Bug 1081307 – VUL-1: CVE-2018-7174: xpdf: An infinite loop in XRef::Xref allows an attacker to cause denial of service because loop detection exists only for tables, not streams.

Bug 1081307 - (CVE-2018-7174) VUL-1: CVE-2018-7174: xpdf: An infinite loop in XRef::Xref allows an attacker to cause denial of service because loop detection exists only for tables, not streams.

(CVE-2018-7174)
Summary:	VUL-1: CVE-2018-7174: xpdf: An infinite loop in XRef::Xref allows an attacker...

Status:	NEW

Classification:	Novell Products
Product:	SUSE Security Incidents
Classification:	Novell Products
Component:	Incidents
Version:	unspecified
Hardware:	Other Other

Priority:	P3 - Medium Severity: Normal
Target Milestone:	---
Assigned To:	Peter Simons
QA Contact:	Security Team bot

URL:	https://smash.suse.de/issue/200310/
Whiteboard:	CVSSv3:SUSE:CVE-2018-7174:5.5:(AV:L/A...
Keywords:

Depends on:
Blocks:	1133493
	Show dependency tree / graph

Create test case

Clone This Bug

Reported:	2018-02-16 09:01 UTC by Karol Babioch
Modified:	2019-12-09 15:51 UTC (History)
CC List:	1 user (show)

See Also:
Found By:	Security Response Team
Services Priority:
Business Priority:
Blocker:	---
Marketing QA Status:	---
IT Deployment:	---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Karol Babioch 2018-02-16 09:01:09 UTC

CVE-2018-7174

An issue was discovered in xpdf 4.00. An infinite loop in XRef::Xref allows an
attacker to cause denial of service because loop detection exists only for
tables, not streams.

References:
https://bugzilla.redhat.com/show_bug.cgi?id=1546052
http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2018-7174
http://www.cvedetails.com/cve/CVE-2018-7174/
https://forum.xpdfreader.com/viewtopic.php?f=3&t=605
https://forum.xpdfreader.com/viewtopic.php?f=3&amp;t=605

Comment 1 Peter Simons 2018-06-21 09:20:04 UTC

No fix was published for this issue. Upstream says they fixed it in their own source code but they did not share the actual diff. There is a vague description saying

> The XRef class has loop detection code, but it was only 
> checking xref tables, and not xref streams. The fix was
> simple: just move the detection code from XRef::readXRefTable()
> to XRef::readXRef().

but I am not sure how to translate that into an actually applicable patch. Red Hat doesn't have a fix either yet; both of their tracker bugs remain open.