<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div style="-webkit-text-size-adjust: auto;"><span></span></div><div><meta http-equiv="content-type" content="text/html; charset=utf-8"><div style="-webkit-text-size-adjust: auto;"><span></span></div><div><meta http-equiv="content-type" content="text/html; charset=utf-8"><div style="-webkit-text-size-adjust: auto;">[thread reconstructed]</div><blockquote type="cite" style="-webkit-text-size-adjust: auto;"><span></span></blockquote><div><blockquote type="cite"><font color="#000000"><span style="-webkit-text-size-adjust: auto; background-color: rgba(255, 255, 255, 0);"></span></font></blockquote><blockquote type="cite"></blockquote><blockquote type="cite"><blockquote type="cite"><font color="#000000"><span style="-webkit-text-size-adjust: auto; background-color: rgba(255, 255, 255, 0);">On 09/30/2014 06:28 AM, AFO-Admin wrote:</span></font></blockquote></blockquote><div><blockquote type="cite" style="-webkit-text-size-adjust: auto;"><blockquote type="cite"><blockquote type="cite">if we would get multithread support, this would boost the bandwith<br>that is avaible, there i'm sure.<br>All my relays run in CPU limit because i don't think wasting even more<br>ipv4 addresses is great, today you get more and more cpu cores that is<br>not linier with the IPC increase.<br><br>E.g. you have a Server with 2x E5-2683 v3 v3 and a 10 Gbit/s pipe you<br>would need atleast 14 IP's to use most of the CPU. And every IP's gets<br>blacklistet, with that much Tor nodes in the same /24 maybe the entire<br>/24. So most ISP's wouldn't be happy with that and with the IPv4<br>shortage this days im also not very happy with that.<br>We really shouldn't waste more IPv4 IP's then needed and the only<br>solution is to change the max amount of Tor Processes from 2 to a<br>higher number or move to IPv6 or get multithreading working.</blockquote></blockquote></blockquote><br></div>[Moritz quoted]<br><blockquote type="cite"><blockquote type="cite"><font color="#000000"><span style="-webkit-text-size-adjust: auto; background-color: rgba(255, 255, 255, 0);"></span></font></blockquote><blockquote type="cite"><blockquote type="cite"><font color="#000000"><span style="-webkit-text-size-adjust: auto; background-color: rgba(255, 255, 255, 0);">E.g. you have a Server with 2x E5-2683 v3 v3 and a 10 Gbit/s pipe you<br></span></font></blockquote><blockquote type="cite"><font color="#000000"><span style="-webkit-text-size-adjust: auto; background-color: rgba(255, 255, 255, 0);">would need atleast 14 IP's to use most of the CPU.</span></font></blockquote></blockquote></blockquote><blockquote type="cite"><br></blockquote></div><blockquote type="cite" style="-webkit-text-size-adjust: auto;"><span>Moritz Bartl transcribed 0.5K bytes:</span><br><blockquote type="cite"><span>Raising the limit from 2 relays per IP to x per IP has been discussed in the</span><br></blockquote><blockquote type="cite"><span>past and would be an easy change.</span></blockquote></blockquote><blockquote type="cite"><br></blockquote><div>Date: Tue, 30 Sep 2014 20:35:21 +0000</div><div><blockquote type="cite"><div style="display: none;"><br></div></blockquote><span></span></div><div><div>From: isis <<a href="mailto:isis@torproject.org">isis@torproject.org</a>></div></div><div><blockquote type="cite"><span>We *still* have that limit? I thought we killed it a long time ago.</span><br></blockquote></div><blockquote type="cite" style="-webkit-text-size-adjust: auto;"><span></span><br><span>Can we kill it now? It's not going to do anything to prevent Sybils, it'll</span><br><span>only prevent good relay operators on larger machines from giving more</span><br><span>bandwidth.</span><br></blockquote><div style="-webkit-text-size-adjust: auto;"><div><br></div><div style="-webkit-text-size-adjust: auto;">Allowing multiple (>2) tor instances would alleviate AFO's issue in the short term, although in this particular case, they might need 14(!) instances (or AES-NI).</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">Given the shortage of IPv4s, and the availability of multi-processor, high-bandwidth servers, we could trial raising the Tor instance limit per IP. (As this is an authority parameter, the change could happen much sooner than predominantly-IPv6 tor or multithreaded tor.)</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">4 would allow 1 tor process per logical processor in many server machines (e.g. 4x1 and 2x2). At ~320Mbps per tor process (the maximum bandwidth in the current network), this could saturate a 1 Gbps link with 1 IP.</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">8 would allow 1 tor process per logical processor in almost all servers (e.g. 4x2 and 8x1), and could saturate a 2.5 Gbps link with 1 IP. In AFO's 10 Gbps case, they'd need 32 processes, or 4 IPs. (Which doesn't seem as unreasonable as 14 IPs.)</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">The only drawback I can see is that IPs with slow connections/few CPUs could then launch 4 or 8 instances, and slow down the network. This could exacerbate the "wasted consensus entry" issue, where the consensus bytes used for a router outweigh its contribution. (But this seems unlikely.)</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">In the short term, can we trial raising the Tor instance limit per IP to 4 or 8?</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">(In the longer term, I'm happy to help with (network) performance, multithreading, or IPv6 - probably in that order.)</div><div style="-webkit-text-size-adjust: auto;"><br></div><div style="-webkit-text-size-adjust: auto;">T</div><div style="-webkit-text-size-adjust: auto;"><br><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><span style="background-color: rgba(255, 255, 255, 0);">teor<br>pgp 0xABFED1AC<br><a href="hkp://pgp.mit.edu/">hkp://pgp.mit.edu/</a></span></div><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><font color="#000000"><span style="background-color: rgba(255, 255, 255, 0);"><a href="https://gist.github.com/teor2345/d033b8ce0a99adbc89c5">https://gist.github.com/teor2345/d033b8ce0a99adbc89c5</a><br><a href="http://0bin.net/paste/Mu92kPyphK0bqmbA#Zvt3gzMrSCAwDN6GKsUk7Q8G-eG+Y+BLpe7wtmU66Mx" x-apple-data-detectors="true" x-apple-data-detectors-type="link" x-apple-data-detectors-result="2">http://0bin.net/paste/Mu92kPyphK0bqmbA#Zvt3gzMrSCAwDN6GKsUk7Q8G-eG+Y+BLpe7wtmU66Mx</a></span></font></div><div><br></div><div><br></div></div><div style="-webkit-text-size-adjust: auto;"><br></div></div></div></div></body></html>