Colleagues!

Earlier in this thread, we learned that installing xserver-xorg-legacy
allows you to run X the old way, as a setuid script. 

The default upgrade path from jessie -- in which X11 was
setuid-only -- migrates to a new xserver-xorg in which the
setuid mechanism is replaced. In order to run X with user
permissions in the dist-upgrade'd environment one needs to
pull in a stack of dependencies including dbus, polkit,
libpam-elogind, and elogind. 

I think it may be a bug that in the case of my upgrade
experience, neither xserver-xorg-legacy (a wrapper that
enables setuid X) nor this pam stack were installed, so
startx failed for me. Perhaps the experience is different
with a display manager installed.

I have and use dbus apps on my system, However, as far as
I'm aware, none of these programs has root privileges. 

As the pam/dbus/elogind/polkit mechanism is capable of
handing out root authority, and as all software has bugs, I
think we _can_ anticipate that bugs that create security holes
will be uncovered in this stack. How much scrutiny did the
developers devote? Did anyone ever consider security at
through the whole stack? Probably the developers of each
component do consider security in their own code.

openssl had a big hole for years, and before that debian's random
number generator was broken. Showstopping
holes, but the show goes on...

Will someone who scrutinizes closer have a back door,
is that likely be true for the foreseeable future?

In a way, running others' code is like driving: putting
oneself in the hands of strangers you've never met and
might not trust for minute in person.

I read about the art of "fuzzing" programs with various
combinations of random inputs, to discover bugs such as
buffer overflows. This technique has been used to find bugs
and improve security in many languages.  It was also used to
find hidden instructions and other attributes of
microprocessors. 

https://github.com/xoreaxeaxeax/sandsifter/blob/master/references/domas_breaking_the_x86_isa_wp.pdf

I see fuzzing tools for dbus also available. 

I think it's an interesting security question, since the default
state of a distribution is so influential.

That PAM is finely grained, I get, so on the surface, it
looks superior to the big club of root permissions.

I'd be interested to links to any discussions of these
topics. I see the CVEs are published, in this example,
smb4k is being careless in arguments it passes to dbus,
leading to an exploit. 

https://nvd.nist.gov/vuln/detail/CVE-2017-8849


cheers


-- 
Joel Roth
  

_______________________________________________
Dng mailing list
[email protected]
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Reply via email to