RE: [saml-dev] X509SubjectName and relationship to RFC2253?

From
Hal Lockhart <>
Date
2002-04-24T14:40:39+00:00
ID
Thread
RE: [saml-dev] X509SubjectName and relationship to RFC2253?
Title: RE: [saml-dev] X509SubjectName and relationship to RFC2253?

 

> The core-00 version of SAML 1.0 describes the format attribute,

> #X509SubjectName , 

> as follows:

> 

>        Indicates that the content of the NameIdentifier element is in

> the form specified for 

>        the contents of        <ds:X509SubjectName> element in [DSIG].

> Implementors should 

>        note that [DSIG] specifies encoding rules for X.509 

> subject names

> that **differ** 

>        from the rules given in RFC2253 [RFC2253]. 

> 

> 

> There is no [DSIG] reference in this SAML document's bibliography. The

> closest I have found is [XMLSig].

That is correct. You found a bug, I will pass it along to the document editor.

 

> Is that the proper one? When I follow that link I end up at RFC3275

> which seems to contradict the SAML

> text:

> 

>         ... The X509SubjectName element, which contains an 

> X.509 subject

> distinguished name 

>         that SHOULD be compliant with RFC 2253 [LDAP-DN] ...

We have it on good authority that this statement is not completely true, alhough it may hold for many common cases. 

> 

> Can someone explain what implied differences would exist in encoding

> between SAML/DSIG and RFC2253

> for this field?  Most importantly --- can I use this field 

> for LDAP DNs,

> or should I add an additional

> format tag (such as #RFC2253DistinguishedName)?

The differences as I understand them  have to do with the escaping of international (multi-byte) characters. The details can be found by a careful reading of the two specifications. 

There is a discussion thread in the SAML archives entitled "Proposed text for <NameIdentifier> in early March of this year, that discuses this issue. It begins with this messsage: 

http://lists.oasis-open.org/archives/security-services/200203/msg00002.html 

Look especially at the messages from Stephen Farrell.

Hal