parallel write queries causing the node to get stuck

[drumfeet]

I have observed this issue only occurs if I run multiple write queries in parallel.
The node cannot recover and will stop committing transactions to warp.

[drumfeet]

I submitted a test case for this issue. #32