I have decided to just mine the log files created on the Qlik replicate server for each task, bumping logging for performance to "trace" so I can get the latency metrics.
It’s a pretty great monitoring tool. Requires less toil to maintain and is easier to implement than, say, Prometheus + Grafana. Sort of the “it just works” of observability.
It’s a huge **bleep**ing cost though. I’ve worked at places where we migrated to it, saw the recurring bill, and migrated away from it again within the same year.
