duckdb: Memory leak, source unknown

What happens?

We have a Java service that operates an in-memory DuckDB database for long periods of time, creating, loading, and dropping tables all the while. We have observed what we believe to be a memory leak, but have so far been unsuccessful in locating the source of the leak.

To Reproduce

We’ve written a minimal reproduction and shared it as a GitHub repository. The reproduction runs for a specified period of time. During execution it creates and loads tables at a rate of 5 tables per second. It drops tables as necessary to maintain a “live” set of 100 tables. Tables are loaded using the Appender class.

The reproduction code itself doesn’t maintain any notable, long-term state beyond a primary DuckDBConnection object.

Environment (please complete the following information):

• OS: Ubuntu Linux 16.04
• DuckDB Version: 0.3.2
• DuckDB Client: Java

Before Submitting

• Have you tried this on the latest master branch?

• Python: pip install duckdb --upgrade --pre
• R: install.packages("https://github.com/duckdb/duckdb/releases/download/master-builds/duckdb_r_src.tar.gz", repos = NULL)
• Other Platforms: You can find binaries here or compile from source.

• Have you tried the steps to reproduce? Do they include all relevant data and configuration? Does the issue you report still appear there?