I have two Splunk queries, each of which uses the _rex command to extract the join field.
Example:
QUERY 1
index=index1 "Query1" | rex field=_raw "abc(?<MY_JOIN_FIELD>def)"
QUERY 2
index=index2 "Query2" | rex field=_raw "ghi(?<MY_JOIN_FIELD>jkl)"
I want to use the Transaction command to correlate these two queries, but I can't figure out how to do it.
Thanks!
Jonathan
Hi @jbrenner ,
you should try something like this:
index=index1
| rex field=_raw "abc(?<MY_JOIN_FIELD>def)"
| append [ search
index=index2
| rex field=_raw "ghi(?<MY_JOIN_FIELD>jkl)"
]
| transaction <MY_JOIN_FIELD>
| ...
otherwise, you could extract both the MY_JOIN_FIELD before so you don't need to extract them in the search, in other words:
index=index1 OR index=index2
| transaction <MY_JOIN_FIELD>
| ...
But i invite you to think in a different way:
At first extract the two fields before and then try to use stats command instead of transaction:
index=index1 OR index=index2
| stats values(field1) AS field1 values(field2) AS field2 values(field3) AS field3 BY <MY_JOIN_FIELD>
where field1, field2 and field3 are the fields you need in your events.
This solution is very much performant than using transaction.
Ciao.
Giuseppe
Hi @jbrenner ,
you should try something like this:
index=index1
| rex field=_raw "abc(?<MY_JOIN_FIELD>def)"
| append [ search
index=index2
| rex field=_raw "ghi(?<MY_JOIN_FIELD>jkl)"
]
| transaction <MY_JOIN_FIELD>
| ...
otherwise, you could extract both the MY_JOIN_FIELD before so you don't need to extract them in the search, in other words:
index=index1 OR index=index2
| transaction <MY_JOIN_FIELD>
| ...
But i invite you to think in a different way:
At first extract the two fields before and then try to use stats command instead of transaction:
index=index1 OR index=index2
| stats values(field1) AS field1 values(field2) AS field2 values(field3) AS field3 BY <MY_JOIN_FIELD>
where field1, field2 and field3 are the fields you need in your events.
This solution is very much performant than using transaction.
Ciao.
Giuseppe
Hi @jbrenner,
good for you, see next time!
Ciao and happy splunking
Giuseppe
P.S.: Karma Points are appreciated 😉
@gcusello Thanks Giuseppe! Worked like a charm! I originally considered your alternative solution, but this is a one-off scenario, and it's unlikely we will ever use these extracted fields again.