On 16/02/11, Max Timchenko wrote:
> On Wed, Feb 10, 2016 at 9:30 PM, Richard Guy Briggs <[email protected]> wrote:
> 
> > On 16/02/10, Max Timchenko wrote:
> > > Has anyone tried that before? What would actually happen if two different
> > > audit clients tried to use the same interface to the audit subsystem in
> > the
> > > kernel?
> >
> > With recent changes upstream, the second would be denied with -EEXIST.
> >
> > Before that, the older one would be starved out.  And versions even
> > older might actually have the newer one orphaned in the very occasional
> > race where the older one shuts down after the second one starts.
> >
> > To quote Highlander, "There Can Be Only One".
> 
> Thanks Richard and Paul for your quick responses. It's great to hear
> that support for containers is being worked on.
> 
> I have read the docs on audispd(8) - is it something auditd and the
> other client could use to enable multiple access? It sounds like
> audispd does support multiple clients, but I would guess all clients
> would have to use the audispd plugin interface instead of the usual
> kernel API.
> 
> What is missing from the documentation for me is the relationship
> between audispd and auditd - whether audispd is an optional component
> of auditd that can run concurrently, or audispd is a replacement of
> auditd when configured (and then auditd cannot run on the same machine
> without running into the same multi-client issues).

I will defer to Steve Grubb on this quesition as the userspace tools are
his domain of expertise.

> Yours,
> --
> Max

- RGB

--
Richard Guy Briggs <[email protected]>
Senior Software Engineer, Kernel Security, AMER ENG Base Operating Systems, Red 
Hat
Remote, Ottawa, Canada
Voice: +1.647.777.2635, Internal: (81) 32635, Alt: +1.613.693.0684x3545

--
Linux-audit mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-audit

Reply via email to