Hi,
are you useing cat7 cable?
Did you configure a DNS fall-back 8.8.8.8 to be on the safe side?

Ok, with (1000 mBit x2=) 2000mBit you are handling 114,3 MB/s

16,8
+2,7
+9,6
+19,7
+26,2
+7,0
+22,6
+9,7
= 114,3 MB/s

Thats already good. Did you declare a cut-off?
You can experimentally try my follwing advice:
For each Tor process set in torrc:
BandwidthRate 13000000 bytes
BandwidthBurst 13375000 bytes

So you end up with
Server1: 4x 13000000 bytes = 4x13MB/s
Server2: 4x 13000000 bytes = 4x13MB/s
(in each direction)
104+104=208 MB/s
You hopefully will end up with ~540TB per month.

(oh, don't miss your upgrade to 0.2.7.6)
(-:


 
Am Sonntag, 13. Dezember 2015 21:18 schrieb Dirk Eschbach <tor-relay.dirk@o.banes.ch>:
 
Hello all,

we are administrating 2 Tor Exits for privacyfoundation.ch [1].

In the last month we realized that the traffic which is going through
our servers / throughput is
by far less than the machines and the providers network can handle.
We were internally and recently with Moritz from torservers.net looking
at this problem but could
not find a way to improve throughput.
Even on the same machine the throughput between servers varies a lot
with out any visible reason.

I hope you can help.

Our setup is two servers with a maximum of 4 tor processes (exit) each.
Looking at CPU, Disk, RAM and so on the machines are not busy at all.

Konsole output
top - 09:57:49 up 16 days, 11:57,  1 user,  load average: 1.14, 0.91, 0.81
Tasks:244 total,  3 running,241 sleeping,  0 stopped,  0 zombie
%Cpu0  :12.5 us, 3.4 sy, 0.0 ni,79.7 id, 0.0 wa, 0.0 hi, 4.4 si, 0.0 st
%Cpu1  :15.9 us, 5.4 sy, 0.0 ni,74.3 id, 0.3 wa, 0.0 hi, 4.1 si, 0.0 st
%Cpu2  :14.3 us, 2.7 sy, 0.0 ni,77.0 id, 0.0 wa, 0.0 hi, 6.0 si, 0.0 st
%Cpu3  : 9.5 us, 3.4 sy, 0.0 ni,80.3 id, 0.0 wa, 0.0 hi, 6.8 si, 0.0 st
KiB Mem:   3877624 total, 2739880 used, 1137744 free,    1288 buffers
KiB Swap: 4026364 total,  364264 used, 3662100 free.   10752 cached Mem

Looking at the network connection it is without any problem possible to
start big downloads without reducing TOR throughput.
The servers are connected with 1 Gbit/s each.

The big question now is:
Why do the machines do not have more throughput ?
Is the reason for this the way the distribution through the Tor network
works.
Moritz hinted it might have to do with the way the tor "bandwidth
scanners" measure the ability of a server to handle traffic.

Can you explain me / point me to documentation where this process is
described and how this can be optimized.
What are the criteria for tor exit node server traffic distribution ?
How do the clients choose the exit ?

I would be very happy if you could provide some answers / documentation
which will set me on the right track.

Thanks very much for you help.

Dirk

[1] https://globe.torproject.org/#/search/query=digiges


_______________________________________________
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays