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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20070607225945.1aa71a93@the-village.bc.nu>
Date:	Thu, 7 Jun 2007 22:59:45 +0100
From:	Alan Cox <alan@...rguk.ukuu.org.uk>
To:	Miloslav Trmac <mitr@...hat.com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>, dwmw2@...radead.org,
	linux-kernel@...r.kernel.org, Alan Cox <alan@...hat.com>,
	Steve Grubb <sgrubb@...hat.com>,
	Alexander Viro <aviro@...hat.com>
Subject: Re: [PATCH] Audit: Add TTY input auditing

On Thu, 07 Jun 2007 16:20:07 +0200
Miloslav Trmac <mitr@...hat.com> wrote:

> Alan Cox napsal(a):
> >>> +			if (filp->f_op->read == tty_read) {
> >>> +				disable = 0;
> >>> +				break;
> > Why says a tty will always have f->op->read == tty_read ?
> AFAICS from tty_io.c, it will always be tty_read or hung_up_tty_read.
> Normal user processes would exit after SIGHUP and not reopen a TTY.
> 
> (I have copied the condition from __do_SAK().  That of course doesn't
> mean it's correct.)
> 	Mirek

Right it may be hung_up_tty_read that was what bothered me. I've had a
think through the different scenarios and I can't think of a simple one
where I can abuse this as the vhangup() path is current root triggered
and loses the tty (so I can't reopen on it)

There are more complex questions - what happens when the much needed
revoke() goes mainstream [and we fix all the security issues its lack
causes], and the case where I do

	login on tty1
	login on tty2

On tty1 run a process which sets nohup and causes a vhangup then opens
tty2 while tty2 command line is running some long running program that
doesn't take input that I could plausibly run legitimately (eg a long
complex sql query, or a slow security check etc)

Alan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/