On Sat, Aug 8, 2015 at 1:41 AM, torrry@safe-mail.net wrote:
I ran some tests against your node. While performance is generally very good, it has very low performance connecting to some exit nodes.
Thanks for running the tests. Which exit nodes led to poor performance? I would like to try to reproduce any performance problems.
How would you measure performance between my node and a given exit without being influenced by the properties of the middle relay? You can only set me as an entrynode, and you can't pick a specific middle, so how would you know that the low performance was my node and not the random middle relay?
The problem is likely that your ISP is routing some traffic via an
overloaded peering point.
This is certainly possible! It's the most compelling theory for me at the moment, although I'm not convinced.
Some ISPs have been know to do that on purpose, e.g. to make connections to Netflix or Youtube extremely slow while claiming not to do throttling.
Fortunately those practices are mostly coming to end in the US, with the FCC's "Open Internet" rules adopted earlier this year. There is a way to file complaints about it, but I'd need more data.
The bandwidth auths probably downrate the measurement results of your
server severely because of those slow connections.
Probably? How can we investigate further?
As I write this one of the relays has the Guard flag again and is averaging around 2 Mb/s. That still stinks, but the measured bandwidth shows only 15.0 Kb/s! Something is still ridiculously off. I'll continue digging. I appreciate all the help thus far.