Dapr Test Infra

This repo includes test apps and infrastructure tools.

Test apps for long haul tests:

Feed Generator : contains feed-generator logic. Exercises the pubsub component (publishing).
Message Analyzer : contains message-analyzer logic. Exercises pubsub component (subscribing) and output binding.
HashTag Counter : contains hashtag-counter. Exercises input binding and actor component.
HashTag Actor : hashtag-actor logic. Exercises actor component.
Pubsub Workflow : azure service bus pubsub logic. Exercises pubsub component (publishing and subscribing).

Test analytics :

Test Crawler : A Python script scrapes the Dapr E2E tests results.

Solution overview and app dependency

sequenceDiagram
FeedGenerator --) MessageAnalyzer : publish a SocialMediaMessage <br> async using pubsub <br> topic "receivemediapost"
MessageAnalyzer --) HashTagCounter: invoke output binding "messagebinding" <br> to publish a message to a queue <br> shared with HashTagCounter
HashTagCounter ->> HashTagActor: receives messages from shared queue <br> using input binding "messagebinding" <br> increase counter by invoking a <br> per (hastag,sentiment) actor
Snapshot ->> HashTagActor: get counts from HashTagActor
PubSubWorkflow --) PubSubWorkflow: publish messages at different rates <br> to different topics it also subscribes to.

TODO what about Pubsub Workflow ? who is publishing to it? And validation workflow?

Deploying this application

Locally with Dapr Multi-App run

This is not properly a deployment, but it is a way to run all the longhaul applications locally using Dapr Multi-App run.

# Start base components such as redis and zipkin, which we depend on
dapr init
# Start RabbitMQ which is used for our input/output binding tests.
docker run -d -p 5672:5672 --hostname dapr_rabbitmq --name dapr_rabbitmq rabbitmq:3-alpine
# Run all the apps
dapr run -f deploy/dapr-multi-app/dapr.yaml

On Azure Kubernetes Service (AKS)

Define the resource group we in which will we create the cluster and other resources and the name of the cluster we will create.

export resourceGroup='myNewResourceGroup'
export location='eastus'
export clusterName='test-infra'

Create a resource group four your new cluster

az group create --name ${resourceGroup} --location ${location}

Deploy a cluster with test apps to this resource group:

az deployment group create --resource-group ${resourceGroup} --template-file ./deploy/aks/main.bicep --parameters "clusterName=${clusterName}"

Done! Explore your new AKS cluster with the sample applications

az aks get-credentials --admin --name ${clusterName} --resource-group ${resourceGroup}

Regaining control of your AKS cluster's Dapr setup.

Our AKS Bicep templates offer a convenient way to quickly start up an AKS cluster running our longhaul applications for E2E testing. In this setup, Dapr is installed on the AKS cluster by means of an AKS extension. Using this extension is great if you want a fully managed Dapr setup on your AKS cluster. However, its use puts some constraints on the user's ability to control the installed Dapr setup. Using Helm or dapr init -k directly to manage, upgrade or uninstall Dapr on your AKS cluster won't work.

If you want to fully control Dapr setup on your cluster, you need to uninstall the extension first:

az k8s-extension delete --yes --resource-group ${resourceGroup} --cluster-name ${clusterName} --cluster-type managedClusters --name ${clusterName}-dapr-ext

Then you can re-install Dapr using Helm or dapr init -k:

export DAPR_VERSION_TO_INSTALL='1.11.3'
helm upgrade --install dapr dapr/dapr \
    --version=${DAPR_VERSION_TO_INSTALL} \
    --namespace dapr-system \
    --create-namespace \
    --wait

Don't forget to do a rolling restart in all apps to pick up the new manually-installed Dapr version: 😉

for app in "feed-generator-app" "hashtag-actor-app" "hashtag-counter-app" "message-analyzer-app" "pubsub-workflow-app" "snapshot-app" "validation-worker-app" "workflow-gen-app"; do
    kubectl rollout restart deploy/${app} -n longhaul-test || break
done

Accessing the Grafana dashboards

There's currently a managed Grafana in the resource group in Azure, but it is not set up. The actual Grafana dashboards are running inside of the AKS cluster. You can access them by port-forwarding the Grafana service to your local machine:

For longhaul release:

 az aks get-credentials --name aks-longhaul-release --resource-group aks-longhaul-release
 kubectl port-forward svc/grafana 8080:80 --namespace dapr-monitoring

For longhaul weekly:

az aks get-credentials --name aks-longhaul-weekly --resource-group aks-longhaul-weekly
kubectl port-forward svc/grafana 8081:80 --namespace dapr-monitoring

The username for the Grafana dashboard is "admin" and you can get the password by running:

kubectl get secret --namespace dapr-monitoring grafana -o jsonpath={.data.admin-password} | base64 --decode

dapr / test-infra

readme