Bug 1381526 (CVE-2016-7798)

Summary:	CVE-2016-7798 ruby: IV Reuse in GCM Mode
Product:	[Other] Security Response	Reporter:	Andrej Nemec <anemec>
Component:	vulnerability	Assignee:	Red Hat Product Security <security-response-team>
Status:	CLOSED WONTFIX	QA Contact:
Severity:	medium	Docs Contact:
Priority:	medium
Version:	unspecified	CC:	cpelland, dajohnso, gblomqui, gmccullo, gtanzill, hhorak, hhudgeon, jfrey, jhardy, jorton, jprause, mtasaka, obarenbo, roliveri, ruby-maint, simaishi, s, strzibny, tagoh, vanmeeuwen+fedora, vondruch
Target Milestone:	---	Keywords:	Security
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-06-08 02:59:22 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1381527
Bug Blocks:	1381528

Description Andrej Nemec 2016-10-04 11:25:32 UTC

An IV reuse bug was discovered in Ruby's OpenSSL library when using
aes-gcm. When encrypting data with aes-*-gcm, if the IV is set before
setting the key, the cipher will default to using a static IV. This creates
a static nonce and since aes-gcm is a stream cipher, this can lead to known
cryptographic issues.

References:

http://seclists.org/oss-sec/2016/q3/562

Upstream bug:

https://github.com/ruby/openssl/issues/49

Upstream patch:

https://github.com/ruby/openssl/commit/8108e0a6db133f3375608303fdd2083eb5115062

Comment 1 Andrej Nemec 2016-10-04 11:26:13 UTC

Created ruby tracking bugs for this issue:

Affects: fedora-all [bug 1381527]

Comment 2 Huzaifa S. Sidhpurwala 2016-11-28 07:44:17 UTC

Analysis:

As explained in https://github.com/ruby/openssl/issues/49#issuecomment-248171371
Calling cipher.key after calling cipher.iv zeros out the value of the iv, resulting in a possibly weak cryptographic operation. 


Mitigation:

A possible workaround to this flaw is, when using aes-256-gcm mode, always set the key first and then the iv. For example when setting random keys and iv use the following code segment:

key = cipher.random_key
iv = cipher.random_iv