[tor-talk] Multiple CPU cores?

[nusenu]

(this thread might be better suited for tor-relays@)

> I've over some time ran some different tor relays - and I really like the
> Tor project. It really makes sense to me.

Thank you for running exit relays!

> Here is the problem I've ran into. On my exit node, speedybacon2500, I have
> 2.5 Gbit/s interface and 8 CPU cores. Tor uses ~60 mbps and 1 CPU core. I
> do also have a few other things running on the server, but I'd really like
> to be able to take advantage of more CPU cores and more bandwidth.

As others have pointed out already the current best solution is to run
multiple tor instances (2 per IPv4).

Also please consider setting the MyFamily parameter in your torrc with
your relay fingerprints so tor users do not choose your relays for more
than one position in a circuit. This is especially relevant in your case
since you apparently run exits and non-exits.

https://atlas.torproject.org/#search/contact:rathhansen

After setting MyFamily you can check this page (if you are no longer
listed there you fixed it properly)
https://raw.githubusercontent.com/ornetstats/stats/master/o/potentially_dangerous_relaygroups.txt
(regenerated daily)


If you are lazy (you should be :) you might want to automate all that
(creating multiple tor instances, setting MyFamily, ...), here is an
ansible role to do just that (including OfflineMasterKeys):

https://github.com/nusenu/ansible-relayor



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: OpenPGP digital signature
URL: <http://lists.torproject.org/pipermail/tor-talk/attachments/20160627/03cce2e4/attachment.sig>