You'll have to clone the original event with the clone filter and add the extra fields only to the clone that you send to graphite. Use conditionals in the output block to send events with extra fields to graphite and all other events to elasticsearch.
For the graphite fields I use the ruby filter to generate them, if I use the clone filter would I still be able to use the ruby filter as well?
I had a look at the clone documentation but it is not clear so I will have to try it out but just checking if you might know. So what I'm thinking is that I use the clone filter and add a custom tag and fields I need. And then later in the filter pipeline I use a conditional check against that custom tag and then use a ruby filter to do the work of adding the Graphite fields. Would this be possible ?
For the graphite fields I use the ruby filter to generate them, if I use the clone filter would I still be able to use the ruby filter as well?
Yes, of course.
So what I'm thinking is that I use the clone filter and add a custom tag and fields I need. And then later in the filter pipeline I use a conditional check against that custom tag and then use a ruby filter to do the work of adding the Graphite fields. Would this be possible ?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.