Bug #7903

Document that one shall restart Unsafe Browser after reconnecting to a different network

Added by emmapeel 2014-09-16 03:30:32 . Updated 2018-03-14 11:06:27 .

Status:
Resolved
Priority:
Low
Assignee:
Category:
Target version:
Start date:
2014-09-16
Due date:
% Done:

100%

Feature Branch:
feature/8243-meek
Type of work:
End-user documentation
Blueprint:

Starter:
Affected tool:
Unsafe Browser
Deliverable for:

Description

reported to WhisperBack:

Exact steps to reproduce the error
—————————————————

1. Connect to a network with a given resolver.
2. Start the Unsafe Browser.
3. Connect to a different network with another resolver.
4. Load a page in the Unsafe Browser.

Actual result and description of the error
—————————————————————

The Unsafe Browser will not be able to load any pages because /etc/resolv.conf
is stuck with the address of the resolver of the first network. Is not fixed when restarting the Unsafe browser neither.

I think that this can be a very annoying usability issue. Let’s say that
you are in an airport and want to try out several captive portals. You should restart Tails every time?


Subtasks


History

#1 Updated by intrigeri 2014-09-16 10:16:13

> Is not fixed when restarting the Unsafe browser neither.

This is very surprising to me, and may indicate a more severe bug.
How exactly can you reproduce this?

> I think that this can be a very annoying usability issue.

Agreed… in corner cases.

> Let’s say that you are in an airport and want to try out several captive portals. You should restart Tails every time?

I think that having to restart the Unsafe Browser only would be acceptable. Having to restart Tails entirely is really problematic, indeed.

#2 Updated by intrigeri 2014-09-16 10:16:44

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

#3 Updated by emmapeel 2014-09-17 13:54:00

I left some people without Internet for a while, and could reproduce the bug.

Restarting the Unsafe Browser doesn’t help.

#4 Updated by intrigeri 2014-09-17 14:25:08

emmapeel wrote:
> Restarting the Unsafe Browser doesn’t help.

Again, how exactly can I reproduce this?

#5 Updated by emmapeel 2014-09-18 00:41:12

Well, if you have two different networks available with internal nameservers you can reproduce it with:

1. Connect to a network with a given resolver.
2. Start the Unsafe Browser.
3. Connect to a different network with another resolver.
4. Load a page in the Unsafe Browser. Not found
5. Restart Unsafe Browser
6. Load a page in the Unsafe Browser. Not found

#6 Updated by BitingBird 2014-09-18 03:00:25

The unsafe browser is not meant to surf the internet, only connect (Feature #7774). Is the safe browser able to connect ?

#7 Updated by Anonymous 2014-09-18 07:25:30

I could not reproduce this behaviour.

#8 Updated by intrigeri 2014-09-18 09:16:48

> The unsafe browser is not meant to surf the internet, only connect (Feature #7774).

Indeed. And the main use case described in this ticket is about connecting, e.g. trying out different Wi-Fi hotspots in a airport.

#9 Updated by sajolida 2014-09-18 14:20:28

I tried this. If visit something.com, then change network and visit somethingelse.com I get an error.

But if I restart the unsafe browser and visit again somethingelse.com it works.

So, from what I see, restarting the unsafe browser is enough to get a working DNS resolver.

If it is technically difficult to update the DNS resolver without restartin the unsafe browser, I would be find with adding a note about that in the documentation.

#10 Updated by intrigeri 2014-09-19 08:07:07

> If it is technically difficult to update the DNS resolver without restartin the
> unsafe browser, I would be find with adding a note about that in the documentation.

I think it might be quite easy to update resolv.conf in the unsafe
browser chroot. Things might get a bit more hairy when we migrate it
to a stronger sandbox, though.

#11 Updated by BitingBird 2015-04-10 19:36:34

  • Assignee deleted (emmapeel)
  • QA Check deleted (Info Needed)

The info has been provided :)

#12 Updated by Anonymous 2017-06-29 14:59:24

  • Priority changed from Normal to Low

I was not able to test this again because I currently run Tails in a VM.

I’d say we should document this, instead of trying to solve it as it might get hairy anyway. People are mostly used to “turning [the browser] off and on again” when something doesn’t work and no other reports about this problem have been posted to this ticket since 3 years.

What do others think?

#13 Updated by intrigeri 2017-06-29 15:38:06

  • Subject changed from Unsafe Browser does not update resolv.conf when network changes to Document that one shall restart Unsafe Browser after reconnecting to a different network
  • Type of work changed from Code to End-user documentation

Agreed.

#14 Updated by anonym 2018-01-26 12:32:05

  • Status changed from Confirmed to In Progress
  • Assignee set to intrigeri
  • Target version set to Tails_3.6
  • % Done changed from 0 to 50
  • QA Check set to Ready for QA
  • Feature Branch set to feature/8243-meek

The feature branch fixes the issue, so no workaround is needed.

#15 Updated by intrigeri 2018-01-30 08:06:36

  • Assignee changed from intrigeri to bertagaz

#16 Updated by bertagaz 2018-02-24 19:23:48

  • Status changed from In Progress to Fix committed
  • Assignee deleted (bertagaz)
  • % Done changed from 50 to 100
  • QA Check changed from Ready for QA to Pass

anonym wrote:
> The feature branch fixes the issue, so no workaround is needed.

nice and clean! :)

#17 Updated by bertagaz 2018-03-14 11:06:27

  • Status changed from Fix committed to Resolved