Good afternoon,
I want to ask if there is a way how to create own _raw data and to fire some kind of SPL query when some new event come to splunk ( the best would be if it is possible to make it before indexing ).
The whole idea is: When a new event comes, splunk will create a brand new event and save it in different index with little set of information from previous event + some extra info (for ex. actual timestamp).
I tried to make this work with alert, by "alerting" each new event to a new index. But the problem is that solution is extremely slow. I tried to upload 20k of new data at once and it took about hour to parse all of these.
Here is the pic of my idea:
That line between server and indexes is splunk server, the best would be if it gonna happen at pre or during indexing time, but if it happens after I wouldn't mind. 🙂
Hey@Kozokkon,
You can try using summary indexing but this can be done after indexing data for the first time.
http://docs.splunk.com/Documentation/Splunk/7.1.1/Knowledge/Usesummaryindexing
Let me know if this helps!!
When you say alerting, do you mean alert search with summary index option? If not, have you explored option of summary indexing for this?
That was kinda a hack, i set alert per every new event and made a spl where I totally recreate my _raw output as I want and set it to save to another index.
What problem are you trying to solve? Perhaps there is another way to accomplish your goal.
In short: I'm receiving JSON so everything is parsed and saved as field. I need to anonymized some information and reduce some fields for search optimization.
Data should be anonymized before it is indexed. Writing the anonymized data to a different index means the original version is still available in another index.
Consider using a scripted or modular input to process the data as you wish before indexing.
Mine idea was to parse already received information to another index with own redefined structure and variables and make searches on them.