col1985/raincatcher-demo-load-testing

:chart_with_upwards_trend: Scripts for load-testing the demo cloud app with load-runner

RainCatcher demo load testing scripts

This project contains scripts that are consumed by load-runner, for load-testing the demo cloud app.

Setup

There's a setup script at ./setup/usersWorkorders.js which creates a specified number of users, and creates a workorder for each one. Running it with node on the command line will show details of the parameters that it expects:

node setup/usersWorkorders.js
Options:
  --app, -a          Cloud app base URL to target            [string] [required]
  --username, -u     Username to use to login to the app     [string] [required]
  --password, -p     Password to use to login to the app     [string] [required]
  --numUsers, -n     The number of users/workorders needed   [number] [required]
  --concurrency, -c  The concurrency at which to create resources
                                                             [number] [required]

Missing required arguments: app, username, password, numUsers, concurrency

See the below section on how to run the test to see how this setup script can be executed by load-runner before a test run, with the same parameters as the test run itself.

Here's an example of how to run the setup script standalone to create 100 users at a concurrency of 10:

./setup/usersWorkorders.js -a https://mycloudapp.mydomain.tld -u trever -p 123 -n 100 -c 10

npm is also configured to be able to run this script as setup, just remember the -- to separate npm args from the script args:

npm run setup -- -a https://mycloudapp.mydomain.tld -u trever -p 123 -n 100 -c 10

Running the test

Here's an example of running the test with 4 users, at a concurrency of 2, with 2 portal users and 2 device users:

load-runner --output true \
    --rampUp 1 \
    --numUsers 4 \
    --pattern 0 1 \
    --concurrency 2 \
    --script scripts/sync-workflow.js \
    --before ./setup \
    -- -a https://wfmdemocloudapponlyexfj.local.feedhenry.io -u trever -p 123

Collecting Results

There's a collection script at util/collectResults.js which takes a results directory to pluck relevant information from, and a csv file (and an optional flag to write the CSV header). It can be run as follows:

node util/collectResults.js --dir runs/<result dir>/ --csv results.csv --header

# or as the npm 'results script'
npm run results -- --dir runs/<result dir>/ --csv test.csv --header