This fixes up commit e3781aff
(Reduce probability of dying when the system lacks memory").

Some servers can only be accessed via their re6st IP. re6st itself uses little
memory so it should not die when it fails to fork.

The way peer addresses were exchanged polluted caches with information about
dead nodes. In particular, bootstrapping often took a long time because the
cache of the primary node was mostly useless.

This also fixes bootstrap of registry.

This is a common misconfiguration that may break internet acces for other peers.

We also stop checking for child process termination when used without tunnel
manager (i.e. with --client or --client-count=0) because it conflicts with the
'ip route' command that is called every minute if --table=0 is used.
Anyway, with a tunnel manager, only openvpn client are watched.

- authenticated communications with registered clients
- XML-RPC is dropped
- multi-threaded server

The previous broadcast model is replaced by a query-response one.
During normal operation, the cache of peers is not used anymore to select
peers to connect to. It now only used for bootstrapping and avoid querying
an already known address.

A lot of bug fix

Removing of the bandwidth calculation

When 2 tunnels are estabilshed between the same 2 peers, one of them is deleted

Check for dead tunnels evry 5 seconds instead of waiting for the next tunnel_refresh

Added exception catching in db, and modified utils.log so tht it can be used with anything ( and not only a string )