I have a running server where logstash is working. In my pipelines.yml there's a line path.config: "/etc/logstash/conf.d/*.conf" - this points to a directory where my configuration is.
The directory contains following files:
filter-nb.conf
grok_patterns - that is a directory where is a file containing grok patterns
input-kafka.conf
output-elasticsearch.conf
The content of output-elasticsearch.conf is as follows:
Check input-kafka.conf, there would be details for the connection.
If is the Kafka connection, then LS is connecting/subscribing to Kafka. So there is no listening port on LS side, except TCP 9600 for internal LS monitoring like event statistics: curl -XGET 'localhost:9600/_node/stats/events?pretty .
Logstash allows structuring the pipeline over multiple files so all *.conf files are read and loaded as a single pipeline. Therefore, you can find the input in the file input-kafka.conf. I this case, Logstash will not open a port for listening but instead connect to a Kafka server and subscribe to a Kafka topic.
Will it work?
The point is to leave the current process unchanged (production environment) except when post contains field named 'application' and the value is 'fointe-test' then save the log in elasticsearch host in index 'fointe-test-%{+YYYY.MM.dd}'.
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.