Bench runner fl-run-bench usage

fl-run-bench [options] file class.method

fl-run-bench launch a FunkLoad unit test as load test.

A FunkLoad unittest uses a configuration file named [class].conf, this configuration is overriden by the command line options.

See for more information.


fl-run-bench MyTestCase.testSomething
Bench MyTestCase.testSomething using MyTestCase.conf.
fl-run-bench –config ~/test.conf MyTestCase.testSomething
Bench MyTestCase.testSomething using test.conf located in home directory.
fl-run-bench -u http://localhost:8080 -c 10:20 -D 30
Bench MyTestCase.testSomething on localhost:8080 with 2 cycles of 10 and 20 users during 30s.
fl-run-bench -h
More options.


--version show program’s version number and exit
--help, -h show this help message and exit
 Path to alternative config location. Otherwise the configuration file is expected to be named after test case class, located either next to test module or path defined by environment variable FL_CONF_PATH
 Base URL to bench.
 Cycles to bench, this is a list of number of virtual concurrent users, to run a bench with 3 cycles with 5, 10 and 20 users use: -c 5:10:20
 Duration of a cycle in seconds.
 Minimum sleep time between requests.
 Maximum sleep time between requests.
 Sleep time between tests.
 Startup delay between thread.
--as-fast-as-possible, -f
 Remove sleep times between requests and between tests, shortcut for -m0 -M0 -t0
 Python dotted import path to BenchRunner class to use.
--no-color Monochrome output.
 Do not fail if css/image links are not reachable.
--simple-fetch Don’t load additional links like css or images when fetching an html page.
--label=LABEL, -l LABEL
 Add a label to this bench run for easier identification (it will be appended to the directory name for reports generated from it).
 Instantiates a debug HTTP server which exposes an interface using which parameters can be modified at run-time. Currently supported parameters: /cvu?inc=<integer> to increase the number of CVUs, /cvu?dec=<integer> to decrease the number of CVUs, /getcvu returns number of CVUs
 Port at which debug server should run during the test
--distribute distributes the CVUs over a group of worker machines that are defined in the section [workers]
 this parameter will over-ride the list of workers defined in the config file. expected notation is uname@host,uname:pwd@host or just host...
 this parameter is for internal use only. it signals to a worker node that it is in distributed mode and shouldn’t perform certain actions.