YT synch has infrastructure which has many external dependencies, like the YT API or QN or Joystream chain, each of which can halt at any time, and also the core infra itself can halt. We need to examine how it recovers from such halts, whether manual or not, and make sure no state is corrupted, and sufficient information is recorded to be able to continue where things left of. Its not obvious, to me anyway, how exactly one would test this, so making a plan sounds like a reasonable first step.

┆Issue is synchronized with this Asana task by Unito