Hello Virgil, An lun., juin 30, 2014, Virgil Griffith schrieb: >It's already established that, for clients, onion-pi's are >discouraged---onion-pi wifi doesn't protect enough (I.e., at all) from >browser-based attacks. > >Given that, The question is now, "Are onion-pi's are good enough to be >useful relays?" Roger said no. Is there a more informed opinion on this >matter---particularly from someone who has actually tried this? Are there >any relays that are known to run on onion-pi? > >If an onion-pi is insufficient for a useful Tor relay, what is the limiting >reagent? What more does it need to be useful? > Sorry for the bad answer to your good question, but you might like to know that tor proxy packages exist for even devices with MIPS processors like the 400 MHz MIPS Atheros AR9331 SoC. While I haven't tested just yet (going to soon), this is considerably less powerful than what the Raspberry Pi packs and seems to still work. Also, if you're willing to struggle you could port Onion Pi to the Cubieboard (second HW revision.). If the thing is processor bound, that would just about double your performance (judging that the newer boxes run a dual core ARM Cortex A7 1GHz and 1 Go RAM. Don't know about networking, as these boards integrate Allwinner A2 SoCs. A third (and most powerful of all) option would be to port to the Intel Galileo. That would be cheapest as well as exposing a IA32 instruction set for easy porting of all that Onion Pi depends on. Cheers, Michael
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ tor-dev mailing list tor-dev@xxxxxxxxxxxxxxxxxxxx https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev