1. 13
    1. 6

      That is a delightful writeup!

      I will caution folks that believe in the bAtTlE-tEsTeD pOwEr AnD sTaBiLiTy of the Erlang Virtual Machine that you can still blow-up a perfectly good prod server if your supervisors aren’t setup just so, and can directly cite a time where we had to compensate for this by wrapping everything in a systemd unit to keep popping it back up until we could properly solve the real problem. Even BEAM instances can use supervision trees. :)

    2. 4

      It achieved that by merging D-Bus into the systemd itself, and then making all service to be D-Bus daemons (which are started on request), and additionally it provides a bunch of triggers for that daemons.

      Erm, come again? This is either a terrible characterization or a gross misunderstanding of what a systemd unit’s relationship is with its underlying daemon.

      1. 4

        I have removed that part as it was confusing leftover from the early drafts.