Re: [tor-bugs] #12585 [Tor]: Implement new option SocksSocket

["Tor Bug Tracker & Wiki" <blackhole@xxxxxxxxxxxxxx>]

#12585: Implement new option SocksSocket
-----------------------------+-----------------
     Reporter:  ioerror      |      Owner:
         Type:  enhancement  |     Status:  new
     Priority:  normal       |  Milestone:
    Component:  Tor          |    Version:
   Resolution:               |   Keywords:
Actual Points:               |  Parent ID:
       Points:               |
-----------------------------+-----------------

Comment (by ioerror):

 Replying to [comment:3 dgoulet]:
 > The "is_socks_socket" is added to an entry_connection_t object which is
 then used in src/or/relay.c +1325 to bypass the check to the IPv4/IPV6
 traffic OK variable.

 I think this is fine - though perhaps I'm mistaken - is there a
 preference? Should I signal that one or the other is preferred?

 > However, it seems that SocksSocket is parsed as a SocksPort option which
 means that the entry connection created is set with the possible flag that
 can be passed to the SocksPort option like "IPv6Traffic" or "CacheDNS",
 etc...

 True.

 >
 > I see that the man page entry you've added does not mention "flags" or
 "isolation flags" like SocksPort does but they are still set with the
 default value. Is this the intended behavior to have one single possible
 way to configure that Unix port and don't offer to the user the
 possibility to deny IPv6 traffic through it for instance like a normal
 SocksPort?

 Yes. That seems like a feature to add later - I just wanted a basic thing
 landed before I try to make it fancy. This is similar to how SOCKSPort was
 developed. :)

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/12585#comment:4>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs