Testing
The following sections explain how to unit test LinuxForHealth connect and load test the LinuxForHealth ecosystem.
Required Software
Make sure you have the required software installed from Developer Setup. To run the load tests you’ll also need to install locust:
pip install locust
Run Unit Tests
LinuxForHealth connect uses pytest for unit testing. To run the unit tests, run pytest from the root directory of connect:
pipenv run pytest
Run Load Tests
Example 1
To run the load tests, make sure you have a LinuxForHealth environment running (see Developer Setup for instructions), then run the load tests from the connect/load-test directory:
cd load-test
locust --host https://localhost:5000 --run-time 5m --users 500 --spawn-rate 20 --headless
This set of steps runs the default load test defined by the connect/load-test/locustfile.py. You can adjust the length of the test (–run-time), the number of simulated concurrent users (–users) and rate at which those users are added to the test (–spawn-rate). Furthermore, you can adjust which endpoints are tested by adding/removing endpoints from locustfile.py. See the locust documentation for additional load test configuration options.
Example 2
The default load test POSTs to the /fhir endpoint using randomly selected FHIR resources from connect/load-test/messages/fhir[1…NUM_FILES].json files. You can replace the contents of these fhir*.json files with your own FHIR resources or add fhir resource files to connect/load-test/messages using the same naming convention. To separate FHIR resources, place them in separate files. To change the number of files used by the load test (default = 2):
NUM_FILES=1 locust --host https://localhost:5000 --run-time 5m --users 500 --spawn-rate 20 --headless
In the test output, you will see that only the first of the two sample fhir messages is loaded:
Loading message file = ./messages/fhir1.json
Example 3
To use the locust UI to run load tests, run locust from the command line:
locust --host https://localhost:5000
then point your browser to http://127.0.0.1:8089 to run the test, specifying the number of users and spawn rate via the UI.