Logstash consumes file descriptors for multiple reasons, and on larger deployments the default limits set by Operating Systems may not be enough and create production outages.
There should be documentation guiding users on how to size the max number of file descriptors. A source of inspiration is the documentation for Elasticsearch, https://www.elastic.co/docs/deploy-manage/deploy/self-managed/file-descriptors.
Logstash should have a similar documentation section in its docs.