GitLab

In the past, we had "exit 1" in there. This problem was identified (#8571) and
fixed (commit 1b46b5b1) in Tails 1.2.3, by replacing this "exit 1"
with "return 1".

Then, while working on another, minor problem (#8687), the "exit 1" was
reintroduced (commit 4ea050ab) in Tails 1.3.2, presumably because we pasted
sample code from the ticket, that had been drafted *before* #8571 was fixed.
Sadly, nobody noticed in time.

I cannot easily reproduce the bug on Tails/Wheezy (because I lack hardware whose
MAC address fails to be spoofed there), but I could reproduce it with
Tails/Jessie ("thanks" to the fact we're currently shipping the buggy 'wl'
kernel module in there).

Will-Fix: #9531

Together with `set -e` that error has prevented this script from
restarting Vidalia, like it should.

Will-fix: #9328

We haven't sourced the Tails shell library's loggin module, so log()
is undefined. Since we `rm -f` the file, log():ing if it still exists
afterwards *should* be pointless, so let's just not do it all.

... cause the control port language doesn't care about it.

(We could have skipped the whole business of using regexes and just
upper():ed the `line` variable and then do a normal string comparison,
but that would also upper() any arguments. It doesn't matter now since
we only care about commands, but if we ever extend this filter so that
arguments do matter, we won't get shot by potentially missing that.)

By checking that the command is ended with a word boundary.

Will-fix: #8687

Will-fix: #7400

The purpose of these settings was to show a green TorButton button in
the toolbar. Since this button is now hidden these settings can be
removed.

Will-fix: #8893

That feels a bit more sane. Also give it a more consistent name.

Currently we don't pass this function anything that requires quoting,
but let's be a bit more defensive.

Will-fix: #8991

According to NetworkManager(8) IP4_NAMESERVERS is a space separated
list.

We do an actual login (via `su`) and extract the resulting environment
so that e.g. LANG is set and /etc/environment gets sourced which sets
PATH so that we look for executables in /usr/local/*bin, which truly
matters since some of them are wrappers around executables found in
some other PATH-dir.

This might be useful to debug weird setups.

Without this, the "Network card $NIC disabled" notification is
sometimes lost when MAC spoofing fails.

If macchanger fails we have to treat the situation as undefined. We
can't trust that get_current_mac_of_nic() returns the actual state of
the hardware; perhaps it instead reports the driver's state that was
successfully spoofed but not applied in the hardware?

If we do this, the panic mode code will never be executed, so we will
in fact potentially let the device leak the real MAC address on the
network.

We still do it for `try_for()` blocks, since they quite quickly can
puke out screenfulls of errors.

So it's not confused with Tor Browser's default search eninge,
Startpage.

We were already ignoring directories in process_IN_OPEN, but it looks like it's
not enough anymore. Let's try harder to skip them.