locust-loadgen-api

Locust Helm Chart

This is a templated deployment of Locust for Distributed Load testing using Kubernetes.

Pre Requisites:

• Requires (and tested with) helm v2.1.2 or above.

Chart details

This chart will do the following:

• Convert all files in tasks/ folder into a configmap
• If an existing configmap is specified, it will be used instead of building one from the chart
• Create a Locust master and Locust worker deployment with the Target host and Tasks file specified.

Installing the chart

To install the chart with the release name locust-nymph in the default namespace:

helm install -n locust-nymph --set master.config.target-host=http://site.example.com stable/locust

Parameter	Description	Default
Name	Locust master name	locust
image.repository	Locust container image name	greenbirdit/locust
image.tag	Locust Container image tag	0.9.0
image.pullSecrets	Locust Container image registry secret	None
service.type	k8s service type exposing master	NodePort
service.nodePort	Port on cluster to expose master	0
service.annotations	KV containing custom annotations	{}
service.extraLabels	KV containing extra labels	{}
extraVolumes	List of extra Volumes	[]
extraVolumeMounts	List of extra Volume Mounts	[]
extraEnvs	List of extra Environment Variables	[]
master.config.target-host	locust target host	http://site.example.com
master.nodeSelector	k8s nodeselector	{}
master.tolerations	k8s tolerations	[]
worker.config.locust-script	locust script to run	/locust-tasks/tasks.py
worker.config.configmapName	configmap to mount locust scripts from	empty, configmap is created from tasks folder in Chart
worker.replicaCount	Number of workers to run	2
worker.nodeSelector	k8s nodeselector	{}
worker.tolerations	k8s tolerations	[]
worker.affinities	k8s affinities	{}

Specify parameters using --set key=value[,key=value] argument to helm install

Alternatively a YAML file that specifies the values for the parameters can be provided like this:

$ helm install --name my-release -f values.yaml stable/locust

Creating configmap with your Locust task files

You're probably developing your own Locust scripts that you want to run in this distributed setup. To get those scripts into this deployment you can fork the chart and put them into the tasks folder. From there they will be converted to a configmap and mounted for use in Locust.

Another solution, if you don't want to fork the Chart, is to put your Locust scripts in a configmap and provide the name as a config parameter in values.yaml. You can read more on the use of configmaps as volumes in pods here.

If you have your Locust task files in a folder named "scripts" you would use something like the following command:

kubectl create configmap locust-worker-configs --from-file path/to/scripts

Interacting with Locust

Get the Locust URL following the Post Installation notes. Using port forwarding you should be able to connect to the web ui on Locust master node.

You can start the swarm from the command line using port forwarding as follows:

for example:

export LOCUST_URL=http://127.0.0.1:8089

Start / Monitor & Stop the Locust swarm via the web panel or with following commands:

Start:

curl -XPOST $LOCUST_URL/swarm -d"locust_count=100&hatch_rate=10"

Monitor:

watch -n 1 "curl -s $LOCUST_URL/stats/requests | jq -r '[.user_count, .total_rps, .state] | @tsv'"

Stop:

curl $LOCUST_URL/stop