I think that a network based to much on remotes VMs, with closed source software running on the most deep machine level, is not very resilient and secure.
So the reason why I was thinking to do so is that I wanted to run a small exit relay on a device running only open source software, like Olimex Lime2 does, and under my direct control.
The latency from my home and the VM is not so high (45-50 ms), and I was pretty sure that with a proper configuration I didn't risk that users exit through my home connection. But If you say that with a so small bandwidth It can't run properly, I trust you, so I keep a non-exit relay.
Anyway thanks for your advicestor-relay@riseup.net:I'm running a non exit relay on a debian machine (in the next few months I will switch to *BSD) on a Lime2.I assume you are referring to a relay run at home.I'm running an exit relay too on a remote VM. I would turn my non-exit relay in an exit one, but for obvious reasons, I don't want to run It from my shitty ISP IP. I could give 10-14 mbps from my home connection, so I think that the lime2 would be powerful enough to run It properly.I would discourage such a setup for the following reasons: - this setup includes the risk that users will exit through your home broadband IP address (bad!) if tunnels break down - such setups that introduce an additional hop decrease the user-experience - most users will not be happy with an "10-14mbps" exit at a home broadband connection - it is not clear to me why you would involve your home IP at all for your exit if you have a VM in a datacenter nonetheless, thanks for running relays, nusenu
_______________________________________________ tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays