Resizable Table Legends and Multiple Terms aggregations

Hello,

In Grafana we have the option to show legends as a table, showing a lot of useful aggregated info, for example:

I have tried many things to achieve the same in Kibana, but never managed to get even close to what Grafana can do.

There seem to be multiple issues:

  1. Resizable legends

I find several GH issues for resizable legends:



Imho it's really important to be able to customize the legend width, as currently some legends show the same values, see for example Custom legend with for Vertical Bar Chart

  1. Legend tables

I can find the following issue: https://github.com/elastic/kibana/issues/16096

This i a really important one, as it allows to show the total average, max, min and current in one graph. Currently we'd have to create multiple series to achieve the same in Kibana.

  1. Nested Terms aggregations

Afaik there seems to be no easy way to use multiple terms aggregations in one graph in a usable way with Kibana TSVB. In Grafana, that's a piece of cake, for example:

This allows us to visualise (for example) a combination of host name, process name and process id. it's of course usable for many other things, such as a combination of host name and disk name, etc..

The combination of the above points is really powerful and allows us to show in one Grafana graph the same as where we'd need at least 2 or 3 graphs for in Kibana.

It would be nice to see some evolution in the above, as currently we are forced to keep using Grafana for metric related stuff. Neither Kibana's Vertical bar charts, TSVB, nor Lens has these features.

Willem

Hi @willemdh,

thanks a lot for this great summary! I totally understand the need for these features, I will raise this internally with the team.

Side note: Feature 3 (nested terms aggregations) should be possible using the "regular" line/bar/area chart in Kibana.

Wel, you are right that it is possible, but did you ever try it? The non-customizable legend width / table completely screws that up actually.