@Toyoo is right. This isn’t for customers. There was already a bug in server side copy that caused a race condition, resulting in bloom filters being sent out that didn’t contain new pieces, making the nodes delete them incorrectly. In scenarios like that, Storj Labs has 7 days to become aware of the issue, identify the pieces that need to be restored and write the code to restore them. 7 days is not a lot of time to do that. I don’t see a way to lower that and not risk data loss in such scenarios.