Closed GoogleCodeExporter closed 9 years ago
Issue 209 has been merged into this issue.
Original comment by mike.auty@gmail.com
on 12 Feb 2012 at 9:01
We also need to include windows 7 in there, and we've got people starting to
try out trunk running into this...
Original comment by mike.auty@gmail.com
on 12 Feb 2012 at 9:02
Actually for systems Vista+ they should be using netscan, not any of these
plugins ;-)
Original comment by jamie.l...@gmail.com
on 14 Feb 2012 at 12:31
Few things regarding this:
* Gleeda's right technically for Win 7 the one network-related plugin is
netscan. Although netscan has been ported to x64 there may be some issues on
Vista x64 so we might as well include netscan for x64 in this issue (AW and I
are aware of this and will be spending some time on it shortly)
* If we don't get to the whitelist/blacklist plugins before 2.1 we should
remember to at least add statements to connections/connscan/sockets/sockscan to
back off >= vista systems and to netscan to back off < vista systems (but there
is already an issue for this so its just duplication here...but a good
reminder).
Original comment by michael.hale@gmail.com
on 14 Feb 2012 at 2:57
Original comment by michael.hale@gmail.com
on 26 Mar 2012 at 1:04
Ok, so I recently added some plugin whitelisting/blacklisting code. At the
moment it just hides the plugins from --help (or shows them with -v), but
doesn't actually prohibit the plugin from running. I can add in stronger
blocks, but it's the kind of thing I can see irritating people if they really
know what they're doing and have a situation where it would all work if only we
weren't stopping them. How do people want to handle this?
Original comment by mike.auty@gmail.com
on 6 May 2012 at 11:37
Hey Mike,
I think hiding them from --help is a good start. I think whether we add
stronger blocks depends on the plugin. For example using netscan on xp/2003
will never work, even if you know what you're doing. Likewise,
connscan/connections/sockscan/sockets will never work on 2008/vista/7. So those
are perfectly fine to block strongly. We might want to make a list somewhere of
any plugins that don't work on various profiles (something like the OS Support
column of http://code.google.com/p/volatility/wiki/FeaturesByPlugin but that
probably needs updating itself) which could make it easier to decide per plugin
what is the best choice.
Also here's a patch that fixes connections, connscan, sockets, and sockscan on
windows 2003 sp1 and sp2 x64. I'll talk with Gleeda tomorrow about testing it
on some of her images, but it works on mine so its a step in the right
direction for closing this issue.
Support for xp x64 (same as 2003 sp0 x64) will come in a separate patch, as
will any updates needed for vista, 2008, and 7 to the netscan plugin.
Original comment by michael.hale@gmail.com
on 9 May 2012 at 3:00
Attachments:
Oh I forgot to mention, if Gleeda or anyone else gets to try this patch and it
doesn't work, please moddump your tcpip.sys and send it to me (or if you have
access to the disk of the system, send the real tcpip.sys from
c:\windows\system32\drivers). -Thanks
Original comment by michael.hale@gmail.com
on 9 May 2012 at 3:11
After some more testing by Gleeda, which resulted in us adding support for one
more tcpip.sys module from win2003 x64 (thanks Gleeda!), this patch has been
committed.
Note: one particular win2003sp2x64 we've seen has a 32-bit tcpip.sys module
loaded. thus the _AddrObjTable and _TCBTable symbols are 32-bit pointers, but
our code in network.py uses the 'address' type which for the 64-bit profile
will be a 64-bit
value. until we have a way of saying "use a 64-bit profile, but for this
particular plugin, an address should be 32-bits" then the list walking network
plugins (sockets and connections) will fail. also since the 32-bit tcpip.sys
structures are going to differ in size, the connscan and sockscan will find the
right pools but print bad data for several members.
Original comment by michael.hale@gmail.com
on 9 May 2012 at 6:12
The netscan command should work on vista/2008/7 all service packs, x86 and x64.
I've built an alternate method of finding connections based on tcpip.sys
partition tables and hash buckets (the way the OS enumerates connections when
you use netstat), but the netscan fixes should suffice for closing this issue.
Let me know if you have questions or issues using the plugins ;-)
Original comment by michael.hale@gmail.com
on 18 May 2012 at 10:49
Original issue reported on code.google.com by
jamie.l...@gmail.com
on 24 Jan 2012 at 7:28