Getting Data In

How to extract two values from two same fields in two events and then find the difference in timestamp

vanakkam777
New Member

Log data example:

{'job_no':'1','begin_build_time':'2019-08-15 11:00:00','event_type':'staging'}
{'job_no':'1','end_build_time':'2019-08-15 11:10:00','event_type':'staging'}
{'job_no':'1','begin_execution_time':'2019-08-15 11:10:01','event_type':'transaction'}
{'job_no':'1','end_execution_time':'2019-08-15 11:20:00','event_type':'transaction'}
{'job_no':'1','begin_artifact_time':'2019-08-15 11:20:01','event_type':'upload'}
{'job_no':'1','end_artifact_time':'2019-08-15 11:30:00','event_type':'upload'}

I need data table output like:
Job# 1 staging 10 mins transaction 10 mins upload 10 mins total 30 mins

Each line is coming as a event in Splunk.
I know we have to use coalesce and streamstats, but I am new to splunk not able to figure out how to get to this level.

Thanks in advance.

0 Karma

mayurr98
Super Champion

Try this run anywhere search

| makeresults 
| eval data="{'job_no':'1','begin_build_time':'2019-08-15 11:00:00','event_type':'staging'} | {'job_no':'1','end_build_time':'2019-08-15 11:10:00','event_type':'staging'}" 
| makemv data delim="|" 
| mvexpand data 
| table data 
| rex field="data" "'job_no':'(?<job_no>[^\']+)','(begin_build_time':'(?<begin_build_time>[^\']+)|end_build_time':'(?<end_build_time>[^\']+))','event_type':'(?<type>[^\']+)'"| eval begin_build_time=strptime(begin_build_time,"%Y-%m-%d %H:%M:%S"),end_build_time=strptime(end_build_time,"%Y-%m-%d %H:%M:%S") 
| stats values(begin_build_time) as begin_build_time values(end_build_time) as end_build_time by type job_no
| eval time_diff_in_min=round((end_build_time-begin_build_time)/60)

In your env, you should try

index=<your_index>
| rex field=_raw "'job_no':'(?<job_no>[^\']+)','(begin_build_time':'(?<begin_build_time>[^\']+)|end_build_time':'(?<end_build_time>[^\']+))','event_type':'(?<type>[^\']+)'"| eval begin_build_time=strptime(begin_build_time,"%Y-%m-%d %H:%M:%S"),end_build_time=strptime(end_build_time,"%Y-%m-%d %H:%M:%S") 
| stats values(begin_build_time) as begin_build_time values(end_build_time) as end_build_time by type job_no
| eval time_diff_in_min=round((end_build_time-begin_build_time)/60)
0 Karma
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!

Make Your Machine Data AI-Ready in Splunk

Every team is getting ready for AI agents.   Thursday, November 12, 2026  |  11AM  PST / 2PM EST   ...

Splunk Asynchronous Forwarding Explained

Splunk asynchronous forwarding is often misunderstood as simply setting autoLBVolume. That is not quite right. ...

How to Monitor AI Agents with Splunk Agent Observability

Let’s say I’m running a travel planning AI app in production. A user asks for three concise hotel options in ...