[Vol-users] Re: Bug or documentation error - linux_dump_map
edwin.smulders at gmail.com
Fri Mar 8 04:24:57 CST 2013
The whole plugin is definitely much better now, and it works. Thanks!
On 7 March 2013 17:03, Michael Hale Ligh <michael.hale at gmail.com> wrote:
> Hi Edwin,
> You're right, it looks like there were a few different issues with that
> plugin. I applied a patch  similar to yours that should make things
> easier. It now accepts -p, but also takes a --dump-dir argument instead of
> --outfile. You can dump all VMAs from all processes, all VMAs from a
> specific process (or multiple processes with -p 1,2,3) or a specific VMA
> from a specific process. The output file name includes the task ID as well
> as the VMA starting address. There's a summary printed to stdout to minimize
> For example:
> Task VM Start VM End Length Path
> ---------- ---------- ---------- ---------- ----
> 1 0x08048000 0x0811d000 0xd5000 dumps/task.1.0x8048000.vma
> 1 0x0811e000 0x08124000 0x6000 dumps/task.1.0x811e000.vma
> 1 0x08124000 0x08125000 0x1000 dumps/task.1.0x8124000.vma
> 1 0x08125000 0x081b0000 0x8b000 dumps/task.1.0x8125000.vma
> 1 0xb7522000 0xb752e000 0xc000 dumps/task.1.0xb7522000.vma
> Can you try the patch and let me know if it works for you?
> . https://code.google.com/p/volatility/source/detail?r=3169
> On Thu, Mar 7, 2013 at 7:06 AM, Edwin Smulders <edwin.smulders at gmail.com>
>> After some more research, I think it is a bug and the attached patch
>> hopefully fixes it.
>> On 7 March 2013 11:35, Edwin Smulders <edwin.smulders at gmail.com> wrote:
>> > Hi,
>> > Yesterday during a challenge we had to use the linux_dump_map plugin
>> > to dump a process stack, and the documentation at
>> > https://code.google.com/p/volatility/wiki/LinuxCommandReference23#linux_proc_maps
>> > says it has the -p option to select a process.
>> > However, as far as I can tell looking in the svn history, this plugin
>> > never had the -p option. And it's definitely not working currently.
>> > I've heard a confirmation that the option was working in version
>> > 2.2-rc1, so maybe it was a global option?
>> > The reason I'm mailing this is because, if the -s is virtual memory,
>> > would you not get possible overlap in areas? How do you know it dumped
>> > the correct VMA? Note that every time I tried, I got the correct area.
>> > Cheers,
>> > Edwin
>> Vol-users mailing list
>> Vol-users at volatilityfoundation.org
More information about the Vol-users