oss-sec mailing list archives

Re: CVE-request: Roundcube XSS issues


From: Kurt Seifried <kseifried () redhat com>
Date: Mon, 20 Aug 2012 10:11:28 -0600

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



On 08/20/2012 05:24 AM, Jan Lieskovsky wrote:
Hi Jon,

this is due the recent roundcubemail XSS issues post: [1] 
http://www.openwall.com/lists/oss-security/2012/08/20/2

which detailed leads into: 1, issue #1 New larry skin & literal in 
Subject header display Ticket: 
http://trac.roundcube.net/ticket/1488519 Upstream patch:

http://trac.roundcube.net/changeset/a7d5e3e8580466639a18da35af13b97dc3765c16/github

Upon code review, I don't think this issue affects 0.7.x versions, 
we ship in Fedora and EPEL (iilc the Larry skin was introduced
only in 0.8.x version and in 0.7.x version the related code looks 
different). I don't have filed RH bug for this based on the above. 
Could you have a look and confirm this?

Please use CVE-2012-3507 for this issue.


2, Issue 2a: Description: Stored XSS in e-mail body. Ticket: 
http://trac.roundcube.net/ticket/1488613 Upstream patch:

https://github.com/roundcube/roundcubemail/commit/5ef8e4ad9d3ee8689d2b83750aa65395b7cd59ee

Upon code review doesn't seem to affect rcmail we ship in Fedora / 
EPEL -> haven't filed RH bug for it. Could you double-check and 
confirm that?,

Issue 2b: Self XSS in e-mail body (Signature). Ticket: 
http://trac.roundcube.net/ticket/1488613 Upstream patch:

https://github.com/roundcube/roundcubemail/commit/c086978f6a91eacb339fd2976202fca9dad2ef32

The 'program/js/app.js' rcube_webmail() upstream change from the 
patch above seems to be applicable to Fedora / EPEL rcmail 
versions. Thus I have filed: 
https://bugzilla.redhat.com/show_bug.cgi?id=849615

to track this. But not sure whole 'Self XSS in e-mail body 
(Signature).' upstream patch would apply with its logic to 0.7.x 
versions: https://bugzilla.redhat.com/show_bug.cgi?id=849615#c3

Therefore this needs review by someone more familiar with 
rcube_webmail() routine code to decide if apply that patch or not. 
Could you do that?

Please use CVE-2012-3508 for these two issues (same version, same type
of vuln so cve merge).

Thank you && Regards, Jan. -- Jan iankko Lieskovsky / Red Hat 
Security Response Team


- -- 
Kurt Seifried Red Hat Security Response Team (SRT)
PGP: 0x5E267993 A90B F995 7350 148F 66BF 7554 160D 4553 5E26 7993

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=wELK
-----END PGP SIGNATURE-----


Current thread: