Hi,
I'd like to use Persistent Queues (PQ) as an internal overflow buffer when we get spikes of logs into our redis instance (input redis {} in each pipeline). But the PQ does not grow despite our redis instance being saturated with logs. There is a key in redis for each pipeline. Redis keys are updated by filebeat clients (7.5.x)
How to debug to find the bottleneck?
Environment: Ubuntu + ELK stack : 7.9.0 + pipelines
Note: PQ used to buffer fine when we used ELK 6.6.8 and a single logstash.conf (no pipelines).