Use a csv filter to parse the input. Store the timestamp values in the same field so that they form an array. Then use the split filter to split each input event into multiple events with the timestamp field being variable. Finally use a file output (if you indeed want to produce files) that uses a line codec with a custom format. Something like this might work:
Is it possible to send in this case also straight to ES? (I believe it is I just want to be sure before I start)
or this is from some reason a special case (since it is CSV)?
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.