I’m trying to figure out how to use the Status Panel plugin to create a high level status view of Kubernetes deployments, as reported by x-pack into Elasticsearch. I want to see one block per kubernetes.deployment.name
and color the block based on the value of kubernetes.deployment.replicas.unavailable
.
After reading through the documentation provided for this plugin, I am absolutely clueless as to how it works. It’s also possible that it just doesn’t work with Elasticsearch as a datasource.
My first thought was to do something like this:
- Use the query
metricset.name: "state_deployment"
. - Use the metric
Max
ofkubernetes.deployment.replicas.unavailable
. - Use Group by, Terms and use
kubernetes.deployment.name
. - Use a Relative time of
1m
.
With these settings, the panel is completely blank. If I make some innocuous change to any value in the Visualization settings, it will turn into one giant green OK block. If I refresh the page, it will be completely blank again.
How do I use this panel? Do I need to do something fancy within the query to populate some variables? If I Group by, Date histogram or Group by both Terms, then Date histogram, I get the same result with a blank panel.
I am open to alternatives such as other panel plugins, or dashboard templates, etc. I have similar results with the Polystat panel plugin. I use a query with a Group by field, and the panel is completely blank.