[OASIS Issue Tracker] (EBXMLMSG-104) PMode[1].Security.X509.Encryption.MinimumStrength

[OASIS Issues Tracker <workgroup_mailer@lists.oasis-open.org>]

    [ https://issues.oasis-open.org/browse/EBXMLMSG-104?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=66288#comment-66288 ] 

Theo Kramer commented on EBXMLMSG-104:
--------------------------------------

Agreed on point 3 that the sender should use the value specified by the PMode[1].Security.X509.Encryption.Algorithm with the following addition

A receiver may optionally accept or reject any algorithm that does not conform to the specified PMode[1].Security.X509.Encryption.Algorithm

> PMode[1].Security.X509.Encryption.MinimumStrength
> -------------------------------------------------
>
>                 Key: EBXMLMSG-104
>                 URL: https://issues.oasis-open.org/browse/EBXMLMSG-104
>             Project: OASIS ebXML Messaging Services TC
>          Issue Type: Bug
>          Components: Core Spec
>            Reporter: Pim van der Eijk
>
> This parameter is inherited from CPPA2,  and its definition is taken from there.  It has several problems:
> 1) the paragraph assumes more effective bits are better,  but this depends on key/algorithm type.  A short EC key can be more effective than a long RSA key.
> 2) Algorithms typically define the key size. In that case the bit size is determined by the PMode[1].Security.X509.Encryption.Algorithm.  The bit size is not an independently selectable parameter in an algorithm.  Algorithms differ not just on key size.  E.g. the differences between AES 125, 192 and 256 are not just key size.  
> 3) A P-Mode is like an agreement.  In an agreement parties specify what they agree to use,  e.g. AES128,  not what they agree to use minimally. If the agreement is AES128, the sender should not use AES256.



--
This message was sent by Atlassian JIRA
(v6.2.2#6258)