[framework] Auxiliary Module Questions

[Kashif Iftikhar]

Hello again,

  Okay here I am again with more questions :-)

1. The Auxiliary::Report mixin has the "report_service" method,
Looking at the DB schema, the "services" table has a "info" text
field. However, the parameters passed to report_service (an opts hash)
don't seem to take info into account. I can't find a way to populate
the info field in the DB with extra info about the service.

2. The "vulns" and "vulns_refs" tables in the DB can get populated
automatically when used with Exploit modules. But how about auxiliary
modules? Is there a way to put data into those tables from within
auxiliary modules. For example, I try to scan web servers for HTTP PUT
and DELETE methods and when I find one that is exploitable, how do I
put the extra info into the DB. One way is to utilize the info field
in services but I guess the the vulns table is meant for that. So how
can I use that?


3. The Auxiliary::Scanner mixin supports three methods "run_host(ip)",
"run_range(range)" and "run_batch(batch)". My understanding is that
these methods are exclusive and only one should be included in a
module that wants scanner functionality. The run_host method gets
called for each IP sepecified in by the RHOSTS option. But for
run_range this gets tricky,  the range parameter passed to run_range
is a string so it is not iterable. How do I iterate through a range
then? Do I need to manually process the string or is there another
way? The Rex::Socket::RangeWalker is called for run_host and run_batch
in scanner.rb but not for run_range. run_batch works fine for me as
the batch is passed in as an array.  Can someone explain why run_range
doesn't get an array too so that it is easily iterable.

4. The second question about Auxiliary::Scanner is, run_host seems to
process one IP at a time and goes in the traditional sequential flow.
My understanding is that run_batch creates one thread for each batch
to facilitate parallel scanning, am I correct about this one? And what
about run_range again? How does parallel scanning/execution work with
that? Does it create a thread for each IP in the range? All these 3
methods can be made to run in passive by specifying the auxiliary
action in PassiveActions. Having these 3 different type of methods
seems to point that they are meant for different scan requirements. So
any suggestions or tips when to use which one? I guess when we want to
limit the max no of threads instantiated (and bandwidth usage) then we
use run_batch. I am still a bit unclear about run_range though.

  Thank you in advance for your help.

- Kashif.




On 4/17/07, mmiller at hick.org <mmiller at hick.org> wrote:
> On Tue, Apr 17, 2007 at 05:48:50PM +0000, Kashif Iftikhar wrote:
> > Hello,
> >
> >  I am trying to put together an HTTP PUT/GET auxiliary module. I have
> > a few questions regarding the working of auxiliary modules.
> >
> > 1. What is the difference between auxiliary actions (of class
> > AuxiliaryAction) and auxiliary commands? Looking at the source code,
> > the auxiliary_commands accessor function returns an empty dictionary (
> > { } ). Does this mean that auxiliary commands have not yet been
> > implemented?
>
> Auxiliary commands allow you to add custom commands to the console
> interface.  For example, there are builtin commands like 'exploit' and
> 'check' for exploit modules.  Auxiliary modules by default all have the
> 'run' command.  You can add additional commands through the
> auxiliary_commands method.  Take a look at admin/backupexec/registry.rb
> for an example.  These additional commands are only applicable to the
> one specific auxiliary module that defines them.
>
> > 2. Regarding auxiliary actions, I was able to add auxiliary actions by
> > passing either
> >
> >  'Actions'        => [AuxiliaryAction.new('PUT_FILE', {}),
> >                                 AuxiliaryAction.new('DELETE_FILE', {})],
> >
> >    OR
> >
> >  'Actions'     =>
> >                               [ ['My Aux action'] ],
> >
> >  Looking at the constructor of AuxiliaryAction class, there is also
> > an options dictionary/hash named opts that can be passed. For the time
> > being I am passing it empty. What keys/values can come into opts? How
> > do I add description for an action.
>
> This hash is to let you have dynamic variables associated with your
> actions.  These only have meaning within your auxiliary module, and
> aren't displayed to end-users.  For example, if you look at exploit
> modules, the Targets information hash element has a name and a hash of
> options.  The concept is the same with auxiliary modules.
>
> >  doing 'show actions' in msfconsole just displays the action name
> > without any description. I tried adding a 'Description' key to the
> > opts passed to AuxiliaryAction class but it had no effect.
>
> There isn't a way at the moment to provide a description with actions.
> We could look into doing this by standardizing on some hash element keys
> such as 'Description'.
>
> > 3. How does an action become passive? Do I just include it in the
> > PassiveActions list/array and it automatically gets to run as passive
> > or are there other requirements?
>
> Yep, I believe this is all you need to do.
>
> > 4. Is there a way to monitor the status/progress of auxiliary scanner
> > modules? Can this progress be monitored from the database without
> > calling msfcli or msfconsole etc?
>
> You should be able to select from various tables, but I don't think
> there is a way to figure out "percentage" complete, for example.
>
> > 5. The auxiliary scanner modules provide three ways (scan host, scan
> > range and scan batch) to do a scan. Exploring the source code, I was
> > able to figure out these:
> >
> >  * To scan a host the module should implement the run_host() method.
> >  * To scan a range, the run_range method should be implemented.
> >  * To scan a batch, the run_batch() method and also a batch_size
> > accessor function should be implemented.
> >
> >  My question is from where does the run_host(ip) method gets the IP,
> > the run_range(range) method gets the range and the run_batch(batch)
> > method gets the batch? How can a user specify a range or a batch using
> > msfconsole?
>
> If I recall correctly, ranges are specified by setting RHOSTS to
> something like 10.0.0.0/8 or some other network address form.
>
> >  I am planning to write a tutorial on auxiliary modules developments
> > and on some useful mixins like HttpClient, any other tips suggestions
> > will also be greatly appreciated.
>
> Great!  We look forward to seeing it.  Feel free to send more questions
> to the list as you have them.
>
>
>