We have a shared aggregation cluster running Logstash 5.6 that has persistent queues enabled, but as far as we can tell, this resource is shared between all configured outputs. Is there a way for us to configure our outputs in a way to quota how much of the persistent queue that they consume?
i.e. If a destination ES cluster fills up on space, the queue will fill up, and eventually start affecting our S3 export outputs as well. I'd like to restrict the ES export output to 1/2 of the persistent queue as to prevent a scenario where the S3 export is not blocked.