LWN.net Logo

Security

OpenID Connect

By Jake Edge
June 2, 2010

When last we looked in on OpenID, it was close to finalizing the OpenID 2.0 specification. That was in 2007; since that time, various other identity management solutions have come about and have been more widely adopted, OAuth in particular. One of the architects of OpenID, David Recordon, has put out a idea (or "strawman" as he calls it) for a new API that combines the best of OpenID and OAuth into "OpenID Connect".

There are a number of shortcomings of OpenID that Recordon and others would like to see addressed. In the three years since the last revision, the internet has not stood still, but OpenID has. OpenID works reasonably well for web sites, but is much more difficult to use for things like desktop widgets or mobile applications. A bigger problem is that OpenID's user-centric nature has made those users less "valuable" to web sites, which results in fewer sites adopting OpenID.

OpenID is structured such that users need only share a limited amount of data (typically just a URL) with a site in order to register with it. That is good from a privacy perspective, but doesn't give site owners information that they want, like name, email address, photo, and so on. According to Chris Messina—who originated the OpenID Connect concept and name—that makes OpenID users into second-class citizens: "Because OpenID users share less information with third parties, they are perceived as being 'less valuable' than email-based registrants or users that connect to their Facebook or Twitter accounts."

More and more sites are farming out their identity management to big sites like Facebook and Twitter using OAuth. Messina and Recordon's idea is to reimplement OpenID atop OAuth 2.0, which would leverage that existing—widely adopted—API for identity management. It would also allow OpenID to become simpler for web site operators to implement. Recordon pointed out some of the problems he has heard about:

I've heard story after story from developers implementing OpenID 2.0 who don't understand why it is so complex and inevitably forgot to do something. With OpenID Connect, discovery no longer takes over 3,000 lines of PHP to implement correctly. Because it's built on top of OAuth 2.0, the whole spec is fairly short and technology easy to understand. Building on OAuth provides amazing side benefits such as potentially being the first version of OpenID to work natively with desktop applications and even on mobile phones.

Based on that, one might wonder why OpenID doesn't just adopt OAuth, rather than build atop it, but there is an important distinction between the two. OpenID Connect would still decentralize the storage of user information and allow the user-centric nature of OpenID to survive. Users would be able to choose their provider or run their own that stored their personal information. That way, users would get to choose whom to trust or to only trust their own server.

Another problem that OpenID Connect hopes to solve is to simplify things for users. Right now, users have to remember and type in a URL that corresponds to their OpenID provider, or click on multiple buttons for popular providers (which leads to the so-called NASCAR problem where there are multiple logos as buttons). OpenID Connect would allow for simpler URLs or even email addresses as identifiers.

It is important to recognize that this proposal is being driven by the fact that OpenID adoption has largely stalled. That has happened because the sites that folks want to use want a little—or a lot—more information about those who are signing up for or using their services. There is a trade off, clearly, as it is not unreasonable for site owners to require more information as a kind of payment, as long as they are up front about it. But the privacy conscious are likely to still be marginalized as the demands for information increase.

While there currently is a lot of noise being made about privacy concerns for sites like Facebook, there appears to be little actual action about it by most users. Privacy just does not seem to be something that is high on most users' priority lists, or, perhaps, Scott McNealy was right: "You have zero privacy anyway ... Get over it." OpenID Connect seems like a reasonable idea, overall, but as long as the majority are happy with the current OAuth-based systems, it is a little hard to see it making any headway. Yes, it may be used by a small minority of internet users, but it is likely to require just enough effort that most will not take advantage of it. It would seem that many are already "over it".

Comments (4 posted)

Brief items

Quote of the week

The activist siphoned more than a million documents as they traveled across the internet through Tor, also known as "The Onion Router," a sophisticated privacy tool that lets users navigate and send documents through the internet anonymously.

The siphoned documents, supposedly stolen by Chinese hackers or spies who were using the Tor network to transmit the data, were the basis for Wikileaks founder Julian Assange's assertion in 2006 that his organization had already "received over one million documents from 13 countries" before his site was launched, according to the article in The New Yorker.

-- Wired

Comments (3 posted)

New vulnerabilities

clamav: multiple vulnerabilities

Package(s):clamav CVE #(s):CVE-2010-1639 CVE-2010-1640
Created:May 27, 2010 Updated:March 14, 2011
Description:

From the Mandriva advisory:

The cli_pdf function in libclamav/pdf.c in ClamAV before 0.96.1 allows remote attackers to cause a denial of service (crash) via a malformed PDF file, related to an inconsistency in the calculated stream length and the real stream length (CVE-2010-1639).

Off-by-one error in the parseicon function in libclamav/pe_icons.c in ClamAV 0.96 allows remote attackers to cause a denial of service (crash) via a crafted PE icon that triggers an out-of-bounds read, related to improper rounding during scaling (CVE-2010-1640).

Alerts:
Fedora FEDORA-2011-2741 2011-03-05
Fedora FEDORA-2011-2743 2011-03-05
Gentoo 201009-06 2010-09-07
SUSE SUSE-SR:2010:014 2010-08-02
Ubuntu USN-945-1 2010-05-27
Mandriva MDVSA-2010:110 2010-05-27
Pardus 2010-72 2010-06-04
openSUSE openSUSE-SU-2010:0414-1 2010-07-22

Comments (none posted)

kdenetwork: arbitrary code execution

Package(s):kdeaccessibility CVE #(s):CVE-2010-1511
Created:May 27, 2010 Updated:April 21, 2011
Description:

From the KDE advisory:

In some versions of KGet (2.4.2) a dialog box is displayed allowing the user to choose the file to download out of the options offered by the metalink file. However, KGet will simply go ahead and start the download after some time - even without prior acknowledgment of the user, and overwriting already-existing files of the same name. (CVE-2010-1511)

Alerts:
Fedora FEDORA-2011-5211 2011-04-12
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Pardus 2010-68 2010-06-04
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15
Fedora FEDORA-2010-8577 2010-05-15
Fedora FEDORA-2010-8544 2010-05-15
Fedora FEDORA-2010-8547 2010-05-15

Comments (2 posted)

kernel: unexpected file access in btrfs

Package(s):kernel CVE #(s):CVE-2010-1636
Created:May 31, 2010 Updated:September 23, 2010
Description: From the Red Hat bugzilla:

The existing [btrfs] code would have allowed you to clone a file that was only open for writing. Not an expected behaviour.

Alerts:
openSUSE openSUSE-SU-2010:0664-1 2010-09-23
Ubuntu USN-966-1 2010-08-04
Fedora FEDORA-2010-9183 2010-05-28
Pardus 2010-94 2010-07-08
Fedora FEDORA-2010-9209 2010-05-28

Comments (none posted)

perl-POE-Component-IRC: arbitrary IRC command execution

Package(s):perl-POE-Component-IRC CVE #(s):
Created:May 31, 2010 Updated:June 2, 2010
Description: From the Red Hat bugzilla:

A vulnerability was reported to Debian for POE::Component::IRC, where it did not remove carriage returns and line feeds. This affects tools or IRC bots using the perl module, and can be used to execute arbitrary IRC commands by passing an argument such as "some text\rQUIT" to the 'privmsg' handler, which would cause the client to disconnect from the server.

Alerts:
Fedora FEDORA-2010-8911 2010-05-22
Fedora FEDORA-2010-8904 2010-05-22

Comments (none posted)

rhn-client-tools: privilege escalation

Package(s):rhn-client-tools CVE #(s):CVE-2010-1439
Created:June 2, 2010 Updated:June 2, 2010
Description: The rhn-client-tools utilities fail to set secure permissions on the loginAugh.pkl file, allowing local users to manipulate it. The result can be unwanted package downloads or the manipulation of action lists associated with the system's profile.
Alerts:
Red Hat RHSA-2010:0449-01 2010-06-01

Comments (none posted)

transmission: denial of service

Package(s):transmission CVE #(s):CVE-2010-1853
Created:June 1, 2010 Updated:June 2, 2010
Description: From the Gentoo advisory:

Multiple stack-based buffer overflows in the tr_magnetParse() function in libtransmission/magnet.c have been discovered. A remote attacker could cause a Denial of Service or possibly execute arbitrary code via a crafted magnet URL with a large number of tr or ws links.

Alerts:
Gentoo 201006-06 2010-06-01

Comments (none posted)

Page editor: Jake Edge
Next page: Kernel development>>

Copyright © 2010, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds