I think the issue here is that if fast initialization crashes, upon reboot, the service will assume everything is great and start processing from tail, even though fast processing hadn't reached the genesis block yet.
A fix would to have an explicit flag in DB indicating fast init state.
I think the issue here is that if fast initialization crashes, upon reboot, the service will assume everything is great and start processing from tail, even though fast processing hadn't reached the genesis block yet.
A fix would to have an explicit flag in DB indicating fast init state.