- Start jobs according to the start time, jobs are now started when they are received - Add DB - Store jobs in DB until start time? - Handle failures in the underlying system (i.e. bitcoind crashes) - Add checks related with OP_CSV in justice tx and dispute_delta provided once the blob is decrypted - Do not accept new appointments if the locator has already been used - Check all the interactions with core, figure out the edge cases and error codes i.e: The justice transaction can already be in the blockchain the first time we push it - Handle reconnection with ZMQ in case of broken pipe. The current version of the code fails if it does happen