I am doing a search query where there will be a dynamic client ID with either a success or a failure result code -- I want to look at all the successes and failures based on a client ID.
The client ID always comes after the method name, a comma, the code (for success or failure), a space and comma, then the client ID followed by a comma. This is an example of what the event logs look like:
<A bunch of random logging stuff> ~aStaticMethodName,SuccessOrFailureCode5555 ,ClientID, <Another bunch of random logging stuff>
I am using 'rex' to find the Client IDs for both the successes or failures by client ID like this:
"My Search Here"
| rex "~myMethod,successCode0000 ,(?<success>[^,]+),"
| rex "~myMethod,failureCode1111 ,(?<failure>[^,]+),"
I then add a | chart count by success
(or change the success to failure) and can get an output looking like this:
The successes and failures will have common client IDs (client A, client B, etc).
What I want to do, is combine both success and failure into one chart grouped by client ID like this:
Is there a way to do this in Splunk? Help is greatly appreciated!
try something like this:
"My Search Here"
| rex "~myMethod,successCode0000 ,(?<success>[^,]+),"
| rex "~myMethod,failureCode1111 ,(?<failure>[^,]+),"
| eval clients=coalesce(success,failure)
| stats count(success) as success count(failure) as failure by clients
Try like this
"My Search Here"
| rex "~myMethod,(?<status>[^,]+) ,(?<clientID>[^,]+),"
| replace "successCode0000" with "Success" "failureCode1111" with "Failure" in status
| chart count over clientID by status
try something like this:
"My Search Here"
| rex "~myMethod,successCode0000 ,(?<success>[^,]+),"
| rex "~myMethod,failureCode1111 ,(?<failure>[^,]+),"
| eval clients=coalesce(success,failure)
| stats count(success) as success count(failure) as failure by clients