linux-kernel - Re: [PATCH] Audit: Add TTY input auditing

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 7 Jun 2007 23:09:48 +0200 (CEST)
From:	Jan Engelhardt <jengelh@...putergmbh.de>
To:	Miloslav Trmac <mitr@...hat.com>
cc:	casey@...aufler-ca.com, Steve Grubb <sgrubb@...hat.com>,
	dwmw2@...radead.org, linux-kernel@...r.kernel.org,
	Alan Cox <alan@...hat.com>, Alexander Viro <aviro@...hat.com>
Subject: Re: [PATCH] Audit: Add TTY input auditing


On Jun 7 2007 21:28, Miloslav Trmac wrote:
>Casey Schaufler napsal(a):
>>> If we do not get commands typed at a prompt, we have to audit by execve. 
>> I would suggest that you'll have to do that as well so that you can tell
>> the difference between typed actions like these:
>> 
>> # cat > /dev/null
>> badprogram --badthing --everyone
>> ^D
>> #
>> 
>> # badprogram --badthing --everyone
>> 
>> where the same typed line is a Bad Thing in one case and completely
>> irrelevent in the other.
>The proposed patch audits each process separately, and includes a part
>of the command name in the audit event, so it is easy to distinguish
>between data entered into (cat > /dev/null) and the shell.
>
>The command name can be faked, but the actions necessary to fake the
>command name would be audited.

Someone please enlighten me why a regular keylogger² that captures
both input and output could not do the same. (Auditing what one has done.)

² http://ttyrpld.sf.net (there are others too)



	Jan
--