jvb-autoscaler.md

Resource name could be:

1. jitsi_conference - Metrics based on active JVB conference count for 15m.
2. jitsi_participants - Metrics based on active JVB participants count for 15m.
3. cpu - Metrics based on "Container_Cpu_Usage" (not working with influx right now).

Prometheus example:

apiVersion: meeting.ko/v1alpha1
kind: AutoScaler
metadata:
  name: jas-sample
spec:
  monitoringType: "prometheus"
  host: "http://172.28.174.90:9090/"
  interval: "60s"
  scaleTargetRef:
    name: jitsi-sample
  minReplicas: 1
  maxReplicas: 3
  metric:
    name: jitsi_participants
    targetAverageUtilization: 40

InfluxDB example:

apiVersion: meeting.ko/v1alpha1
kind: AutoScaler
metadata:
  name: jas-influx-sample
  annotations:
    jas.influxdb/token: "e9XEamUiqMeoa0HmsXiZ8zIe5BDr2p8D"
    jas.influxdb/org: "influxdata" # This could be either the organization name or the ID.
    jas.influxdb/bucket: "jitsi"
spec:
  monitoringType: "influxdb"
  host: "http://influx-influxdb2:80/"
  interval: "60s"
  scaleTargetRef:
    name: jitsi-sample
  minReplicas: 1
  maxReplicas: 3
  metric:
    name: jitsi_participants
    targetAverageUtilization: 40

For InfluxDB, you can set up next fields in annotations:

1. jas.influxdb/token = InfluxDB auth token. Field is being required.
2. jas.influxdb/org = InfluxDB organization name. If field not provided, then it would be equal to "influxdata".
3. jas.influxdb/bucket = InfluxDB bucket with jitsi metrics. If field not provided, then it would be equal to "jitsi".