1

I am trying to monitor my Vespa cluster (with the help of the Prometheus exporter), but I can't find the right metrics to observe to know the space my index is taking, nor the space my replicas are taking. And I would also like to find a simple way of visualizing the number of documents that are indexed in my cluster, but I can't find a simple way of doing that. I have found the vespa_container_documents_total metric, but its value if always zero. The only way I've found to get its real value is to perform a search request on the cluster, then, this metric is populated. But only for a few time (like one minute), and then it gets back to zero.

So, is there a way to simply monitor those two metrics ?

dkurzaj
  • 346
  • 4
  • 13

1 Answers1

3

Take a look at https://docs.vespa.ai/documentation/reference/metrics-health-format.html, you want to gather metrics from searchnode and not the container. If you fetch metrics from the search node metric port you'll find a ton of metrics related to disk usage, documents indexed, documents active +++.

Jo Kristian Bergum
  • 2,984
  • 5
  • 8