I'm trying to use multiple config files located in /etc/filebeat. The default config is stored as filebeat.yml, with additional configs stored as $service.yml. I've verified that all the configs are working and functional by using the "-c config.yml -e -d "*"" switches against each .yml. When starting filebeat, only one prospector and accompanying paths from filebeat.yml are active. I'm using filebeat version 5.0.0-rc1.
I'm trying to use the same approach with packetbeat and metricbeat, whereby system defaults are located in *beat.yml, and additional services are defined in $service.yml via configuration management (ansible).
This does help indeed, thanks! The docs that I referenced previously did not mention the config_dir option, and eluded to the -path.config switch instead.
Regarding the other beats, how does one take a modular approach with the configs?
For the other beats currently the best approach is using the chaining of config files with -c. Or as you alreasy use ansible it should be possible to use ansible to generate one big config file.
What docs were you referencing to initially where the config_dir was not mentioned?
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.