Syncing blocks #18
Description
It would be useful to transfer blocks between systems for eg. backups:
tr1pctl send @.. | ssh foo tr1pctl recv- would transfer the blocks of the current session to the remote system. This might need to be extended to avoid uploading blocks that are already known by the receiving system.- sftp - blocks are uploaded to a remote server. For sandboxing reasons, this should be done in a separate process and probably needs tr1pctl tail -f #8.
- broadcast - tr1pd could be used as a system that announces events to multiple other systems in a secure way.
There's also the usecase to have "merger" nodes that receive multiple inputs and write them to their own ledger. It's important that the information where a block is coming from isn't lost, and that adding this data doesn't exceed a size limit. It might be useful to have a separate field for this. An open question is how blocks should be handled when merged twice, so it would have two "from" fields.