I've built a ELK stack that loads a CSV on an hourly basis. The CSV is of all our tasks from jira, and so every change in an issue gets updated in the CSV, and needs to be reloaded to Kibana.
The stack works fine on loading, every new row gets sent, but logstash doesn't reload the entire file unless i manually restart it. I've defined the start point and sincedb_path correctly to my knowledge.
That is correct. Setting 'sincedb_path => “NUL”' prevents the in-memory sincedb being persisted across restarts, but the file input still maintains the sincedb and there is no way to stop that.
Can you include the date/time in the filename when you extract tasks from jira?
I was thinking the other way around. If the last line did not have a newline at the end then the file input would continue tailing the file, waiting for the newline to appear so that it could ingest the last line.
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.