I was trying to follow the steps described here to access Qovery installed Grafana and search my logs but it seems it’s no longer possible or the steps have changed. I get the following error:
namespaces “prometheus” not found
Hello,
You can get it with
kubectl port-forward --namespace prometheus svc/grafana 8080:80
(be sure you target the correct cluster with your kubeconfig)
connect with your browser on localhost:8080 and the login is admin and password qovery-only
If you are using an EC2 instance (not a managed EKS) grafana is not available/installed.
This is producing error:
Error from server (NotFound): namespaces “prometheus” not found
Would you mind giving me the cluster uuid of the cluster ?
It’s cluster b7b471a2-bf49-4dcd-9dc0-bd30792290bd
You have it installed now.
Sorry for the back and forth, I was not aware that we don’t install it anymore by default.
If you need it on another cluster, feel free to ping us.
Thanks @Erebe. We’re considering moving from Datadog to Grafana Cloud to reduce costs. Is it possible to expose the existing Loki instance so we don’t have to duplicate our logs?
Hello @prki,
Sadly, no, because we don’t have any authentication/authorization in place to secure the access of loki. For now it is only accessible internally to the cluster without restriction.
Hi @Erebe, we are interested in deploying our own Grafana instance on the cluster and then hooking it up with the existing Loki instance. What are the settings like hostname and port for the Loki instance? Could you help us to configure this?
Nevermind, we figured it out. All you need is to point Grafana to http://loki.logging:3100 to access your logs from Loki.
This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.