Splunk Enterprise

Different timestamp values for the same file indexed.

peterkn
Explorer

Hi, 

I have a data input (Directory Monitor) for /opt/splunk/data/test

Everyday a new csv file is copy pasted in this directory, and Splunk would start indexing them. 

However all rows in this csv file are indexed with different timestamp (_time) values. Eg the file has 3382 events indexed, but doing a 

 

 

index=caseload host=XXXX source="/opt/splunk/data/test/testfile.csv" | stats count by _time

 

 

would yield something like

_timecount
2015-04-17 04:56:4922
2016-01-08 19:51:4933
2016-01-18 12:20:0911
2016-02-07 21:15:0918

 

shouldn't it be all current time which is "2020-06-17 10:10:10" for instance, instead of various different timestamps, I'm thinking it is trying to find some value per row that represents a timestamp and parse it, but I don't even see any "2015-04-17" in those 22 rows. 

How do I make all the Directory Monitors to index each event using current timestamp? 

0 Karma
1 Solution

rnowitzki
Builder

Hi @peterkn ,

Can you share a few lines of the csv? 

Splunk tries to get a time for each event / row. It seems that is uses a field that not really is a timestamp or fails to get the correct values from it. Maybe some number that could be interpreted as a unix/epoch timestamp.

You have to tell Splunk where the timestamp is  and how to interprete it. Either in the UI or in props.conf.
=> Explained here. 

The "current time" / time when indexing is the last option being used.

If there is no timestamp, you could add one to each row with (e.g.) sed. If a script copies the file, it would be an easy enhancement.
Or you configure props.conf as decribed here to really use the current time/index time.

--
Karma and/or Solution tagging appreciated.

View solution in original post

rnowitzki
Builder

Hi @peterkn ,

Can you share a few lines of the csv? 

Splunk tries to get a time for each event / row. It seems that is uses a field that not really is a timestamp or fails to get the correct values from it. Maybe some number that could be interpreted as a unix/epoch timestamp.

You have to tell Splunk where the timestamp is  and how to interprete it. Either in the UI or in props.conf.
=> Explained here. 

The "current time" / time when indexing is the last option being used.

If there is no timestamp, you could add one to each row with (e.g.) sed. If a script copies the file, it would be an easy enhancement.
Or you configure props.conf as decribed here to really use the current time/index time.

--
Karma and/or Solution tagging appreciated.

peterkn
Explorer

You're an absolute champion. 

Due to the nature of the data in the file I'm not legally allowed to share it's content unfortunately. 

For those who are interested, instead of modifying/creating props.conf, I changed the timestamp setting from when adding a new file to an index (Settings>Add Data>Upload), select the sourcetype (csv in my case), under Source Type there is a dropdown for Timestamp, my defaulted to "Automatic" so I changed it to "Current", upon clicking "Next" it will ask me to save, select yes, overwrite. This setting will apply to all input monitors, please make sure you restart Splunk. 

Thanks again @rnowitzki 

0 Karma

rnowitzki
Builder

You're welcome.

And you actually changed props.conf with that, but using the UI instead of CLI/vi 🙂

Happy splunking.

--
Karma and/or Solution tagging appreciated.
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!

Index This | What travels the world but is also stuck in place?

April 2026 Edition  Hayyy Splunk Education Enthusiasts and the Eternally Curious!   We’re back with this ...

Discover New Use Cases: Unlock Greater Value from Your Existing Splunk Data

Realizing the full potential of your Splunk investment requires more than just understanding current usage; it ...

Continue Your Journey: Join Session 2 of the Data Management and Federation Bootcamp ...

As data volumes continue to grow and environments become more distributed, managing and optimizing data ...