(OTS) Race Condition in Parent Block Update

[l-monninger]

Summary

The parent_block update code here has a race condition if two threads both read the same parent_block together and only then both update the parent_block. The issue is caused by the read lock being released before the write, creating a small window where other thread might jump-in. This is currently impossible though since the sequencer only runs on a single thread, but it needs to be fixed if the single thread assumption might ever change.

[l-monninger]

@khokho