Thank you Tim for your response. I am running Tor 0.2.9.11 on ubuntu linux. I have not implemented a caching DNS at this point and may not do so. I'm not sure how this problem progressed but I notices that arm was slower and more difficult to call up in recent weeks. This is not an old exit. It's been running about 10 weeks. A recent back down of bandwidth did little to resolve the problem.
I get the sense that Tor will not use the second CPU since I tried to reset the server recently. nothing else on the VPS uses more that 0.8% of CPU.
Unlikely that it will help but the fingerprint of this relay is
B06F093A3D4DFAD3E923F4F28A74901BD4F74EB1 Torexitmoldova. Thanks so much for your interest. Larry
On 30 Jun 2017, at 04:52, Arisbe <arisbe@cni.net> wrote: I have a small Tor exit relay (VPS) running at +90% CPU capacity.What version of Tor? Has this only happened recently? Did you upgrade your Tor version? Are you running a caching DNS server?Memory usage is negligible. I am unable to launch ARM as the single CPU cannot process both Tor and arm. Traffic is mostly high for a small node. I have CPUsNum set at 2 in my torrc from its original setup.Have you tried removing this from your torrc?However, pidstat 5 -p <pid> shows that only 1 CPU is employed. lscpu shows that the VPS is allotted 2 cpus.2 cores, or hyperthreading on the same core? Do you have any CPU masking options set? Tor uses threads, rather than using multiple processes.CPU speed is around 2.44 M. I'm hoping someone has some experience with this issue and can give me advice.This is what I'd expect for a relay. Cryptography is expensive. T -- Tim Wilson-Brown (teor) teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org ------------------------------------------------------------------------
_______________________________________________ tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays