This setup requires an existing installation of Minishift and follows the installation instructions for Hono on EnMasse using S2I: https://github.com/ctron/hono/tree/feature/support_s2i_05x/openshift
This tutorial will assume that you have a Unix-ish operating system and at least the following command line tools installed:
- minishift, oc
- curl, wget, jq, bash, tar
First you will need to start up minishift:
minishift start --cpus 4 --memory 16GB --disk-size 40GBThis readme file will assume that you have cloned (checked out) out the repository and are executing commands from inside the working space of your clone. Cloning is as easy as:
git clone http://github.com/ctron/hono-demo-1Or the following command if you want to check out the develop branch instead:
git clone http://github.com/ctron/hono-demo-1 -b developCreate a new EnMasse project:
oc new-project enmasse --display-name='EnMasse Instance'Download, unpack and deploy EnMasse:
wget https://github.com/EnMasseProject/enmasse/releases/download/0.17.0/enmasse-0.17.0.tgz
tar xzf enmasse-0.17.0.tgz
cd enmasse-0.17.0
./deploy-openshift.sh -n enmasse -m "$(minishift console --url)"
cd ..Note: After the deployment of EnMasse the system might need a while to download and install container images. Please wait for all pods to start before continuing. You can track the progress using the OpenShift Web UI.
The following step will create the addresses for DEFAULT_TENANT:
curl -X PUT --insecure -T src/openshift/addresses.json -H "content-type: application/json" https://$(oc -n enmasse get route restapi -o jsonpath='{.spec.host}')/apis/enmasse.io/v1/addresses/defaultoc new-project hono --display-name='Eclipse Hono™'
oc -n hono create configmap influxdb-config --from-file=src/openshift/influxdb.conf
oc process -f https://raw.githubusercontent.com/ctron/hono/feature/support_s2i_05x/openshift/hono.yml \
-p "ENMASSE_NAMESPACE=enmasse" \
-p "GIT_REPOSITORY=https://github.com/ctron/hono" \
-p "GIT_BRANCH=feature/support_s2i_05x"| oc create -f -Note: After executing the template container images will be downloaded and builds will be triggered in order to start up Hono. Please wait for all pods to start before continuing. You can track the progress using the OpenShift Web UI.
By default Hono limits the number of devices to 100 for its examples device registry.
This limit can be set by executing:
oc env -n hono dc/hono-service-device-registry HONO_REGISTRY_SVC_MAX_DEVICES_PER_TENANT=10000Note: Please remember that this device registry is held in-memory and flushed to disk using JSON. So performance might become an issue with too many devices.
Create a new project for the simulator and deploy it:
oc new-project iot-simulator --display-name='IoT workload simulator'
oc process -f src/openshift/demo.yml | oc create -f -Grafana can be deployed in order to watch the metrics of Hono and also see the simulated payload. It can be installed by executing the following commands:
oc new-project grafana --display-name='Grafana Dashboard'
oc process -f https://raw.githubusercontent.com/ctron/hono/feature/support_s2i_05x/openshift/grafana.yml \
-p ADMIN_PASSWORD=admin | oc create -f -The following command line requests require the use of the Grafana URL and will use the
environment variable GRAFANA_URL for that. You can set the URL in your local shell using:
GRAFANA_URL="$(oc -n grafana get route grafana --template='{{ .spec.host }}')"
echo "http://$GRAFANA_URL"It is also possible to open the URL with a web browser in order to view dashboards and configurations.
The credentials for this instance are admin / admin.
Create two new datasources by executing the following commands:
curl -X POST -T src/grafana/ds_hono.json -H "content-type: application/json" "http://admin:admin@$GRAFANA_URL/api/datasources"
curl -X POST -T src/grafana/ds_payload.json -H "content-type: application/json" "http://admin:admin@$GRAFANA_URL/api/datasources"
curl -X POST -T src/grafana/dashboard_hono.json -H "content-type: application/json" "http://admin:admin@$GRAFANA_URL/api/dashboards/db"
curl -X POST -T src/grafana/dashboard_payload.json -H "content-type: application/json" "http://admin:admin@$GRAFANA_URL/api/dashboards/db"Now you have an IoT simulator running, which will stream a data set to the Eclipse Hono instance. The IoT consumer will consume the simulated payload and store it in the metrics database of Hono.
You can check the following URLs:
- $GRAFANA_URL/dashboard/db/ampds2
- This shows the payload as it is stored in the InfluxDB payload instance.
- $GRAFANA_URL/dashboard/db/hono
- Shows the metrics of the Eclipse Hono instances. Initially this should show a "stunning" 1 message/second throughput.
You can install Eclipse Che to the setup by executing the following commands:
DEPLOY_ROOT_URL=https://raw.githubusercontent.com/eclipse/che/6.1.x/dockerfiles/init/modules/openshift/files/scripts/
curl -fL "$DEPLOY_ROOT_URL/deploy_che.sh" -o get-che.sh
curl -fL "$DEPLOY_ROOT_URL/che-openshift.yml" -o che-openshift.yml
curl -fL "$DEPLOY_ROOT_URL/che-config" -o che-config
export CHE_IMAGE_TAG=6.1.1
bash ./get-che.shAfter Che is initialized you can import the example project and start working on it:
CHE_URL="http://$(oc -n eclipse-che get route che --template='{{ .spec.host }}')"
echo "Open browser at: $CHE_URL/f?url=https://github.com/ctron/hono-demo-1"You can now open this project in Che and start making changes using the IDE.
It is possible to run the project directly from within the Che IDE by running the "buildAndRun" command. The URL to the application is showing on top of the console output.
It is also possible to create a standard OpenShift project from the sources in a few steps with the following commands:
oc new-project demo-gauge
oc new-app https://github.com/ctron/hono-demo-1 --context-dir=demo-gauge
oc expose svc/hono-demo-1
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent