Metrics Not Showing in Grafana with Kong Ingress Controller in Database Mode

1

Hi Kong Nation,

I set up monitoring for the Kong Ingress Controller using Prometheus and Grafana, following the documentation provided. [Ref] While everything works fine in DB-less mode, I’m encountering an issue with database mode: the Grafana dashboard shows “no data” for all metrics.

Here are the steps I took:

  1. Installed Kong Ingress Controller in a Kubernetes cluster using Helm.
  2. Configured Prometheus with a scrape interval of 10 seconds and installed Grafana.
  3. Enabled the Prometheus plugin in Kong globally.
  4. Set gateway.serviceMonitor.enabled=true in the Helm upgrade command.
  5. Generated traffic using curl commands to test metrics collection.

Despite these configurations, metrics aren’t appearing in Grafana when using the database mode.

Could you provide insights or troubleshooting steps to resolve this issue?

2

I used db less mode and removed the gateway image tag so that it will pull the latest one by default.

helm install kong kong/ingress -n kong --create-namespace --set gateway.image.repository=kong --set gateway.image.tag="3.9.0"

also I removed the version tag when executing the below command so that i can get the latest one.

helm install promstack prometheus-community/kube-prometheus-stack --namespace monitoring --version 52.1.0 -f values-monitoring.yaml

After doing these, when I run

helm upgrade kong kong/ingress -n kong --set gateway.serviceMonitor.enabled=true --set gateway.serviceMonitor.labels.release=promstack

Kong creates a new gateway pod but its not getting ready. I tried the same with the versions everything works fine. Any insights on this? on how to use the latest version atleast for promstack since it has better GUI?