Getting Data In

How to extract sum totals for data in a nested JSON object?

stephenmelrose
Engager

Hi there,

I have the following log line format (slightly edited for anonymity),

2013-08-14T08:54:10.098+0100 [INFO] {process-0.1735} spin-processor - [TIMER] some.request.complete {"id":"a1b2c3d4e5","sequenceNo":2,"fooId":"f6g7h8i9j0","duration":2.981,"totalActions":170,"actions":{"Foo.UPDATED":1,"Bar.UPDATED":169}}

The bit my question relates to is the actions key in the JSON object. Ultimately what I want is to get a table containing the sum total for each action type over a given time period, e.g.

+----------------------+
| Action       | Total |
+----------------------+
| Foo.INSERTED | 2     |
| Foo.UPDATED  | 56    |
| Bar.INSERTED | 169   |
| Bar.UPDATED  | 5678  |
+----------------------+

Any idea how I would do this?

The closest I've got so far is to manually do a sum() on each type (as I'm already SPATHing the JSON), but the table it gives me is the wrong way round to what I want, e.g.

+-----------------------------------------------------+
| sum(actions.Foo.UPDATED) | sum(actions.Bar.UPDATED) |
+-----------------------------------------------------+
| 56                       | 5678                     |
+-----------------------------------------------------+

Thanks.

Tags (3)

russellliss
Path Finder

I was sort of able to do this myself. You may want to try

Based on your example above:

| spath input=some.request.complete actions{} | mvexpand actions{} | spath input=actions{} | stats sum("Bar.UPDATED") by "Foo.UPDATED"

0 Karma
Get Updates on the Splunk Community!

Splunk Enterprise Security 8.0.2 Availability: On cloud and On-premise!

A few months ago, we released Splunk Enterprise Security 8.0 for our cloud customers. Today, we are excited to ...

Logs to Metrics

Logs and Metrics Logs are generally unstructured text or structured events emitted by applications and written ...

Developer Spotlight with Paul Stout

Welcome to our very first developer spotlight release series where we'll feature some awesome Splunk ...