Templating init failed Datasource named ${DS_default} was not found

[liangyuanpeng]

I have run docker image of streamnative/apache-pulsar-grafana-dashboard and I just got some error Datasource named ${DS_default} was not found when I access the topic dashboard, have something missed?
Thanks any help!

[sijie]

@liangyuanpeng have you figured the problem?

[abatilo]

Can we re-open this? I just had this happen on a clean install.

[sijie]

@abatilo Can you share your command to run the dashboard?

[abatilo]

Hi @sijie
In the context of a Kubernetes cluster deployed with this helm chart, I accessed the dashboard with kubectl -n pulsar port-forward service/pulsar-grafana 3000 and then navigating to http://localhost:3000

[milos-matijasevic]

Is something happening about this ?

[sijie]

@abatilo @milos-matijasevic Did you see any logs from the Grafana pods?

[abatilo]

I don't remember at this point, to be honest.

[sijie]

@abatilo how did you resolve the problem?

[milos-matijasevic]

Im working on it right now, and i don't see any error neither relevant logs from grafana pod

By the way it works if i manually change datasource for each panel, but i can't save changes since it's provisioned dashboard which is not editable

Also what i can notice is that DS_default variable is missing, when i add it, it works, but again i can't save it

edit
This dashboard is loaded automatically from API, and it has "__input" field in it, on this thread (https://community.grafana.com/t/set---inputs-values-when-pushing-a-dashboard-json-config-to-grafana-via-the-api/2836) it is said that "__input" field is for importing dashboard from UI, and it should not be used when importing dashboards from API

I can confirm that when i import other dashboard (https://github.com/kubernetes/ingress-nginx/blob/master/deploy/grafana/dashboards/nginx.json) with "__input" field manually from UI it works

[abatilo]

@abatilo how did you resolve the problem?

By not using Pulsar. I was evaluating it for use. But the whole ecosystem seems too young to rely on for production use.