Tracking All MongoDB Read Operations Without Performance Impact (Free Solution)

I'm trying to implement a system to track all read-only operations on our MongoDB deployment with the following constraints: Requirements: - Need to track all read operations (find, aggregate, etc.) hitting our MongoDB server - Solution must be free (no MongoDB Enterprise/Atlas features) - Minimal performance impact on production workload - Complete coverage of operations from all clients/applications What I've already tried: 1. **system.profiler**: Too much performance impact when tracking all operations 2. **currentOp()**: Missing many read operations due to their short lifecycle 3. **MongoDB logs**: Difficult to parse efficiently and adds I/O overhead 4. **Change streams**: Don't capture read-only operations 5. **Driver command monitoring**: Only captures operations from the specific application instance where it's implemented Has anyone successfully implemented a solution for comprehensive read operation tracking in a similar environment? What approach would you recommend given my constraints?