Splunk Search

Comparing data from 2 days by data type?

Min1025
Explorer

Hi

I want to compare the data from 2 days by data type, my expected result is as below, is it possible?

Data Type Yesterday Yesterday Count Today Today Count Count Change Rate
A 2023/3/26 15 2023/3/27 18 0.20
B 2023/3/26 20 2023/3/27 19 -0.05
C 2023/3/26 16 2023/3/27 35 1.19
D 2023/3/26 21 2023/3/27 40 0.90
E 2023/3/26 30 2023/3/27 25 -0.17
F 2023/3/26 40 2023/3/27 50 0.25
Labels (1)
0 Karma

Min1025
Explorer

Hi @gcusello,

Thank you for the query. I tried this query and encountered 2 issues.

1. The previous date returned both previous day and yesterday.

2. The previous count is same as yesterday count.

Min1025_0-1680172230135.png

Do you have any idea for above issues?

Thanks,

Min

0 Karma

gcusello
SplunkTrust
SplunkTrust

Hi @Min1025,

please try this:

<your_search> earliest=-2d@d latest=@d
| eval 
   day=strftime(_time,"%Y-%m-%d"),
   yesterday=strftime(now()-86400,"%Y-%m-%d"),
   previous_day=strftime(now()-172800,"%Y-%m-%d")
| stats 
   values(previous_day) AS "Previous day" 
   count(eval(if(day=previous_day,1,0))) AS "Previous day Count"
   values(yesterday) AS "Yesterday" 
   count(eval(if(day=yesterday,1,0))) AS "Yesterday Count"
   BY "Data Type"
| eval "Count Change Rate"=round(1-"Previous day Count"/"Yesterday Count",2)

Ciao.

Giuseppe

0 Karma

gcusello
SplunkTrust
SplunkTrust

HI @Min1025,

there's a thing that I don't understand: todays count will always be less than yesterdat count because today is in progress, maybe it could be better to count two days ago with yesterday, to have two full days to compare, anyway, please try something like this:

<your_search> earliest=-d@d latest=
| eval day=strftime(_time,"%Y-%m-%d"), today=strftime(now(),"%Y-%m-%d")
| stats 
   values(eval(if(day!=today,day,"") AS "Yesterday" 
   count(eval(if(day!=today,1,0)))  AS "Yesterday Count"
   values(eval(if(day=today,day,"") AS "Today" 
   count(eval(if(day=today,1,0)))  AS "Today Count"
   BY "Data Type"
| eval "Count Change Rate"=round(1-"Yesterday Count"/"Today Count",2)

Ciao.

Giuseppe

Min1025
Explorer

Hi @gcusello,

Thank you for the query. You are right, it's better to count two days ago with yesterday and compare the results for two full days.  Could you please guide the query for this?

I tried the query you provided but I got the same count for "today" and "yesterday", not sure why.

Min1025_1-1679991124350.png

Thanks,

Min

 

0 Karma

gcusello
SplunkTrust
SplunkTrust

Hi @Min1025,

ok, in this case, you have only to modify the time period and something else, please try this:

<your_search> earliest=-2d@d latest=@d
| eval 
   day=strftime(_time,"%Y-%m-%d")
   previous_day=if(now()-_time<86400,day,"")
   yesterday=if(now()-_time>86400,day,"")
| stats 
   values(previous_day) AS "Previous day" 
   count(eval(if(day=previous_day,1,0))) AS "Previous day Count"
   max(yesterday) AS "Yesterday" 
   count(eval(if(day=yesterday,1,0))) AS "Yesterday Count"
   BY "Data Type"
| eval "Count Change Rate"=round(1-"Previous day Count"/"Yesterday Count",2)

 Ciao.

Giuseppe

Got questions? Get answers!

Join the Splunk Community Slack to learn, troubleshoot, and make connections with fellow Splunk practitioners in real time!

Meet up IRL or virtually!

Join Splunk User Groups to connect and learn in-person by region or remotely by topic or industry.

Get Updates on the Splunk Community!

[Puzzles] Solve, Learn, Repeat: Matching cron expressions

This puzzle (first published here) is based on matching timestamps to cron expressions.All the timestamps ...

Why Splunk Customers Should Attend Cisco Live 2026 Las Vegas

Why Splunk Customers Should Attend Cisco Live 2026 Las Vegas     Cisco Live 2026 is almost here, and this ...

Data Management Digest – May 2026

Welcome to the May 2026 edition of Data Management Digest!   As your trusted partner in data innovation, the ...