Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 8 Feb 1996 17:33:13 +0100
From:      Wolfram Schneider <wosch@cs.tu-berlin.de>
To:        Peter Wemm <peter@jhome.DIALix.COM>
Cc:        Paul Traina <pst@freefall.freebsd.org>, freebsd-bugs@freefall.freebsd.org
Subject:   Re: kern/679 
Message-ID:  <199602081633.RAA01482@localhost>
In-Reply-To: <199602080413.MAA11493@jhome.DIALix.COM>
References:  <199602071717.JAA12960@freefall.freebsd.org> <199602080413.MAA11493@jhome.DIALix.COM>

next in thread | previous in thread | raw e-mail | index | archive | help
Peter Wemm writes:
>>Synopsis: chown(2) ignores set-user-id and set-group-id bits for root
>>This looks appropriate to me and is more in line with the UNIX standard.
>>Peter - would you care to verify my opinion before this is changed?
>
>I tend to agree.  "Root knows what he's doing" is a pretty fundamental
>design "feature".  

My brain-damaged sysop at work change the owner of a file hierarchy
with cron because solaris has no group quota. Everyone with
write access in this file system can break a account, just do 

$ cp /bin/sh .
$ chown u+s sh
[wait a hour for cron]
$ ./sh

>Paul Traina <pst@shockwave.com>
>Let's go with whatever POSIX says and be done with it.

I remember POSIX says for this special case 'undefined'. I think we
should made a compile option, e.g. CHOWN_COMPAT (SysV) or CHOWN_SECURE
(SCO).

Wolfram



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199602081633.RAA01482>