Splunk Search

How to count events that occurred 60 minutes apart (not on the hour)?

srive326
Explorer

Hello everyone,

I need help with a search.
I have a table with the following fields:

VISITDATE           USERNUMBER     WEBSITE
4/19/2020 7:15:11     001    www.google.com
4/26/2020 10:24:03    001    www.google.com
4/26/2020 10:33:03    001    www.google.com
4/26/2020 11:15:12    001    www.google.com
4/26/2020 11:30:12    001    www.google.com
4/28/2020 14:30:12    001    www.google.com

I want to count the number of visits (by usernumber and website) that occurred 60 minutes apart, and return the earliest
time in that 60-minute bucket.

VISITDATE           USERNUMBER     WEBSITE  COUNT
4/19/2020 7:15:11     001    www.google.com      1
4/26/2020 10:24:03    001    www.google.com   3
4/26/2020 11:30:12    001    www.google.com   1
4/28/2020 14:30:12    001    www.google.com   1

For example, using bins or buckets with a 60-min/1-hour time span that snap to the hour 10:00 -11:00 is one hour and that would give me this result, which I don't want:

VISITDATE           USERNUMBER     WEBSITE  COUNT
4/19/2020 7:15:11     001    www.google.com      1
4/26/2020 10:24:03    001    www.google.com   2
4/26/2020 11:15:12    001    www.google.com   2
4/28/2020 14:30:12    001    www.google.com   1

Thank you in advance for your help.

0 Karma
1 Solution

to4kawa
Ultra Champion
| makeresults 
| eval _raw="VISITDATE,USERNUMBER,WEBSITE
4/19/2020 7:15:11,001,www.google.com
4/26/2020 10:24:03,001,www.google.com
4/26/2020 10:33:03,001,www.google.com
4/26/2020 11:15:12,001,www.google.com
4/26/2020 11:30:12,001,www.google.com
4/28/2020 14:30:12,001,www.google.com" 
| multikv forceheader=1 
| eval _time=strptime(VISITDATE,"%m/%d/%Y %T") 
| autoregress _time as prev_time
| streamstats count(eval(_time - prev_time > 3600)) as session1
| streamstats min(_time) as session_min by session1
| eval session2=if(_time - session_min <= 3600, "T","N")
| eval tmp=session1.session2
| streamstats dc(tmp) as session3
| stats count as COUNT min(_time) as _time by USERNUMBER WEBSITE session3
| eval VISITDATE=strftime(_time,"%m/%d/%Y %T")
| table VISITDATE,USERNUMBER,WEBSITE,COUNT

View solution in original post

0 Karma

to4kawa
Ultra Champion
| makeresults 
| eval _raw="VISITDATE,USERNUMBER,WEBSITE
4/19/2020 7:15:11,001,www.google.com
4/26/2020 10:24:03,001,www.google.com
4/26/2020 10:33:03,001,www.google.com
4/26/2020 11:15:12,001,www.google.com
4/26/2020 11:30:12,001,www.google.com
4/28/2020 14:30:12,001,www.google.com" 
| multikv forceheader=1 
| eval _time=strptime(VISITDATE,"%m/%d/%Y %T") 
| autoregress _time as prev_time
| streamstats count(eval(_time - prev_time > 3600)) as session1
| streamstats min(_time) as session_min by session1
| eval session2=if(_time - session_min <= 3600, "T","N")
| eval tmp=session1.session2
| streamstats dc(tmp) as session3
| stats count as COUNT min(_time) as _time by USERNUMBER WEBSITE session3
| eval VISITDATE=strftime(_time,"%m/%d/%Y %T")
| table VISITDATE,USERNUMBER,WEBSITE,COUNT
0 Karma

srive326
Explorer

Hi,

Thank you so much for your answer. I implemented it on my data and I noticed that there is one thing I should add. There can be multiple users/USERNUMBER so using
|autoregress _time as prev_time
on my data has the effect of using another users previous time instead of the prev_time specific to that user. Is there a way I can make the |autoregress _time as prev_time conditional to the USERNUMBERs so that for example each unique USERNUMBER will have null for the first time they visit a site on the prev_time column. Thank you again for your help.

0 Karma

to4kawa
Ultra Champion

use streamstats current=f last(_tme) as prev_time by USERNUMBER

0 Karma

srive326
Explorer

That worked perfectly, thank you!

0 Karma
Get Updates on the Splunk Community!

Splunk Observability Synthetic Monitoring - Resolved Incident on Detector Alerts

We’ve discovered a bug that affected the auto-clear of Synthetic Detectors in the Splunk Synthetic Monitoring ...

Video | Tom’s Smartness Journey Continues

Remember Splunk Community member Tom Kopchak? If you caught the first episode of our Smartness interview ...

3-2-1 Go! How Fast Can You Debug Microservices with Observability Cloud?

3-2-1 Go! How Fast Can You Debug Microservices with Observability Cloud? Learn how unique features like ...