We are trying to deploy Filebeat to a steaming pile of Beanstalk instances. In our case, different instances have different applications, and perhaps different logs to monitor. The process for automating a per-app custom filebeat config file is painful with beanstalk. We were considering whether if might make sense to just list all the possible config files in a single filebeat config. Would filebeat just shrug off the missing files, or would it complain continuously?
Filebeat does not complain if it cannot find a file. However, in case the file is present on the filesystem but it cannot be accessed, FB logs error messages every time it tries to read from it.