- Mark as New
- Bookmark Message
- Subscribe to Message
- Mute Message
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Creating Pivot Chart with two sums?

I am feeling more and more like the new Pivot UI functionality is way too limiting.
Can anyone help me to do a query like this:
sourcetype=xxx-prod (FOO OR BAR) | timechart span=30m sum(FOO) as foo sum(BAR) as bar
I know this seems like it would be an easy pivot to create. Simply auto-extract two fields, then press the + button the columns in the pivot table creation screen and add another column, creating a second columns of sums. Two columns of sums divided by time. The stats table looks great.
Then when I go to turn it into a bar graph, they force me to choose to identify my y-axis with only one sum. I explored the "Color" option and it won't allow me to include anything that is a sum, but rather straight up fields.
Please if anyone can help me understand that would be great. I'm trying to get together a presentation on the usefulness of data models and pivots for my company and I keep finding more reasons not to use it than reasons to use it.
Thanks.
- Mark as New
- Bookmark Message
- Subscribe to Message
- Mute Message
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Unfortunately, you cannot do multiple aggregators in the pivot ui.
You could use the pivot command. ie.
| pivot datamodel object sum(foo) AS "foo" sum(bar) AS "bar" SPLITROW _time AS _time PERIOD auto
Obviously this kind of defeats the purpose of using pivot outside of the acceleration benefits but hopefully they will iterate over the interface and make it more useful.
- Mark as New
- Bookmark Message
- Subscribe to Message
- Mute Message
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content

Thank you for your answer. It clarified whether it is possible or not. Unfortunately, you are right, using the query itself defeating the purpose of getting the acceleration benefits and bypass the use of the pivot UI completely. I have a list of questions/concerns about the data models/pivot system they have in place that I am taking with me to Splunk conference.
