When a worker goes down, there is some form of leak. Neighbours don't get cleaned up, threads may keep running when they shouldn't.
The issue is only apparent when using a restarting node pool; standard workers aren't restarted and don't show an issue (at least not in a noticeable way).
When a worker goes down, there is some form of leak. Neighbours don't get cleaned up, threads may keep running when they shouldn't.
The issue is only apparent when using a restarting node pool; standard workers aren't restarted and don't show an issue (at least not in a noticeable way).