Skip to content

Latest commit

 

History

History
21 lines (11 loc) · 1.13 KB

README.md

File metadata and controls

21 lines (11 loc) · 1.13 KB

KEDA multi metric scaler - using Prometheus Trigger

This sample demonstrates how to use KEDA to scale a Kubernetes deployment based on multiple metrics. Specifically for not scaling up deployment when the error threshold is exceeded, else scaling based on a queue length metric. This

Installation

The setup.sh script installs the required components for this sample. Details of the components have been added as comments in the script.

Key points to note:

  • After the Line in the setup.sh script, which builds the test go application container image, and pushes it to ACR, you should update the ACR_NAME with your created ACR name in deployment.yaml file, for both the deployments.
  • The Prometheus scrape config scrapes the test go apps for the msg_queue_length and rate_429_errors metrics.

Prometheus Query

Details of the prometheus query used, can be found in the keda-scaled-object.yaml

KEDA scaling tests

After the installation, you can follow the steps in tests.sh to test the scaling behavior.