The announcements from other peers can be treated as discovery of those
peers, even if they happen outside of the DHT loop. This is especially
relevant given the timing of the loop iteration (5-10 minutes).
Another motivation could be duplicate detection logic. To prevent
duplicate peers from reconnecting either server or client could mark the
peer with info.reconnect(false) in hyperswarm. For the discovery this
means that only the side with info.reconnect(true) can re-establish
the connection. Detecting re-appearance of the remote peer eagerly
should speed up the connection and avoid long wait times.
The announcements from other peers can be treated as discovery of those peers, even if they happen outside of the DHT loop. This is especially relevant given the timing of the loop iteration (5-10 minutes).
Another motivation could be duplicate detection logic. To prevent duplicate peers from reconnecting either server or client could mark the peer with
info.reconnect(false)
in hyperswarm. For the discovery this means that only the side withinfo.reconnect(true)
can re-establish the connection. Detecting re-appearance of the remote peer eagerly should speed up the connection and avoid long wait times.