=================================================
In the light of these:
[ LXDE replacement question ]
https://forums.gentoo.org/viewtopic-t-9 ... ml#7558950
Tips and tricks for ConsoleKit, PolicyKit, and udev helpers
https://forums.gentoo.org/viewtopic-t-8 ... ml#7164546
and:
[ same thread ]
https://forums.gentoo.org/viewtopic-t-8 ... ml#6544053
also:
[ same thread ]
https://forums.gentoo.org/viewtopic-p-7 ... ml#7559286
(where I asked advice where to post this question)
as well as:
Air-Gapped Gentoo Install, Tentative
https://forums.gentoo.org/viewtopic-p-7 ... ml#7558880
which contains my blunt take on it.
=====================================
EDIT 2014-09-07
Lest users, advanced or regular, not get tired with too many of my citations,
let me call in here a real undisputed authority in GNU/Linux security, because
the link above on my Air-Gapped topic contains these lines.
Now readers, advanced or regular, shouldn't anymore be in lack of understanding...Have a look at another infrastructure introduced in the GNU/Linux kernel
back in late 2010, as it appears in this article by Brad Spender Spengler:
False Boundaries and Arbitrary Code Execution
https://forums.grsecurity.net/viewtopic.php?f=7&t=2522
...[snip]...
But, the suggested 'man capabilities' a stop to fuel up your understanding, and
it's not so hard to get the gist of it. C'mon!
...[snip]...
Did you just read how CAP_SYS_ADMIN can give [*] a (shadow sitting on a)spender wrote: CAP_SYS_ADMIN: generic: among many other things (it's a sort of catch-all capability choice), CAP_SYS_ADMIN grants the ability to mount/unmount filesystems.
So you have the ability to bind mount a new filesystem over an existing one to backdoor any binary on the system.
There doesn't appear to be any DAC check for this operation, so the capability itself is sufficient.
CAP_SYS_ADMIN also grants the ability to use the TIOCSTI ioctl against /dev/tty (a tty not owned by us) and inject commands into an administrator's shell that will be executed without any interaction on their part.
seat (that the user isn't even aware is rummaging in his machine) the:
"ability to bind mount a new filesystem over an existing one to backdoor any
binary on the system"
( [*] Spender talks to them straight. He says: "...you have the ability to bind
mount...". He can confront them. I can't. My defenses work, but they are yet so
very primitive, only based on backup and restore. I'm very much still learning
all the time. )
of what I mean like some of the posters further in this topic. I apologize for
not being cleverer in devising way to convey my meanings.
EDIT 2014-09-07 END
=====================================
I sincerely wish to try and go without dbus, consolekit and associates.
I really really wonder whether that is anymore possible at all, and whether
every user has to absolutely enable multi-seat install, else he/she is not
supported.
(pls. see in different wikis and docs how Gentoo devs only support either
systemd od consolekit/policykit, only those two options are officially
available)
It is good to repeat the most important information about consolekit (systemd
must be even worse, but I haven't studied it):
http://www.freedesktop.org/software/Con ... leKit.html
where there is, at the time of this writing, this precise definition:
Can I build my Gentoo on my box for just a sigle user? Is that banned in GentooA seat is a collection of sessions and a set of hardware (usually at least a
keyboard and mouse). Only one session may be active on a seat at a time.
now?
I hope the answer will be: No it's not. I'll never be.
But I've been trying the whole day today, to rid myself of what I never even
intended to install on my system (see the Air-Gapped Install for that, but it's
a hefty read... sorry, complex issues there... and here, won't be a short hit
and go question solved this one either)...
I spent the whole day so far, trying to rid myself of the dbus that the gtk+
somehow just can't install without...
I'll try and post the whole story, prepared till this moment but developing,
slowly at maybe two or three installments, the text so far, next.
I didn't want to open another topic, But the closest thread is of a misnomer
for this purpose, the LXDE replacement question linked above.
And no advice, at this time, on the Tips and tricks for ConsoleKit, PolicyKit,
and udev helpers thread, linked above as well.
So, after this theoretical introduction, I'll next try and post concrete
problems I have.
Miroslav Rovis
www.CroatiaFidelis.hr
======= cut off from this line to end if verifying hashes =======
File corresponding to this post: Gen_140529-140907_kit_dbus_stuff.txt,
was unedited since 2014-05-29, but was
Publictimestamp-ed today as # 1240700
With today's edit inclusion, the new file Gen_140529-140907_kit_dbus_stuff_NEW.txt
has Publictimestamp # 1240706
--
publictimestamp.org/ptb/PTB-21562 sha256 2014-09-07 15:01:45
2ED87C2B1924255526C65A9161F2543806654DE3F6165809BE0C53FBA37D3A9C
