Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20120831182332.GA2125@jwilk.net>
Date: Fri, 31 Aug 2012 20:23:32 +0200
From: Jakub Wilk <jwilk@...ian.org>
To: oss-security@...ts.openwall.com
Subject: Re: [Notification] CVE-2012-3500 - rpmdevtools,
 devscripts: TOCTOU race condition in annotate-output

* Jan Lieskovsky <jlieskov@...hat.com>, 2012-08-31, 11:22:
>A TOCTOU race condition was found in the way 'annotate-output' (used to 
>execute a program annotating the output linewise with time and stream) 
>tool of rpmdevtools, a suite of scripts and (X)Emacs support files to 
>aid in development of RPM packages, performed management of its 
>temporary files used for standard output and standard error output. A 
>local attacker could use this flaw to conduct symbolic link attacks, 
>possibly leading to their ability in an unauthorized way to alter files 
>belonging to the user running the 'annotate-output' tool.

The vulnerable code appears to be:

OUT=`mktemp --tmpdir annotate.XXXXXX` || exit 1
ERR=`mktemp --tmpdir annotate.XXXXXX` || exit 1
rm -f $OUT $ERR
mkfifo $OUT $ERR || exit 1

But mkfifo will never create a FIFO over a symlink; the underlying 
library function fails with EEXISTS when "pathname already exists. This 
includes the case where pathname is a symbolic link, dangling or not." 
So AFAICS it's just a DoS, not something giving the attacker "ability in 
an unauthorized way to alter files".

-- 
Jakub Wilk

Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.