substitutes for a dependency through an in-process interface. The test calls application code while the substitute responds at that seam; it need not exercise the network client or protocol.
As an Amazon Associate I earn from qualifying purchases.
Service virtualization runs a simulated upstream service that accepts the application’s actual protocol requests and returns controlled responses. For an HTTP integration test, the application can use its real HTTP client and serialization path while the simulated service stands in for the remote API. WireMock supports standalone and containerized simulations, and Docker’s guide demonstrates this HTTP-level approach in contrast with mocking Java methods.
Neither technique is inherently the better choice: the test boundary should match the behavior you want to verify. A virtual service is still a simulation, not proof that the real provider works or that its current behavior matches the simulation.
When should you use each approach?
| Approach | Best fit | Question to ask |
|---|---|---|
| Hand-written in-process mock or fake | Narrow unit-level isolation at an application interface | Is the test about business logic, without checking network or client behavior? |
| Service simulator, such as WireMock or Mountebank | Controlled protocol-level interaction with a dependency | Does the test need request matching, response bodies, stateful sequences, or injected failures? |
| Real dependency service or container | Integration with the actual dependency implementation | Is the implementation practical to run, and is validating that implementation the goal? |
The real dependency option is not universally superior: it answers a different question. Choose a simulator when the protocol interaction matters but the real upstream is unavailable, costly, unreliable, or unsuitable for repeatable testing. Choose a real service when the test must validate that implementation. Contract validation or a real integration is also needed when the question is whether the provider’s current contract or production behavior is correct.
What can a service simulator test?
Requests and responses across the protocol boundary
WireMock can match requests and return configured responses. That lets a test exercise the application’s HTTP client path and request/response serialization rather than replacing a Java method call. The simulator controls the upstream side; the application still makes the protocol request.
Sequences and changing state
WireMock scenarios can return different responses as scenario state changes. This is useful when a test needs a sequence—for example, an initial response followed by a different result—rather than the same canned reply on every request.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Failures and delays
WireMock documents fault and latency simulation, including an HTTP 503 response with a configured delay. Controlled failures make it possible to exercise application behavior under conditions that can be difficult or risky to reproduce against a real service.
Request verification
Mountebank can record incoming requests for later checks, but recording must be enabled. Verification is not automatic: client-side test code must inspect the recorded request array and implement the assertions. Mountebank notes that long-running instances with many recorded requests may need disk persistence to avoid a significant memory-performance constraint.
How do you run a virtual service in an integration test?
One option is a standalone simulator; another is to run it in a container as part of the test environment. WireMock documents standalone and Docker distributions. Its Testcontainers guidance lists reviewed modules for JVM, Python, and Go, and recommends a generic container where no language-specific module exists. Module availability can change, so check the current documentation for the language you use.
Rank #4
- Identify the boundary. Decide which protocol request and response the test must exercise, and configure the application to send that traffic to the simulator rather than the real upstream.
- Start the simulator. Run it as a standalone process or provision it as a test container. Containerization can make the simulator part of the integration-test environment.
- Define the behavior. Configure matching requests and responses; add scenario state, delays, or faults only when the test needs those behaviors.
- Run the application test and assert outcomes. Check the behavior under test, and, when necessary, verify the request received by the simulator. For Mountebank, enable recording and write the verification logic in the test client.
How should you compare tools and approaches?
Compare options on the dimensions that affect the test, rather than assuming a simulator is automatically faster or cheaper:
Recommended Free Tools
- Boundary: Does the test need to cross an HTTP or other protocol boundary, or isolate code at an in-process seam?
- Protocol and language support: Does the tool support the protocol, and is there a suitable integration for your test language?
- Scenario needs: Are simple fixed responses enough, or must responses change with scenario state?
- Failure behavior: Do you need to inject errors, latency, or other controlled faults?
- Setup and deployment: Will a standalone process, container, or hosted simulator fit the test environment?
- Maintenance: How much configuration and verification code will the team need to keep accurate?
The official documentation describes capabilities and deployment choices, but does not establish a neutral benchmark for speed or total cost. Base those decisions on your own environment rather than treating a vendor capability list as comparative performance evidence.
Best Value
What do WireMock and Mountebank provide?
WireMock
WireMock’s documentation describes standalone JAR and Docker distributions, JSON and Java stub authoring, request matching, response templating, scenarios, and fault or latency injection. The examples in the documentation show version 3.13.2; verify the current version and features in the official WireMock documentation before adopting version-specific instructions. WireMock also describes a hosted service virtualization offering for teams that want managed hosting.
WireMock with Testcontainers
The WireMock Testcontainers guide covers provisioning WireMock in a container from test code. It lists reviewed JVM, Python, and Go modules and describes using a generic container when a language-specific module is not available.
Mountebank
Mountebank’s documentation describes an over-the-wire test double with stubbing and request recording. Account for the explicit recording setup and client-written verification when deciding whether its approach fits your tests.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWhat a simulated dependency does not prove
A passing test shows how the application behaved against the responses and request rules you configured. It does not establish that the real upstream currently accepts those requests, returns those responses, or behaves the same way in production. Keep simulation for repeatable controlled scenarios; use contract validation or a real integration when compatibility with the actual provider is the thing being tested.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
