After much searching through Google and these forums, I've decided I've pulled my hair out enough to try to figure out if what I'm trying to accomplish is possible and am making this post.
All components are version 6.2.2
I want to, within a single gauge, display an average of data, and the count of that data across multiple hosts. The gauge itself will fill based on the average, and want to have a label of the number of hits that went into the average calculation. Hope that makes sense.
Here's a quick mockup of what I'm shooting for:
Can this even be accomplished in a single gauge in Kibana?
The goal here is to display the relevant information (average response time and number of hits) about a host in one spot on the dashboard - as opposed to displaying one visualization for each metric we care about.
You can do everything you've asked for except describe how many hits were used to calculate the number average. To get that information you'll need another visualization that can be presented next to this one on a dashboard. I recommend a data, which could provide a pretty easy to skim breakdown of the number of hits per host (or whatever the data is).
That said, I think it would make for an interesting feature request if you want to file an issue at github.com/elastic/kibana/issues/new.
It looks like you know how to get the multi-guage part working, but in case you don't here is the configuration I used to get it:
I don't think I was clear enough though - I shouldn't have said "within a single visualization." (I've updated the original post to be more clear I hope). I can add additional metrics to the visualization to get all the data I want - but I'd like to display both average response time and a count of the hits on a single gauge (as in the mockup I provided).
We have nearly 20 hosts to monitor - so to display count and average separately would consume 40 visuals (20 gauges for the avg response time, and another 20 for the counts).
Hope that clears things up!
Yeah, you won't be able to get the count into the gauges that Kibana has today, but a feature request could potentially get that added in the future. What I would recommend doing today is using gauges for your averages and a data table for your counts. The data table is much more dense and can be used to lookup the count for a specific gauge. (rather than having two banks of gauges, one for averages and one for counts)
If your goal is to manually compare the count with the average so you can ignore gauges with low/high counts you might be able to trim down the list of hosts a bit using
min_doc_count, which will discard hosts that have less than some number of hits. You can do this with the "advanced json" input that's tucked away at the bottom of the terms aggregation config panel. In that text box put:
to only show buckets that have 10 or more hits.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.