Prometheus Question

Hi there - just a quick question - I am getting no information on the Grafana Dashboard - everything is working fine and prometheus services are running normally but I get the error “A is undefined” - any suggestions really appreciated - thanks

can you tell a little more about your setup?

Some questions to help debug:

  • can you curl the node port on the 12798 on the metrics endpoint? What is the result. Eg curl http://<node_ip>:12798/metrics ?
  • when you install prometheus and grafana, you should be able to see also other stats, like the prometheus stats, can you see them?
  • did you setup the prometheus datasource on grafana? It gives you option to test the connection, was it successful?
  • can you see cardano node metrics in prometheus? they are all prefixed with cardano_ prometheus has a nice autocompletion, so if the metrics are correctly ingested, when you start typing their name in prometheus it will give you suggestions.

Thanks for responding. Yes I can run the curl command and I get a whole list of metrics back with prefixes Cardano_node_metrics etc . I followed the install of prometheus exactly by following the Coincashew instructions at
https://www.coincashew.com/coins/overview-ada/guide-how-to-build-a-haskell-stakepool-node#16-setup-prometheus-and-grafana-dashboard - by following this I managed to start and test the services and they all responded as they should. All ports have been open on the relay node, block producing node and firewall as instructed. Connections all appeared ok. When I look at my Dashboard however everything just has No Data in it.

Sorry but I’m not familiar with the coincashew setup.

Do you have access to prometheus? Can you see the metrics there?

Yes I can access to prometheus through http://ip address:9090 - I can see that it is connecting

Great, and can you see metrics there? See screenshot

Yes I get this

cardano_node_metrics_upTime_ns{alias=“block-producer-node”,instance=“My_IP:12798”,job=“prometheus”,type=“cardano-node”} 43952574773156
cardano_node_metrics_upTime_ns{alias=“relaynode1”,instance=“localhost:12798”,job=“prometheus”,type=“cardano-node”}

Righteo,

so it must be something at grafana level.

you said that when you click the “test” button on the prometheus Datasource in grafana, the result is ok, right? In that case it mean the plumbing between prometheus and grafana are alright.

All leads to the actual dashboard. Does it have any variables? Have you tried to manually set them? You can access dashboards variables via teh settings icon… good luck

Thanks for the advice - I will have a look at the dashboard - not really sure what I can do as everything shows that “a” is not defined. I will try anyway and see if I can fix it