Re: micro-virtualization

David-Sarah Hopwood <[email protected]> Sun, 23 Sep 2012 02:20:04 +0100
Newsgroups gmane.comp.lang.e.general
Message-ID <[email protected]>
On 22/09/12 04:37, Paul E Baclace wrote:
> Hi, everyone.  I just saw an announcement about the "micro 
> virtualization" infrastructure from Bromium (founded by Simon Crosby, a 
> XenSource founder).  It appears to be an effort to encapsulate processes 
> in a lightweight way without using a vm for each process playpen.  An 
> overview:
> 
> http://www.brianmadden.com/blogs/guestbloggers/archive/2012/09/20/guest-blog-from-simon-crosby-explaining-what-bromium-is-and-how-it-works.aspx

I'm confused. Is this an implemented system that works for isolation of
tasks within existing Windows applications? What is described doesn't seem
implementable:

# By ensuring that each vulnerable or untrustworthy task (eg: opening a
# web page or an email attachment) is executed in its own micro-VM, Bromium
# can guarantee that a compromised task cannot access enterprise data or
# applications.
#
# Bromium assumes that at some point a task in a micro-VM will be attacked
# and will be compromised. The granular isolation afforded by the Microvisor,
# together with the resource control policies, ensures that any attack will
# be confined to the micro-VM, that no enterprise data will be stolen, and
# that the attack will be automatically discarded.

If tasks are at the granularity of opening a particular web page or
email attachment, that seems as though it would require either completely
new web browsers, email clients, etc., or drastic rearchitecture of existing
ones. Existing apps assume shared memory within a process, so there's no
way to provide isolation between, say, the code that is opening an email
attachment and the code that has access to the rest of the user's email
and contacts, without breaking the app.

This isn't affected by whether you have hardware virtualization support or
not. You *might* be able to reduce reliance on the host OS's security
mechanisms by using hardware virtualization, but I'm skeptical (the hardest
problem isn't implementing isolation; it is providing enough sharing, and
only the required sharing, for software to be able to work).

-- 
David-Sarah Hopwood ⚥

_______________________________________________
e-lang mailing list
[email protected]
http://www.eros-os.org/mailman/listinfo/e-lang
signature.asc (application/pgp-signature, 554 B)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJQXmPJAAoJEAZ/SSCYzydiJ4QH/RDhUPa1CRBl6Dp9BoZbhVzr
kLL+yfmIHq8uXX8qQfBiZkrxyWECVtOuJzUA+iQ7Bvdjweo6jv6y0DxM2LmcfLxZ
6o77xDrudZDJlkRqQQSSVWAGbMDmb8rtAWVZ4Wt/CGOTBIHITANy4M1JDrV+b4AB
rm3H0DgIwR+Xi5FnISlNL6c7nvhJeDUnCYwcfULmg9NwY1XoSXP6gSMnEhHTmtD4
eGpfAzLpF32+ZZVq7K2CTiPAPbKBKXVa4ob3oAXqQLvMDfTO/gCnm+iL8N/29o/l
J40i/3l1A673KuHlfYfTGnc1sRfRweACon7CvJ4PHTa3AxdxYx5dalrbTRFytBk=
=DcKR
-----END PGP SIGNATURE-----