netdev - Re: sysfs^H^H^H^H^Hsysctl warnings, reserved names

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]

Date:	Fri, 17 Feb 2012 14:05:34 +0100
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	"Eric W. Biederman" <ebiederm@...ssion.com>
Cc:	Denys Fedoryshchenko <denys@...p.net.lb>, netdev@...r.kernel.org
Subject: Re: sysfs^H^H^H^H^Hsysctl warnings, reserved names

Le vendredi 17 février 2012 à 04:07 -0800, Eric W. Biederman a écrit :
> Eric Dumazet <eric.dumazet@...il.com> writes:
> 
> > Le vendredi 17 février 2012 à 11:37 +0100, Eric Dumazet a écrit :
> >> Le vendredi 17 février 2012 à 12:14 +0200, Denys Fedoryshchenko a
> >> écrit :
> >> > Hi
> >> > 
> >> > Just did a test:
> >> > 
> >> > ip link set dev eth1 name default
> >> > 
> >> > And got a lot of (expected) sysfs warnings:
> >> > [1068625.677143] sysctl table check failed: 
> >> > /net/ipv4/conf/default/promote_secondaries  Sysctl already exists
> >> > [1068625.677151] Pid: 18106, comm: ip Not tainted 3.2.4-centaur #1
> >> > and etc
> >> > 
> >> > Kernel 3.2.4, but i guess it doesn't matter much.
> >> > Maybe such names (as all and default) should be "reserved"?
> >> > Or it is ok like that?
> >> > 
> >> 
> >> You're right, we should forbid this to happen.
> >> 
> >> 
> >
> > Following is a quick hack, I CC Eric because its probaly better to
> > address this in sysfs.
> 
> Well we are talking about sysctl not sysfs but same difference, I keep
> an on eye on them.
> 
> I expect renaming a network device to "all" or "default" would be a
> problem in any kernel supporting renaming of networking devices.
> 
> At the basic level of handling it.  sysctl checks for this sometimes
> now and as soon as my sysctl tree is merged the checks will become
> unconditionally present.  In what sense were you thinking it would
> be better to address this in the sysctl?
> 

Because problem is : it seems a rename() of "eth3" to "default" is
allowed by sysctl, yet "default" is already in directory.

It seems sysfs_rename_link() / sysfs_rename_dir() / sysfs_rename() might
have a problem ?

Sorry, I leave for vacations and can check all this before 5 days.



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html