Euh Mark...I think you're talking about something else here :-)  We chatted
about a remote check (agent based) and what you're describing looks like
that.  What I'm talking about is/was a option to execute a command on a
remote system (as an alert that is).
 


Dirk.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Mark Bradshaw
Sent: Wednesday, November 24, 2004 1:30 PM
To: [EMAIL PROTECTED]
Subject: RE: [SA-list] Remote command as alert

Let's step back and consider what an agent might be capable of that
psexec isn't.  An agent is capable of advanced checks and logic to check
process, states, etc.  Psexec is just a method of executing remote
processes.  Now, either way you go, someone still has to build an agent
capable of doing a useful check, however going the agent route allows
more flexibility in how data is returned after the check completes via a
structure command and response language that SA would understand.
Another thing to note, psexec is a windows technology, whereas the agent
methodology could be used in linux, solaris, atari, Xbox, etc.  I know
there are other mechanisms to achieve the same result in *nix world, but
I think there's an inherent benefit in keeping it with one centralized
methodology.  I would vote agent.

 . . . If this were a democracy. :o)

Mark Bradshaw
Director of Online Services
DREAM3
http://www.dream3.org
(866) 7DREAM3
-----Original Message-----
From: Alistair Francis [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, November 24, 2004 3:48 AM
To: [EMAIL PROTECTED]
Subject: RE: [SA-list] Remote command as alert

>From what I'm picking up off the list, psexec sounds very similar to 
>the
hacking tool "netcat". Instantiating a remote command prompt on an
NT/2000 system, that is basically accessible to anyone who knows it's
there, should give everyone the willies. It takes about 10 seconds to
fire off an LSASS injector to dump the SAM into a txt which you can then
password crack offline, at your leisure! Be very scared!

Ideally, the client/server setup would be more secure in that the
listener/agent would only accept connections from the appropriate server
(as Dirk mentioned, specified IP, username, password, etc). Nothing is
100% safe but this is a hell of a lot safer. IMHO.

Alistair Francis
Systems Administrator
Comm Express Services SA (PTY) LTD
TEL:    +27 (0)11 475-5567
FAX:   +27 (0)11 475-6238
CELL: +27 (0)84 607-7325

The information contained in this electronic mail message is
confidential to the Matragon group of companies and may enjoy legal
privilege. The contents are intended solely for the addressee and access
thereto by anyone else is unauthorised. Should you not be the intended
recipient, kindly delete the message and inform us. Any disclosure,
copying or distribution is prohibited and may be unlawful. Please also
note that any action taken, or omitted to be taken in reliance on the
information contained herein is done at your own risk.



-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Dirk Bulinckx
Sent: 24 November 2004 09:10
To: [EMAIL PROTECTED]
Subject: RE: [SA-list] Remote command as alert


The way psexec (probably) works isn't therefore something great.
>From my understanding:
        it copies it self (or part) to the remote system
        install it as a service
        starts the service
        service executes the "commandline"
        service stops
        service removes itself
        service file is deleted

For this to work you need not only admin access to the remote system but
you also need netbios access to the remote system.

With an agent approche all would be done via ONE port (TCP probably) and
as such will not require the use of Netbios.


Dirk.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Kurt Buff
Sent: Wednesday, November 24, 2004 1:59 AM
To: '[EMAIL PROTECTED]'
Subject: RE: [SA-list] Remote command as alert

Wouldn't necessarily require an agent.

See http://sysinternals.com for their psexec utility.

All you need is admin rights on the remote box.

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of Alistair Francis
> Sent: Tuesday, November 23, 2004 03:58
> To: [EMAIL PROTECTED]
> Subject: RE: [SA-list] Remote command as alert
>
>
> Yup, it would be useful, even though it requires an agent on the 
> target machine. The only problem I can see is security. It would need 
> to be seriously tight, possibly some sort of encrypted key between 
> server and agent. If it was as simple as telneting to a certain port 
> on the target mac and issuing commands... (shudder!) imagine what a IT

> savvy disgruntled employee could do! Are you thinking of a restricted 
> command set or something along the lines of a remote command prompt 
> type thing?
>
> Alistair Francis
> Systems Administrator
> Comm Express Services SA (PTY) LTD
> TEL:    +27 (0)11 475-5567
> FAX:   +27 (0)11 475-6238
> CELL: +27 (0)84 607-7325
>
> The information contained in this electronic mail message is 
> confidential to the Matragon group of companies and may enjoy legal 
> privilege. The contents are intended solely for the addressee and 
> access thereto by anyone else is unauthorised. Should you not be the 
> intended recipient, kindly delete the message and inform us. Any 
> disclosure, copying or distribution is prohibited and may be unlawful.

> Please also note that any action taken, or omitted to be taken in 
> reliance on the information contained herein is done at your own risk.
>
>
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of Dirk Bulinckx
> Sent: 23 November 2004 13:28
> To: [EMAIL PROTECTED]
> Subject: [SA-list] Remote command as alert
>
>
> Is the possibility to execute a command on a remote system (currently 
> only possible on the system running SA), something that would be of 
> any help knowing that for that you would have to have a 
> "remotecommand" service installed on the system(s) that need to 
> execute those commands.
>
> Dirk.
>
>
>
>
>
>
> -------------------------
>
> [This E-mail scanned for viruses by Declude Virus]
>
> To unsubscribe from a list, send a mail message to [EMAIL PROTECTED] 
> With the following in the body of the message:
>    unsubscribe SAlive
>
>
>
>
> -------------------------
>
> [This E-mail scanned for viruses by Declude Virus]
>
> To unsubscribe from a list, send a mail message to [EMAIL PROTECTED] 
> With the following in the body of the message:
>    unsubscribe SAlive
>







-------------------------

[This E-mail scanned for viruses by Declude Virus]

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive






-------------------------

[This E-mail scanned for viruses by Declude Virus]

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive




-------------------------

[This E-mail scanned for viruses by Declude Virus]

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive



-------------------------

[This E-mail scanned for viruses by Declude Virus]

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive






-------------------------

[This E-mail scanned for viruses by Declude Virus]

To unsubscribe from a list, send a mail message to [EMAIL PROTECTED]
With the following in the body of the message:
   unsubscribe SAlive

Reply via email to