Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Miloslav Trmac
Thanks for the review. Andrew Morton napsal(a): > On Wed, 06 Jun 2007 12:10:28 +0200 Miloslav Trmac <[EMAIL PROTECTED]> wrote: >> +/** >> + * tty_audit_opening - A TTY is being opened. >> + * >> + * As a special hack, tasks that close all their TTYs and open new ones >> + * are assum

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Casey Schaufler
--- Jan Engelhardt <[EMAIL PROTECTED]> wrote: > Someone please enlighten me why a regular keylogger² that captures > both input and output could not do the same. (Auditing what one has done.) 1. shell aliases # innocuous -p 0 2. shell variables # $INNOCUOUS -p 0 3. symlinks # ./innocu

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Alan Cox
On Thu, 07 Jun 2007 16:20:07 +0200 Miloslav Trmac <[EMAIL PROTECTED]> 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_

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Jan Engelhardt
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/n

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Miloslav Trmac
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 > # >

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Casey Schaufler
--- Steve Grubb <[EMAIL PROTECTED]> wrote: > Ok, this is a sample set of requirements we are trying to meet: > > Implement automated audit trails for all system components to reconstruct the > > following events: > All actions taken by any individual with root or administrative privileges M

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Steve Grubb
On Thursday 07 June 2007 11:42, Casey Schaufler wrote: > > tools like rootsh, but that is too easy to detect and defeat. And then it > > does not put its data into the audit system where its correlated with > > other system events. > > The evaluation teams that I have worked with (OrangeBook and CC

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Alan Cox
> logging to meet the audit requirements. You have to log what happened. > Logging what was requested is insufficient and logging what was > typed, which may or may not have resulted in an actual request is > not helpful to meeting security audit requirements. Key information can answer some quest

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Casey Schaufler
--- Steve Grubb <[EMAIL PROTECTED]> wrote: > On Thursday 07 June 2007 04:13:42 Jan Engelhardt wrote: > > >Add TTY input auditing, used to audit system administrator's actions. > > > > _What_ exactly does it audit? > > In theory, it should audit the actions performed by the sysadmin. This patch

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Miloslav Trmac
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. Norma

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Steve Grubb
On Thursday 07 June 2007 04:13:42 Jan Engelhardt wrote: > >Add TTY input auditing, used to audit system administrator's actions. > > _What_ exactly does it audit? In theory, it should audit the actions performed by the sysadmin. This patch doesn't cover actions done via X windows interface. > An

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Alan Cox
> > + if (filp->f_op->read == tty_read) { > > + disable = 0; > > + break; Why says a tty will always have f->op->read == tty_read ? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a mess

Re: [PATCH] Audit: Add TTY input auditing

2007-06-07 Thread Jan Engelhardt
On Jun 6 2007 11:49, Miloslav Trmac wrote: >From: Miloslav Trmac <[EMAIL PROTECTED]> > >Add TTY input auditing, used to audit system administrator's actions. _What_ exactly does it audit? And why does it only audit sysadmin actions? Is this supposed to be a keylogger? >TTY input auditing works o

Re: [PATCH] Audit: Add TTY input auditing

2007-06-06 Thread Andrew Morton
On Wed, 06 Jun 2007 12:10:28 +0200 Miloslav Trmac <[EMAIL PROTECTED]> wrote: > From: Miloslav Trmac <[EMAIL PROTECTED]> > > Add TTY input auditing, used to audit system administrator's actions. > TTY input auditing works on a higher level than auditing all system > calls within the session, which

Re: [PATCH] Audit: Add TTY input auditing

2007-06-06 Thread Miloslav Trmac
I'm sorry, I forgot the patch... From: Miloslav Trmac <[EMAIL PROTECTED]> Add TTY input auditing, used to audit system administrator's actions. TTY input auditing works on a higher level than auditing all system calls within the session, which would produce an overwhelming amount of mostly useless