How can I diagnose intermittent connection pool exhaustion in a .NET app using ADO.NET with SQL Server on Glitch?

The goal is to identify why a connection pool intermittently runs out of available connections in an ADO.NET‑based application hosted on Glitch, while using a specific pooling feature such as Max Pool Size. The application exhibits sporadic time‑outs and exceptions despite seemingly normal traffic, making it unclear whether the exhaustion stems from unreturned connections, a configuration limit, or pool fragmentation. Monitoring tools show fluctuating pool usage, but it is uncertain which metrics or events definitively signal a leak versus a hard limit being reached. What performance counters or ADO.NET events indicate that the pool is exhausted due to a leak versus simply hitting the Max Pool Size threshold? How can I differentiate between a connection that is not being returned to the pool and a legitimate queue wait when the pool size limit is active? Which ADO.NET diagnostics (e.g., ClearPool, Pooling=false, or specific connection string keywords) help isolate the root cause without requiring a process restart?