Splunk Search

Evaluate _time against current value and next value in a foreach and mvfilter ?

dhtran
Loves-to-Learn Lots

Hello,

I need to evaluate my _time against a list of times output from a lookup table and produce a calculated field "nextPeriodTime" which is the next time after _time.

I came quite close to the final desired result by using a combination of eval, forearch and mvfilter.

My search query

index="nxs_mq" 
    | table interstep _time  
    | lookup params_vacations.csv interstep OUTPUT 0900,1000,1100,1200,1300,1400,1500,1600,1700
    | eval nextPeriodTime="" 
    | eval eventTime=strftime(_time,"%H%M") 
    | eval eventDate=strftime(_time,"%Y-%m-%d")
    | foreach *00 [| eval nextPeriodTime = if(isnotnull(mvfilter(match('<<FIELD>>',"Y"))),'<<MATCHSTR>>',"N/A")]

Results
alt text

The problem is, for example on the first record, I need the value of nextPeriodTime to be "2020-04-21 12:00:00" instead of "17" (plus, I didn't quite understand why "17" ?) because
_time is after 11:00 and before 12:00, so the next period time from _time will be 12:00.

Is it possible / does it make sense to enrich the mvfilter() part to achieve something like

mvfilter(match('<<FIELD>>',"Y") AND (_time > previous-'<<MATCHSTR>>' AND _time < next-'<<MATCHSTR>>') )

Thanks in advance for your help

0 Karma
1 Solution

to4kawa
Ultra Champion
 index="nxs_mq" 
| table interstep _time  
| lookup params_vacations.csv interstep OUTPUT 0900,1000,1100,1200,1300,1400,1500,1600,1700
| foreach *00 [eval tmp=mvappend(tmp,'<<FIELD>>')
| eval check=if(mvcount(mvdedup(tmp))=1 AND match(tmp,"Y"),1,0)]
| eval nextPeriodTime = if(check=1, strftime(relative_time(_time,"@h+1h"), "%F %H%M"), "N/A")
| eval eventTime=strftime(_time,"%H%M") 
| eval eventDate=strftime(_time,"%Y-%m-%d")
| fields - tmp check

https://docs.splunk.com/Documentation/Splunk/latest/SearchReference/Foreach

  • matchstr
    • Syntax: matchstr= <string>
    • Description: Replaces <<MATCHSTR>> with part of the field name that matches wildcard(s) in the specifier.

View solution in original post

0 Karma

to4kawa
Ultra Champion
 index="nxs_mq" 
| table interstep _time  
| lookup params_vacations.csv interstep OUTPUT 0900,1000,1100,1200,1300,1400,1500,1600,1700
| foreach *00 [eval tmp=mvappend(tmp,'<<FIELD>>')
| eval check=if(mvcount(mvdedup(tmp))=1 AND match(tmp,"Y"),1,0)]
| eval nextPeriodTime = if(check=1, strftime(relative_time(_time,"@h+1h"), "%F %H%M"), "N/A")
| eval eventTime=strftime(_time,"%H%M") 
| eval eventDate=strftime(_time,"%Y-%m-%d")
| fields - tmp check

https://docs.splunk.com/Documentation/Splunk/latest/SearchReference/Foreach

  • matchstr
    • Syntax: matchstr= <string>
    • Description: Replaces <<MATCHSTR>> with part of the field name that matches wildcard(s) in the specifier.
0 Karma

dhtran
Loves-to-Learn Lots

thank you, although I need to fix some minor details in my lookup file but this works perfectly

0 Karma
Get Updates on the Splunk Community!

Extending Observability Content to Splunk Cloud

Watch Now!   In this Extending Observability Content to Splunk Cloud Tech Talk, you'll see how to leverage ...

More Control Over Your Monitoring Costs with Archived Metrics!

What if there was a way you could keep all the metrics data you need while saving on storage costs?This is now ...

New in Observability Cloud - Explicit Bucket Histograms

Splunk introduces native support for histograms as a metric data type within Observability Cloud with Explicit ...