You also have a target specified as part of your grok ... this will result in any grok'ed fields being sub-fields to the specified top-level field called fecha - ie: your extracted date field will actually be in fecha.fecha. You should remove the target in the grok or if you need it, then to access the date field, your date match syntax needs to be the following - as well as taking @Badger's comment into consideration or use the recommended pattern in LS docs:
match => [ "[fecha][fecha]", "MMM dd HH:mm:ss"]`.
You can also simplify your grok match to leverage the built-in syslog time format:
match => {"message" => "%{SYSLOGTIMESTAMP:fecha}"}`
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.