Reread routing table when a established tunnel breaks
authorJulien Muchembled <jm@nexedi.com>
Mon, 29 Dec 2014 21:14:55 +0000 (22:14 +0100)
committerJulien Muchembled <jm@nexedi.com>
Tue, 30 Dec 2014 15:26:03 +0000 (16:26 +0100)
commit4536d8eb83ee2e2bb93e88f158fe0ebea3629fd8
tree78cc8fbba2e41c9171c6b0ac81d3f1df0c261d2d
parent4ea3e7be0ebec59dfa016fabbfc3c29c9644f67f
Reread routing table when a established tunnel breaks

The main reason is to speed up recovery from temporary network cut:
- by not wasting time trying remaining distant peers that were collected during
  the last read of the routing table.
- by not blacklisting good peers, which would happen if too many of them were
  retried before network is back
re6st/tunnel.py