Splunk Search

How to extract 2 multivalue fields and search the list of each fieldA that has more than one fieldB?

rafiqul_ahsan
New Member

From following search result - I want to extract User-Name and Calling-Station-Id, and both fields have multiple values. Once I am able to extract the fields, I will then need to find the list of Calling-Station-Id and User-Name, that meets following criteria :
"The same Calling-Station-Id has more than one User-Name". I will then need to see the trend of new Calling-Station-Id that matches this criteria. What's the best way to address this problem? Can anyone please help ?

2015/06/12 09:59:12.196 (Notice) Skipping AVP AAA-Value-0=  [1] = {(User-Name)=(!afsheen77@aaa.twcwifi.com),(Framed-IP-Address)=(10.106.42.82),(NAS-IP-Address)=(24.27.229.164),(NAS-Identifier)=(nycmny83-cr02ras03.wifi.rr.com),(NAS-Port)=(0),(NAS-Port-Id)=(0/0/0/201),(NAS-Port-Type)=(Virtual),(Calling-Station-Id)=(78-4b-87-e8-0e-10),(Called-Station-Id)=(F0-B0-52-37-D8-E0:TWCWiFi),(Class)=(VISITEDMSO\=}
0 Karma
1 Solution

maciep
Champion

I didn't it test it at all, but something like this maybe.

[your initial search] | rex "User-Name\)=\((?<user>[^\)]+)" | rex "Calling-Station-Id\)=\((?<station>[^\)]+)" | stats dc(user) as user_count by station | where user_count > 1

View solution in original post

maciep
Champion

I didn't it test it at all, but something like this maybe.

[your initial search] | rex "User-Name\)=\((?<user>[^\)]+)" | rex "Calling-Station-Id\)=\((?<station>[^\)]+)" | stats dc(user) as user_count by station | where user_count > 1

rafiqul_ahsan
New Member

Thanks this worked. the 2nd thing I wanted to do is to chart (may be timechart) the station counts by time. This way I can tell how many new stations are matching this criteria over time.

0 Karma

maciep
Champion

I'm not sure how many stations you have, so the chart could be a little busy, but maybe just something like this?

[your initial search] | rex "User-Name)=((?[^)]+)" | rex "Calling-Station-Id)=((?[^)]+)" | timechart dc(user) as user_count by station

You could still pipe to where at the end too if you only want to see those that stations that have multiple users. It's really just a matter of exactly what you want to see on the timechart.

0 Karma

woodcock
Esteemed Legend

Assuming you already have the field-extractions working, this will work for the first part:

... | stats dc(User-Name) AS numUsers values(User-Name) AS UserNames BY Calling-Station-Id | where users > 1

As for this part:

I will then need to see the trend of new Calling-Station-Id that matches this criteria.

I have no idea what you mean; you will have to be much more clear.

0 Karma
Get Updates on the Splunk Community!

Stay Connected: Your Guide to May Tech Talks, Office Hours, and Webinars!

Take a look below to explore our upcoming Community Office Hours, Tech Talks, and Webinars this month. This ...

They're back! Join the SplunkTrust and MVP at .conf24

With our highly anticipated annual conference, .conf, comes the fez-wearers you can trust! The SplunkTrust, as ...

Enterprise Security Content Update (ESCU) | New Releases

Last month, the Splunk Threat Research Team had two releases of new security content via the Enterprise ...