Project

General

Profile

Bug #12239

Some users have no circuit view in Tor Browser

Added by anonym over 2 years ago. Updated 8 months ago.

Status:
Resolved
Priority:
Normal
Assignee:
Category:
-
Target version:
Start date:
02/15/2017
Due date:
% Done:

0%

Feature Branch:
Type of work:
Research
Blueprint:
Starter:
Affected tool:
Browser

Description

Supposedly this is even without persistence enabled (and in particular without custom persistence enabled for ~/.tor-browser) and without the involvement of IUKs.

tor-controlport-filter.log View (156 Bytes) emmapeel, 02/17/2017 09:07 AM

History

#1 Updated by anonym over 2 years ago

How to debug:

  • Are you running from DVD, USB, SD card? Also:
    • Please try another Tails medium on the same computer.
    • Please try the same Tails medium on another computer.
  • What is the output of: cat /lib/live/mount/medium/live/Tails.module
  • In Tor Browser's about:config, make sure that extensions.torbutton.display_circuit is set to true.
  • Does this fix it?
    sudo systemctl stop tor-controlport-filter
    sudo sed -i 's/9052/9051/' /etc/tor/torrc
    echo "CookieAuthentication 0" | sudo tee -a /etc/tor/torrc
    sudo systemctl restart tor@default
    
    And restart Tor Browser.
  • If not:
    sudo sed -i 's/9051/9052/' /etc/tor/torrc
    sudo systemctl restart tor@default
    /usr/local/lib/tor-controlport-filter --debug > tor-controlport-filter.log
    
    And restart Tor Browser, verify that the circuit view doesn't work again, Ctrl+C in the terminal to kill tor-controlport-filter, and attach tor-controlport-filter.log to this ticket.

#2 Updated by anonym over 2 years ago

  • Assignee changed from anonym to emmapeel
  • QA Check set to Info Needed

Can you please help me debug this, since you are affected?

#3 Updated by emmapeel over 2 years ago

anonym wrote:

How to debug:

  • Are you running from DVD, USB, SD card? Also:

From USB - not possible right now to test on SD card I'm afraid.

  • What is the output of: cat /lib/live/mount/medium/live/Tails.module
filesystem.squashfs
  • In Tor Browser's about:config, make sure that extensions.torbutton.display_circuit is set to true.

It is set to true.

Does this fix it?

Nope.

I attach the tor-controlport-filter.log, not much on it, but on the console, while running `/usr/local/lib/tor-controlport-filter --debug > tor-controlport-filter.log` I got this:

Tor control port filter started, listening on localhost:9051
/usr/local/lib/tor-browser/firefox (pid: 15407, user: amnesia, port: 40534, filter: tor-browser) connected: loaded filter: tor-browser
Final rules:
commands:
  GETCONF:
  - {pattern: (bridge)}
  GETINFO:
  - {pattern: circuit-status}
  - {pattern: 'ns/id/[a-fA-F0-9]+'}
  - {pattern: ip-to-country/\d+\.\d+\.\d+\.\d+}
  SIGNAL:
  - {pattern: NEWNYM}
events:
  STREAM: {}
restrict-stream-events: true

/usr/local/lib/tor-browser/firefox (pid: 15407, user: amnesia, port: 40534, filter: tor-browser) disconnected: client quit
Traceback (most recent call last):
  File "/usr/lib/python3.5/socketserver.py", line 625, in process_request_thread
    self.finish_request(request, client_address)
  File "/usr/lib/python3.5/socketserver.py", line 354, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/usr/lib/python3.5/socketserver.py", line 681, in __init__
    self.handle()
  File "/usr/local/lib/tor-controlport-filter", line 618, in handle
    self.controller = self.connect_to_real_control_port()
  File "/usr/local/lib/tor-controlport-filter", line 554, in connect_to_real_control_port
    with open(global_args.control_cookie_path, "rb") as f:
PermissionError: [Errno 13] Permission denied: '/var/run/tor/control.authcookie'

#4 Updated by emmapeel over 2 years ago

  • Assignee changed from emmapeel to anonym

#6 Updated by anonym over 2 years ago

  • Assignee changed from anonym to emmapeel

Sorry, I messed up a bit in the above shell commands. Let's try this instead (as root):

# make sure Tor Browser is not running
sed -i 's,^ExecStart=.*,\0 --debug,' /lib/systemd/system/tor-controlport-filter.service 
systemctl daemon-reload
systemctl restart tor-controlport-filter
# start Tor Browser, let a page load to 100% and make sure the circuit view is not working
journalctl -u tor-controlport-filter > tor-controlport-filter.log

I guess it's safest if you email me that log, otherwise you might leak your bridge (or persistent entry guard) if you use one.

Hm. If you use a bridge (or persistent entry guard) can you please try without?

#7 Updated by emmapeel over 2 years ago

I didn't used bridges that time... but I have to configure an HTTPS proxy to connect to the Internet.

I can try without it later.

#8 Updated by emmapeel over 2 years ago

  • Assignee changed from emmapeel to anonym

anonym wrote:

Sorry, I messed up a bit in the above shell commands. Let's try this instead (as root):
[...]
  1. start Tor Browser, let a page load to 100% and make sure the circuit view is not working

It works now! (it wasnt working when i started doing this commands)

#9 Updated by anonym over 2 years ago

  • Assignee changed from anonym to emmapeel

emmapeel wrote:

anonym wrote:

Sorry, I messed up a bit in the above shell commands. Let's try this instead (as root):
[...]
  1. start Tor Browser, let a page load to 100% and make sure the circuit view is not working

It works now! (it wasnt working when i started doing this commands)

Ok, please test this then:
  1. boot into a fresh Tails sedssion
  2. start Tor Browser, let a page load to 100% and make sure the circuit view is not working
  3. close Tor Browser is not running
  4. sudo systemctl restart tor-controlport-filter
  5. start Tor Browser, let a page load to 100% and look at the circuit view again
Hm. And also:
  1. boot into a fresh Tails sedssion
  2. sudo systemctl restart tor-controlport-filter
  3. start Tor Browser, let a page load to 100% and look at the circuit view again

(Yes that's the same as the above without steps 2 and 3.)

I'm trying to figure out whether simply restarting tor-controlport-filter is enough. If --debug is needed it's a bit stranger, but might be indicative of some race/timing issue (i.e. the --debug printing's added delays fix the timing issue).

#10 Updated by anonym over 2 years ago

  • Target version changed from Tails_2.11 to Tails_2.12

#11 Updated by emmapeel over 2 years ago

I haven't tested 2.10 and 2.11 with this. I will do.

Incidentaly, I haven't found this problem in Tails 3.0beta[1,2,3].

#12 Updated by intrigeri over 2 years ago

  • Status changed from Confirmed to Resolved
  • Target version changed from Tails_2.12 to Tails_3.0

Incidentaly, I haven't found this problem in Tails 3.0beta[1,2,3].

OK, closing with the appropriate target version then.

#13 Updated by mercedes508 8 months ago

bfd20fbec601b9cc1470db1096eece3

Also available in: Atom PDF