linux-kernel - Re: [PATCH 0/2] struct pid-ify autofs4

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] [day] [month] [year] [list]

Message-ID: <1351124728.3116.8.camel@perseus.themaw.net>
Date:	Thu, 25 Oct 2012 08:25:28 +0800
From:	Ian Kent <raven@...maw.net>
To:	Miklos Szeredi <miklos@...redi.hu>
Cc:	"Eric W. Biederman" <ebiederm@...ssion.com>,
	autofs@...r.kernel.org, linux-fsdevel@...r.kernel.org,
	linux-kernel@...r.kernel.org, sukadev@...ux.vnet.ibm.com,
	serge.hallyn@...onical.com
Subject: Re: [PATCH 0/2] struct pid-ify autofs4

On Wed, 2012-10-24 at 16:59 +0200, Miklos Szeredi wrote:
> > Ian Kent <raven@...maw.net> writes:
> > 
> > > Yeah, the problem with that is that "autofs doesn't work if containers
> > > are used" is ill defined since there are use cases where it does, I
> > > believe. At the very least, ill defined in my view of things.
> 
> Customer says:
> 
>  "There is no interaction between host and the conatainer.  The host use
>   only his own automount and each containers used automount in their
>   container."

That sounds like a sensible requirement to me.

> 
> I think it's a pretty clearly defined use case.  And one which automount
> could easily support since the only requirement is that all namespaces
> are treated equally.

Yep. A problem might be dealing with mounts cloned from the parent
namespace at container creation. Ideally they wouldn't be duplicated so
they wouldn't need to be cleaned up (perhaps that's justified given the
requirement above).

Another thought is, what would happen on just cloning a namespace, not
necessarily as a container (is that even a sensible question)? The user
may actually want the mounts in this case, and can we even tell the
difference at namespace creation?

> 
> But I agree that adding safeguards against cases which don't have such
> easily defined semantics (such as triggers from several different
> namespaces).

Look forward to it.

> 
> I'll post updated patches.
> 
> Thanks,
> Miklos
> 


--
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/