Re: [tor-relays] [Tor Weather] Node Down! - AccountingMax and ORPort 80

2014-11-21 Thread Toralf Förster
On 11/21/2014 09:44 AM, Chuck Peters wrote:
 Nov 16 00:00:00.000 [notice] Opening OR listener on 0.0.0.0:80
 Nov 16 00:00:00.000 [warn] Could not bind to 0.0.0.0:80: Permission denied

As stated in [1] you could try something like

$ setcap 'cap_net_bind_service=+ep' /usr/bin/tor


[1] 
http://serverfault.com/questions/112795/how-can-i-run-a-server-on-linux-on-port-80-as-a-normal-user

-- 
Toralf
pgp key: 0076 E94E

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


[tor-relays] OrPort self-testing sometimes fails

2014-11-21 Thread Julien ROBIN
Hi everybody,

Second time it happen to me so I would like to know if you have ideas about it.
Sometimes after a reboot, I haven't the following message into the log :

Self-testing indicates your ORPort is reachable from the outside. Excellent. 
Publishing server descriptor.

And after a while (fortunately), I receive a warning (Tor Weather) because my 
server is shown as offline.

So I just type /etc/init.d/tor stop and /etc/init.d/tor start, and it works 
fine.

I would like to know how does the test works, in order to check what can fail 
during the test.
If my server's IP (95.130.9.190) is not reachable by a part of the Internet for 
example, in order to open a support ticket with my ISP (do you know how I can 
easily check it ?).

Also, since few month now, the bandwith usage seems to keeps far below what I 
had before (even with a good consensus weight, fresh OS install with new keys 
and name a month ago) so I now realize that may be there is a problem somewhere 
with this machine.

Thank you in advance for your help and ideas !

Best Regards,
Julien ROBIN



Please find below the logs of the failed start, followed by the working 
start


Nov 21 11:56:53.000 [notice] Tor 0.2.5.10 (git-43a5f3d91e726291) opening log 
file.
Nov 21 11:56:53.000 [notice] Parsing GEOIP IPv4 file /usr/share/tor/geoip.
Nov 21 11:56:53.000 [notice] Parsing GEOIP IPv6 file /usr/share/tor/geoip6.
Nov 21 11:56:53.000 [notice] Configured to measure statistics. Look for the 
*-stats files that will first be written to the data directory in 24 hours from 
now.
Nov 21 11:56:53.000 [notice] Caching new entry debian-tor for debian-tor
Nov 21 11:56:53.000 [notice] Caching new entry debian-tor for debian-tor
Nov 21 11:56:53.000 [notice] Your Tor server's identity key fingerprint is 
'MagnetronFR1 145AEE9359A8943189500C5451722CAA7EEEC6C3'
Nov 21 11:56:53.000 [notice] Bootstrapped 0%: Starting
Nov 21 11:56:57.000 [notice] We now have enough directory information to build 
circuits.
Nov 21 11:56:57.000 [notice] Bootstrapped 80%: Connecting to the Tor network
Nov 21 11:56:57.000 [notice] Bootstrapped 85%: Finishing handshake with first 
hop
Nov 21 11:56:58.000 [notice] Bootstrapped 90%: Establishing a Tor circuit
Nov 21 11:57:00.000 [notice] Tor has successfully opened a circuit. Looks like 
client functionality is working.
Nov 21 11:57:00.000 [notice] Bootstrapped 100%: Done
Nov 21 13:56:58.000 [notice] Interrupt: we have stopped accepting new 
connections, and will shut down in 30 seconds. Interrupt again to exit now.
Nov 21 13:57:28.000 [notice] Clean shutdown finished. Exiting.


Nov 21 13:57:39.000 [notice] Tor 0.2.5.10 (git-43a5f3d91e726291) opening log 
file.
Nov 21 13:57:39.000 [notice] Parsing GEOIP IPv4 file /usr/share/tor/geoip.
Nov 21 13:57:39.000 [notice] Parsing GEOIP IPv6 file /usr/share/tor/geoip6.
Nov 21 13:57:39.000 [notice] Configured to measure statistics. Look for the 
*-stats files that will first be written to the data directory in 24 hours from 
now.
Nov 21 13:57:39.000 [notice] Caching new entry debian-tor for debian-tor
Nov 21 13:57:39.000 [notice] Caching new entry debian-tor for debian-tor
Nov 21 13:57:39.000 [notice] Your Tor server's identity key fingerprint is 
'MagnetronFR1 145AEE9359A8943189500C5451722CAA7EEEC6C3'
Nov 21 13:57:39.000 [notice] Bootstrapped 0%: Starting
Nov 21 13:57:39.000 [notice] Bootstrapped 5%: Connecting to directory server
Nov 21 13:57:41.000 [notice] We now have enough directory information to build 
circuits.
Nov 21 13:57:41.000 [notice] Bootstrapped 80%: Connecting to the Tor network
Nov 21 13:57:41.000 [notice] Bootstrapped 85%: Finishing handshake with first 
hop
Nov 21 13:57:41.000 [notice] Self-testing indicates your ORPort is reachable 
from the outside. Excellent. Publishing server descriptor.
Nov 21 13:57:43.000 [notice] Bootstrapped 90%: Establishing a Tor circuit
Nov 21 13:57:44.000 [notice] Tor has successfully opened a circuit. Looks like 
client functionality is working.
Nov 21 13:57:44.000 [notice] Bootstrapped 100%: Done
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


[tor-relays] Website no longer available from Tor

2014-11-21 Thread Marcello Vitagliano
Hi
from a few weeks the site spacemarc.it is no longer accessible via
TOR. My hosting does not apply any filter nor I do it from code.

I do not get any error from web server (forbidden, timeout and so
on...) but the Tor Browser (v4.0.1) simply tries to connect for
several seconds but to no avail (try yourself please).

I checked the Apache logs and I do not see any request from an IP Tor.
In your opinion why is not it more accessible?
Thanks
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


Re: [tor-relays] building Tor against LibreSSL 2.1.1 fails with undefined reference to `EVP_aes_128_ctr' error

2014-11-21 Thread David Stainton
I am also very interested in hearing from people who have built tor
with LibreSSL...
specifically I'd love it if someone worked out all the details to do
this as a static build in OpenBSD.


On Fri, Nov 21, 2014 at 3:22 PM, Seth l...@sysfu.com wrote:
 I'm trying to build tor-0.2.5.10 from source against LibreSSL 2.1.1 on a
 FreeBSD 9.3x jail system.

 It fails with this message

 ---

   CC   src/tools/tor-gencert.o
   CCLD src/tools/tor-gencert
 src/common/libor-crypto.a(aes.o): In function `aes_new_cipher':
 /usr/local/src/tor-0.2.5.10/src/common/aes.c:100: undefined reference to
 `EVP_aes_128_ctr'
 *** [src/tools/tor-gencert] Error code 1

 Stop in /usr/local/src/tor-0.2.5.10.
 *** [all] Error code 1

 Stop in /usr/local/src/tor-0.2.5.10.

 --

 Has anyone has any luck building Tor against LibreSSL?

 --
 Seth
 I 3 nicely trimmed email replies
 ___
 tor-relays mailing list
 tor-relays@lists.torproject.org
 https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


Re: [tor-relays] Website no longer available from Tor

2014-11-21 Thread grarpamp
 In your opinion why is not it more accessible?

You asked four times. We can't see your systems
or your exits so we don't know.
Run your own restricted exits on your own various computers,
starting with those you can reach it with via clearnet, then
route all your traffic out those exits and do more testing,
beginning with tcpdump or some form of log/traffic analysis
on both the exit and webserver. There's obviously blocking or
misconfiguration involved. Let us know what you find.

https://www.torproject.org/docs/documentation.html.en
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


Re: [tor-relays] OrPort self-testing sometimes fails

2014-11-21 Thread Ryan Getz
On Fri, Nov 21, 2014, at 09:18 AM, Julien ROBIN wrote:
snip
 I would like to know how does the test works, in order to check what can
 fail during the test.
 If my server's IP (95.130.9.190) is not reachable by a part of the
 Internet for example, in order to open a support ticket with my ISP (do
 you know how I can easily check it ?).
snip

I cannot comment on how the test is performed (although it may be in the
documentation) but if you are looking to test connectivity  to your ipv4
address, you could always use a Looking Glass or route server for  some
basic connectivity tests (pinging / traceroute / etc). If you do
discover an issue, it would be very beneficial to provide traceroute
data (if it's not blocked by a firewall somewhere) in *both* directions
(your server to the looking glass ip and looking glass to your server
ip)...  this helps identify any potential routing issues where
asymmetric routing may be in place.

Just search for Looking Glass or route server and you should find
many available from numerous ISPs and locations that may be helpful in
your testing. Good luck and I hope someone else is able to answer with
details regarding the testing of your relay.

Regards,
Ryan Getz
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


[tor-relays] Fast Exit Node Operators - ISP in US

2014-11-21 Thread SiNA Rabbani
Dear Relay Operators,

I noticed there are very few US based exit nodes in the network. And more and 
more
people are jumping on the same set of AS numbers in Europe.

I am not if the reason is lack of Tor friendly ISPs or people are just too 
freaked out about
the summer of Snowden.

I think it's very wrong to assume that EU countries are not part of the 
world-wide-wiretap, packets are 
going through a few internet exchanges anyways.

I have been hosting/operating Faravahar (one of the authority directory 
servers) at Rethem Hosting (rethemhosting.net)
for a couple of years now, and never had any issues.

I also just brought up 2 exit nodes there:
https://atlas.torproject.org/#details/A5B1C342B316C2AE5695B903CED18F619A8361CC
https://atlas.torproject.org/#details/6FFCDF910C32D620FCC6EEF7A8A57F3E9A234649

If anyone is interested in running fast Tor Exit nodes at Rethem Hosting. Feel 
free to contact me directly,
so I can make proper referral/introductions. Rethem Hosting is also able to 
provide hosting In IceLand, but you 
get the most bang for your buck in the US datacenter.

Thank you for contributing to Tor.

All the best,
SiNA 


#   Consensus Weights   Advertised BandwidthGuard Probability   
Middle Probability  Exit ProbabilityNicknameFingerprint 
ExitGuard   Country Autonomous System
1.5958% 0.% 0.% 0.% 4.8946% (11 other relay groups) 

17.6751%0.% 0.% 0.% 54.2124%(total in selection)

1   4.1722% 0.% 0.% 0.% 12.7968%*   (26 relays) 
(26)(22)FR  (4)
2   3.8256% 0.% 0.% 0.% 11.7338%*   (19 relays) 
(19)(16)DE  (6)
3   2.1098% 0.% 0.% 0.% 6.4712% *   (10 relays) (10)
(8) NL  (6)
4   1.4620% 0.% 0.% 0.% 4.4843% *   (5 relays)  (5) 
(5) RO  (1)
5   0.9090% 0.% 0.% 0.% 2.7881% *   (6 relays)  (6) 
(5) SE  (4)
6   0.8788% 0.% 0.% 0.% 2.6955% *   (4 relays)  (4) 
(4) CH  (2)
7   0.8560% 0.% 0.% 0.% 2.6255% *   (3 relays)  (3) 
(3) LU  (1)
8   0.6510% 0.% 0.% 0.% 1.9967% *   (1 relays)  (1) 
(1) LR  (1)
9   0.6433% 0.% 0.% 0.% 1.9731% *   (10 relays) (10)
(8) US  (10)
10  0.5715% 0.% 0.% 0.% 1.7528% *   (2 relays)  (2) 
(2) GB  (1)
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


Re: [tor-relays] Fast Exit Node Operators - ISP in US

2014-11-21 Thread Moritz Bartl
Hi SiNA,

On 11/22/2014 01:08 AM, SiNA Rabbani wrote:
 Dear Relay Operators,
 
 I noticed there are very few US based exit nodes in the network. And more and 
 more
 people are jumping on the same set of AS numbers in Europe.
[...]

Thank you, SiNA. A reminder to relay operators: Diversity is important.
A very good paper everyone should read is Traffic Correlation on Tor by
Realistic Adversaries [1].

Compass [2] is very useful in at least determining country- and AS-level
diversity. It would be nice to have more than just a feeling of when to
rule out a potential ISP and/or country, but I would at least try to
avoid any of the popular AS.

lowendbox.com is not a bad source for virtual machine hosting. If you
plan to run an exit relay, it is imperative that you ask the ISP
beforehand, and you should read the Exit Guidelines [3]. Add the answer
of the ISP to the GoodBadRelays wiki page [4]. For non-exit relays, I
wouldn't ask or tell the ISP, they don't have to know. When you pick a
cheap provider with unlimited (fair use) bandwidth, make sure you
contact the ISP beforehand to find out how much constant traffic they
are actually ok with, and configure your relay accordingly. The
hibernation options are quite useful in that regard.

For larger exits (dedicated, higher bandwidth), webhostingtalk.com can
be a good source. It is generally cheaper to pool money and rent a
bigger server. Ideally, you find some people around you. For example, if
you have a local hackerspace or makerspace nearby, you should leave
contact info and ask if there's interest to collectively run a larger
relay. I always wanted to get Tor User  Relay Operator Groups going.
A quite outdated and lame attempt is a wiki page on the torservers wiki [5].

The next step may be to set up an organization around your exit(s). Many
groups chose the non-profit model [6]. This type of organization is
surprisingly easy to create and manage, but it does produce overhead.
Think a bit about who wants to play accountant and all that.

After a while, you might consider joining the Torservers.net
reimbursement partnership. While the program does not formally require
you to have an organization, we do prefer them, simply because they are
a sign of a more stable environment. For more information, see [7].

[1] http://freehaven.net/anonbib/#ccs2013-usersrouted
[2] https://compass.torproject.org/
[3] https://trac.torproject.org/projects/tor/wiki/doc/TorExitGuidelines
[4] https://trac.torproject.org/projects/tor/wiki/doc/GoodBadISPs
[5] https://www.torservers.net/wiki/usergroups
[6] https://www.torservers.net/partners.html
[7]
https://blog.torservers.net/20130917/reimbursement-for-exit-operators.html

-- 
Moritz Bartl
https://www.torservers.net/
___
tor-relays mailing list
tor-relays@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays


Re: [tor-relays] Website no longer available from Tor

2014-11-21 Thread Andy Isaacson
On Fri, Nov 21, 2014 at 12:50:39PM -0500, grarpamp wrote:
  In your opinion why is not it more accessible?
 
 You asked four times. We can't see your systems
 or your exits so we don't know.

Indeed!  You can increase the Tor client debugging level on the machine
you're trying to TBB from, find out what exit you used to try to
connect, and then try tcptraceroute and other debugging tools from your
server to see if you can debug the connectivity failure.

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


Re: [tor-relays] building Tor against LibreSSL 2.1.1 fails with undefined reference to `EVP_aes_128_ctr' error

2014-11-21 Thread Seth
On Fri, 21 Nov 2014 09:10:11 -0800, David Stainton  
dstainton...@gmail.com wrote:



I am also very interested in hearing from people who have built tor
with LibreSSL...


If you want to try building a FreeBSD port using LibreSSL instead of  
OpenSSL add this to /etc/make.conf


OPENSSL_PORT=security/libressl
WITH_OPENSSL_PORT=yes


specifically I'd love it if someone worked out all the details to do
this as a static build in OpenBSD.


Not sure about static builds, what's the benefit?

I do know OpenBSD 5.6 has LibreSSL baked in and it works with Tor. Just  
install the tor package, edit /etc/tor/torrc and you're up and running.


Next time I stand up another relay or exit node on OpenBSD I think I'll  
kick it up a notch with some chroot and/or systrace sauce.


https://trac.torproject.org/projects/tor/wiki/doc/OperationalSecurity#RunTorandOtherServicesinaRestrictedEnvironment

Am also interested in hearing any tips for minimizing data retention. I  
thought about making a hardlink or symlink from /var/log to /dev/null, but  
I have a feeling there's more to it than that.

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