| Age | Commit message (Collapse) | Author |
|---|
|
significantly simpler way to move all of this logic entirely into the containing service, liberating the core from any concern over the nature of its pipe to the outside world.
|
|
|
|
avoid flooding.
|
|
|
|
run in the foreground as some platforms may not require threads at all.
|
|
direct links.
|
|
|
|
|
|
link, porting over core loop code from old Node.cpp to new CAPI version, etc.
|
|
two-way handshake to prevent half-connects.
|
|
scope.
|
|
|
|
|
|
|
|
simplifies a ton of stuff. Loads of spaghetti logic can die since we no longer have to make these decisions down in the core.
|
|
|
|
|
|
possible fix for "NAT traversal coma" issue. Also fix a typo.
|
|
|
|
since identities are what we really want to cache.
|
|
some more legacy multicast fixes.
|
|
frames in new way.
|
|
|
|
|
|
little more consistent about using _ to denote private member variables.
|
|
anti-symmetric-NAT tactics.
|
|
for making supernode topology hot-updatable.
|
|
connectivity and use this for multicast propagation. Also consolidate adding of active bridges via the same functor as regular multicast next hops.
|
|
|
|
|
|
|
|
since doing so can result in asymmetric failed NAT-t over UDP if one side has a firewall that permits outgoing UDP but not incoming.
|
|
|
|
UDP preference in send().
|
|
VERB_RENDEZVOUS computation.
|
|
|
|
|
|
pretty well now!
|
|
|
|
|
|
supernodes from resynchronizing unless explicitly ordered.
|
|
and there is no point in mallocing the TCP write buffer.
|
|
indeed learn incoming TCP paths. Otherwise the recipient of a TCP connection does not know to reply via TCP! Heh.
|
|
|
|
|
|
|
|
|
|
|
|
all paths are equally dead.
|
|
|
