W3C home > Mailing lists > Public > public-xg-webid@w3.org > December 2011

Re: ExplorerKeygen - keygen element for IE

From: Henry Story <henry.story@bblfish.net>
Date: Wed, 7 Dec 2011 14:47:57 +0100
Cc: public-xg-webid@w3.org
Message-Id: <1E5A4CAF-76E4-4DC6-BE87-911B95A3BFEE@bblfish.net>
To: Kingsley Idehen <kidehen@openlinksw.com>

On 7 Dec 2011, at 14:07, Kingsley Idehen wrote:

> On 12/7/11 7:47 AM, Henry Story wrote:
>> 
>> 
>> On 7 Dec 2011, at 13:15, Kingsley Idehen wrote:
>> 
>>> On 12/7/11 4:36 AM, Henry Story wrote:
>>>> 
>>>> 
>>>> On 7 Dec 2011, at 01:36, Peter Williams wrote:
>>>> 
>>>>>  
>>>>>  
>>>>> My work was about running a validation agent on IIS.
>>>>>  
>>>>> Peter argued that IIS could not process a client certificate (self-signed or otherwise) that was not rooted on the windows host, on which IIS (or other native https server app) is listening.
>>>> 
>>>> Ah good. So originally when I was arguing that Kingsley had misread you I was right. You could have supported my view at that time.
>>>> As you were silent I started taking on Kingsley's position.
>>> 
>>> IIS and IE are sort of doing the same thing. I am saying that IE will not work with Certs. that aren't signed by a Cert. rooted on its host machine or a network server (depending on your Windows network setup). 
>> 
>>> Not working means that the TLS handshake will fail.
>> 
>> Ok, so you are saying with Peter Williams that unless IIS Trusts the Signer of the Certificate it receives, the handshake will fail. So IIS would need to know the Singing authority of every web site.
> 
> No, it just needs a CA trust chain. Thus, you can make you own by generating a CA cert. and storing it in the keystore as a new CA root. Basically, you end up with the default CA trust chain that comes with Windows plus new chains you construct yourself via your own root Certs. 

yes, but then your IIS instance would only be able to accept Certificates signed by yourself right?  Since you are the one in possession of the new CA root you just stored in your keystone. 

In particular Peter Williams is asserting that your IIS instance would not be able to accept a certificate signed by the CA Root I created and stored in my keystore, not that is without having as a policy to add any CA you find automatically to your trust store. The danger of doing that is that unless you change the logic of how all windows apps works with trust stores, you would be essentially disabling all of Windows security by doing that. An evil person ( Dr Evil ) could just create a new CA certificate, sign a WebID certificate with it, authenticate on your IIS with the WebID Cert, and then know that its root CA would be in your trust store. All the other pieces of Windows software that would then automatically rely on that CA certificate  Dr Evil had generated and trust every certificate Dr Evil signed.

In the end this is not a big problem, because people can use Apache, Java or other web servers that are less tied to the kernel, and wait for Windows to update its logic which can only be when WebId is very widely used, as that will be enough to justify the expense required of going through the work.

> 
> This is all about you taking control of your machine or having your machine adhere to trust chains in a Windows network which can even take the form of a forest with many domain controllers etc..
> 
> From a WebID perspective the impact area boils down to IdPs being able to produce a signing cert. (i.e., CA cert) and a signed certificate. You can be the notary and the certificate signer re. Windows.
> 
>> If that is the case then until someone provides a patch for IIS to allow it to work with Null Trust Authorities, then IIS cannot use WebID.
> 
> Of course it can. It just requires the steps above.
> 
>> I am sure if we can do this with Apache and Java they can do it to, but since they built it into the kernel - against the arguments of various leading technologists - things are going to be more difficult to change there. Which is not that much of a problem for us now.
>> 
>> 
>>> As I said, in our brief chat yesterday, I am about to revisit all the scenarios on Windows re. IE. 
>> 
>> There is the IIS server issue discussed above, and the IE Client Certificate issue discussed by Bergi below. These are two different issues I think. Both of them merit attention.
> 
> I just need time to make updated demos and notes. This will come when we are done with iOS5 and Mac OS X variants of what we are doing with Facebook, Twitter, LinkedIn etc..

Ok, What is the best  screen cast you have of your solution? 

Henry

> 
> Kingsley
>>> 
>>>> 
>>>>>  
>>>>> Does anyone claim this argument is invalid?  A simple contradiction claim is fine. I would LIKE to do this, so we can say windows can 100% implement webid. My proof is only 99% (since my validation authority cannot process abitary client  certs, without an act of pre-registration)
>>>> 
>>>> So I think this is a good challenge for Windows Security specialists. Perhaps we should have a challenges section of the wiki.
>>>> There would be this challenge, and the other one would be: how to get Bergi's keygen script to not require users to go and set registry values if their machine is set up for tight security.
>>> 
>>> We seek a one-click solution triggered by a link for making Certs on Windows that terminates with Cert. storage in the native Windows keystore. Then we need to be able to test against any WebID verifier. This is what our Wizard has delivered since its inception more than a year ago. 
>> 
>> So good we then have two solutions. The solution of using the built in ActiveX component that Bruno Harbulot told us about 2 years ago, and your client solution. We should compare them for usability, and see where each one shines.
>> 
>> It's good that we have 2 options.
>> 
>> I think you did a screen cast. Perhaps bergi can do a screen cast of creating a certificate on Windows 7.
>> 
>> Henry
>> 
>>> 
>>> Kingsley
>>>> 
>>>>> ..
>>>>>  
>>>>> Getting IE to perform https client authn and release a supporting client cert to some non-native WIndows webserver (e.g. tomcat running on a windows socket) is a matter about which others made claims. They can stand by them, or not.
>>>>>  
>>>>> ---------
>>>>>  
>>>>> I spent some time this week playing with sip URIs and TLS used in SRTP (and ZImmermans alternative secure transports). Also got to play with cisco phones that use CTLs and certs, binding them to particular call agents, and using particular validation protocols during call processing and multi-media channel setup. More later. It will be fun to see how the semantic web might cooperate with the world of SIP URIs, in running webid-style trust domains for the key management protocols that might complement those used today, supporting SRTP.
>>>>>  
>>>>> anyway, back to research vs programming.
>>>>>  
>>>>> > Date: Tue, 6 Dec 2011 22:49:17 +0100
>>>>> > From: bergi@axolotlfarm.org
>>>>> > To: henry.story@bblfish.net
>>>>> > CC: public-xg-webid@w3.org
>>>>> > Subject: Re: ExplorerKeygen - keygen element for IE
>>>>> > 
>>>>> > Am 06.12.2011 10:42, schrieb Henry Story:
>>>>> > > Great work Bergi!
>>>>> > > 
>>>>> > > Were you able to create a certificate with this from Internet Explorer and then
>>>>> > > log into fcns.eu? Peter Williams declared this was impossible to do last week.
>>>>> > 
>>>>> > Sure. I only tested my own endpoint, but that shouldn't matter.
>>>>> > 
>>>>> > > 
>>>>> > > I think you should definitively copy and paste this e-mail into a wiki page
>>>>> > > linked to from our new HOWTO page. This looks like the place to do ti from
>>>>> > > 
>>>>> > > http://www.w3.org/2005/Incubator/webid/wiki/Creating_Certificates
>>>>> > 
>>>>> > I added a Internet Explorer section.
>>>>> > 
>>>>> > I would be nice if someone with a English version of Windows could add
>>>>> > some screenshots, especially for the "The drawback of this solution"
>>>>> > section to show people how to enable this component.
>>>>> > 
>>>>> > > 
>>>>> > > 
>>>>> > > 
>>>>> > > On 6 Dec 2011, at 00:04, bergi wrote:
>>>>> > > 
>>>>> > >> Internet Explorer doesn't support the keygen element out of the box. The
>>>>> > >> only way to generate certificate request in the browser is the
>>>>> > >> X509Enrollment ActiveX component. I've written some JavaScript code
>>>>> > >> which brings nearly full keygen compatibility to IE. It's based on
>>>>> > >> IEKeygen.js Bruno Harbulot wrote for Clerezza, but it's a little bit
>>>>> > >> more generic.
>>>>> > > 
>>>>> > > very nice.
>>>>> > > 
>>>>> > >>
>>>>> > >> What must be changed:
>>>>> > >> It should require just a conditional include on the client side:
>>>>> > >> <!--[if IE]>
>>>>> > >> <script type="text/javascript" src="explorer-keygen.js"></script>
>>>>> > >> <![endif]-->
>>>>> > >> On the server side PKCS10 support must be added, which is in our case
>>>>> > >> more or less just a different packaging of the public key. I'm using
>>>>> > >> OpenSSL in my PHP code. If you look at the function
>>>>> > >> buildCertificateSpkac and buildCertificatePkcs10 in
>>>>> > >> OpenSslCertificateBuilder.php you will see it's nearly the same code.
>>>>> > >>
>>>>> > >> The drawback of this solution:
>>>>> > >> Microsoft doesn't trust it's own ActivceX components. This means the
>>>>> > >> page must be in the trusted zone or the user has to change
>>>>> > >> initialization of untrusted ActiveX components settings from disabled to
>>>>> > >> ask.
>>>>> > > 
>>>>> > > I think this is the case for the Windows 7 only. I think I tried this a 
>>>>> > > year ago on some other windows and it did not ask me for all this. 
>>>>> > > It will be interesting to have people try this out themselves, and 
>>>>> > > send us feedback.
>>>>> > 
>>>>> > I also added a note on the wiki page.
>>>>> > 
>>>>> > > 
>>>>> > >>
>>>>> > >> A little bit more in detail what the JavaScript code does:
>>>>> > >> On page load it searches for a keygen element and adds a combobox for
>>>>> > >> the key length selection after the keygen element to the DOM. The key
>>>>> > >> length will be written to the keylength attribute in the keygen element.
>>>>> > > 
>>>>> > > I suppose that is to imitate the way keygen works. I did not check but
>>>>> > > does keygen really send the key length in the form to the server, or is
>>>>> > > it not just used to create the public key?
>>>>> > 
>>>>> > Yes, it's to imitate the keygen behavior of other browsers. The combobox
>>>>> > itself doesn't even get a name attribute, which makes it invisible to
>>>>> > the form and the .serialize() function of jQuery.
>>>>> > 
>>>>> > > 
>>>>> > >> Also the action attribute in the form element gets renamed to ekaction
>>>>> > >> to avoid submitting the form. The submit button is replaced with another
>>>>> > >> button that calls some JavaScript code. If the newly created button is
>>>>> > >> pressed, the JavaScript code will call the ActiveX component and create
>>>>> > >> a new certificate signing request. For the CSR a new hidden input field
>>>>> > >> will be created. The jQuery .serialize() function is used to get the
>>>>> > >> form data in www-form-urlencoded format and Ajax is used to send the
>>>>> > >> data to the server. Than the response is forwarded to the ActiveX
>>>>> > >> component. And finally the certificate is installed in the Windows Keystore.
>>>>> > > 
>>>>> > > very nice!
>>>>> > > 
>>>>> > > 
>>>>> > >>
>>>>> > >> The JavaScript code is MIT licensed, the PHP code GPL 3.
>>>>> > > 
>>>>> > > 
>>>>> > > 
>>>>> > >>
>>>>> > >> Link to the SVN repo:
>>>>> > >> https://www.axolotlfarm.org/svn/bergi/bergnet/php/certbuilder/trunk/
>>>>> > >>
>>>>> > > 
>>>>> > > Social Web Architect
>>>>> > > http://bblfish.net/
>>>>> > > 
>>>>> > > 
>>>>> >
>>>> 
>>>> Social Web Architect
>>>> http://bblfish.net/
>>>> 
>>> 
>>> 
>>> -- 
>>> 
>>> Regards,
>>> 
>>> Kingsley Idehen	      
>>> Founder & CEO 
>>> OpenLink Software     
>>> Company Web: http://www.openlinksw.com
>>> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
>>> Twitter/Identi.ca handle: @kidehen
>>> Google+ Profile: https://plus.google.com/112399767740508618350/about
>>> LinkedIn Profile: http://www.linkedin.com/in/kidehen
>>> 
>>> 
>>> 
>>> 
>> 
>> Social Web Architect
>> http://bblfish.net/
>> 
> 
> 
> -- 
> 
> Regards,
> 
> Kingsley Idehen	      
> Founder & CEO 
> OpenLink Software     
> Company Web: http://www.openlinksw.com
> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
> Twitter/Identi.ca handle: @kidehen
> Google+ Profile: https://plus.google.com/112399767740508618350/about
> LinkedIn Profile: http://www.linkedin.com/in/kidehen
> 
> 
> 
> 

Social Web Architect
http://bblfish.net/
Received on Wednesday, 7 December 2011 13:48:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 December 2011 13:48:44 GMT