Documentation for version: latest

Run workloads

Now that the devices are running, we need to run some container workloads on them. The first thing to do is to make sure that devices are online:

---> kubectl get edgedevices
NAME        AGE
camera-ny   33m
--->

To run a workload, we need to create a EdgeWorkload config, that looks like this:

apiVersion: management.project-flotta.io/v1alpha1
kind: EdgeWorkload
metadata:
  name: workload
spec:
  deviceSelector:
    matchLabels:
      app: camera
  type: pod
  pod:
    spec:
      containers:
        - name: workload
          image: quay.io/project-flotta/nginx:1.21.6

This workload will look for devices that has the label app=camera, so we need to tag our device (camera-ny) with the application tag:

kubectl label edgedevice camera-ny app=camera

The result of the edgedevice will be like this:

---> kubectl get edgedevice camera-ny -o json | jq .metadata.labels
{
  "app": "camera",
  "device.cpu-architecture": "x86_64",
  "device.cpu-model": "intelrcoretmi7-8650ucpu1.90ghz",
  "device.hostname": "localhost.localdomain",
  "device.system-manufacturer": "lenovo",
  "device.system-product": "20l8s2n80p",
  "device.system-serial": "pc14lwfg"
}
--->

After that, we need to apply our workload:

---> kubectl apply -f /tmp/deploy.yaml
edgeworkload.management.project-flotta.io/workload created

When the device get the next hearbeat, it’ll retrieve the config, and from there will deploy it. Edge administratior can review workload status in the device status:

---> kubectl get edgedevice camera-ny -o json | jq .status.workloads
[
  {
    "lastTransitionTime": "2022-05-04T15:32:07Z",
    "name": "workload",
    "phase": "Running"
  }
]