We've misused sqlite3_column_name. The SQLite documentation states that the following:

The returned string pointer is valid until either the prepared statement is destroyed by sqlite3_finalize() or until the statement is automatically reprepared by the first call to sqlite3_step() for a particular run or until the next call to sqlite3_column_name() or sqlite3_column_name16() on the same column.

As part of our query_by_name infrastructure we've first received all field names for the prepared statement and stored them as string slices for later use. After that we called sqlite3_step() for the first time, which invalids the pointer and therefore the stored string slice.

If a tokio::sync::oneshot channel is closed (via the oneshot::Receiver::close method), a data race may occur if the oneshot::Sender::send method is called while the corresponding oneshot::Receiver is awaited or calling try_recv.

When these methods are called concurrently on a closed channel, the two halves of the channel can concurrently access a shared memory location, resulting in a data race. This has been observed to cause memory corruption.

Note that the race only occurs when both halves of the channel are used after the Receiver half has called close. Code where close is not used, or where the Receiver is not awaited and try_recv is not called after calling close, is not affected.

See tokio#4225 for more details.