Why is the license usage much larger than the file uploaded?

[sanderwubben]

I have a 10GB developer license on my local Splunk instance. Today I checked my licence usage: 1GB, then I added a data file of 500MB through the 'Add Data' button. Now I got a warning that my license usage today is at 11GB and that a license warning will be generated.

I thought licensing counted the amount of raw data indexed. Is another metric used that could inflate my data by 20x?

For the record, I have almost no other data inputs running.

[emiller42]

Was that 500MB file compressed? a gzipped logfile of that size is likely closer to 3-4GB when uncompressed, and that's what Splunk is considering when doing license calculations.

[somesoni2 · Accepted solution]

I would suggest to look at the license_usage log (index=_internal sourcetype=splunkd source=*license_usage.log type=Usage)to see which host/index/sourcetype is causing all that 11 GB of licensing and go from there.

I stand corrected, thanks @emiller42. Below is not correct, DO NOT FOLLOW
The license usage in Splunk is not only the size of raw data files it monitors but includes all associated files (index files and raw data file) which Splunk has to generate to make it searchable, so depending upon you data and processing rules, it can be larger than the actual data file.

[sanderwubben]

It seems to have been the 1D10T problem, although your answer did help me. I decompressed the full file and split it into 500MB files for uploading, but through the license_usage log I noticed I still had a folder watch on that same folder I decompressed in, so every file was indexed multiple times.
Thanks for your help.

[emiller42]

The license usage in Splunk is not only the size of raw data files it monitors but includes all associated files (index files and raw data file) which Splunk has to generate to make it searchable

That is not correct, License usage is solely based on the volume of data ingested.