Re: Inquiry on the usage of CAdES in S/MIME structures

Dear Ayhan,

 

Archive format of CMS CAdES-A can be implemented with many versions of archive time-stamps.

 

Only archive time-stamp version 3 (ATSv3) is fixed for S/MIME usage.

The ATSv3 must include the ats-hash-index attribute which shall be carried as an unsigned attribute of the signature of the archive-time-stamp-v3 Attribute (see clause 6.4.3 ETSI TS 101 733 V2.2.1 (2013-04)). 

 

The ats-hash-index unsigned attribute provides an unambiguous imprint of the essential components of a CAdES signature for use in the archive time-stamp (see 6.4.3). These essential components are elements of the following ASN.1 SET OF structures: unsignedAttrs, SignedData.certificates, and SignedData.crls. 

 

The redundant CAdES attributes are not required to be included in CMS when ATSv3 is used e.g. now are used CMS fields like the SignedData.crls component as defined in RFC 3852 [4] which can include OCSP and/or CRL revocation information. 

Use of the ats-hash-index attribute makes it possible to add additional certificate / revocation information / unsigned attribute within SignedData.certificates/ SignedData.crls/ unsigned attributes of the CAdES signature (for instance counter signatures), after an archive time-stamp has been applied to a signature. Its use also allows the inclusion of components required by parallel signatures at a later time.

 

Previous archive time-stamp versions before version 3 have some problems e.g. with adding S/MIME unsigned attribute after archive time-stamp or problems with parallel CMS signatures when one of parallel signatures contains archive timestamp.

 

Archive timestamp version 3 is defined in ETSI TS 101 733 V2.2.1 (2013-04)

http://www.etsi.org/deliver/etsi_ts/101700_101799/101733/02.02.01_60/ts_101733v020201p.pdf 

 

Previous versions of CAdES specification have defined many redundant attributes.

The usage of redundant attributes was deprecated in the last profile.

 

CAdES Baseline Profile ETSI TS 103 173 V2.2.1 (2013-04)

http://www.etsi.org/deliver/etsi_ts/103100_103199/103173/02.02.01_60/ts_103173v020201p.pdf  

 

It means the use of the CAdES-A profile within the S/MIME structure is possible without know problems only when ATSv3 is included.

 

More information is from the page 15. It is about: ETSI ESI deprecates long-term formats

http://files.lockitin.webnode.sk/200000078-00104010a3/12th%20Edition%20of%20the%20Conference%20June%2004-06th%202012%20Poland.ppt 

 

Best regards,

Peter Rybar

National Security Authority

Information Security and Electronic Signature Department

Budatinska 30, 850 07 Bratislava 57, Slovak Republic

tel.: +421 2 6869 2163

mob.: +421 902 891 155

fax: +421 2 6869 1701

e-mail: [email protected]

e-mail: [email protected]

 

 

  _____  

From: [email protected] [mailto:[email protected]] On Behalf Of Ayhan Sehrin
Sent: Saturday, July 20, 2013 12:33 PM
To: [email protected]
Subject: [smime] Inquiry on the usage of CAdES in S/MIME structures
Importance: High

 

Dear list Members, 

 

As Verion Technology Group, a technology provider working on Registered E-Mail (REM) solutions since 2009, we are currently in the process of formulating a set of basic implementation principles and tools for REM Management Domain (MD)’s operative in Turkey. We need to resort to your experience and know-how on the subject of the S/MIME specification, and I was kindly prompted by Mr. Blake Ramsdell to address your list for further comments and insight.

The Registered E-Mail Regulatory Authority in Turkey has very recently mandated the use of the CAdES-A profile within the S/MIME structure and as we are a bit perplexed on the issue, we would like to inquire your comments on:

- whether it would be logical/feasible/meaningful to use the CAdES-A profile within an S/MIME structure (for some instances within the REM workflow CAdES-BES is also required)
- whether the above usage would present difficulties/problems in terms of interoperability of systems (i.e. e-mail clients or BPM/ERP systems, where the messages are expected to be utilised - e.g. a REM message received is downloaded and then used as the input of a business process within a core banking component) especially as regards the subsequent verification process within these systems and among different REM Service Providers in Turkey and in Europe (as Turkey has opted in to the ETSI TS 102640).
- whether, from a technical viewpoint, the implementation of an S/MIME structure with the usage of a CAdES signature presents a problem with the S/MIME RFC

- whether the S/MIME RFC should be expected to natively accomodate CAdES profiles in the near future 


We would gladly provide more details on the ETSI 102640 implementation in Turkey, if it is of essense to the above inquiries. 
We will be much obliged, should you be able to provide insight.

Thank you in advance and warm regards,


Ayhan ŞEHRİN
Man. Director
Verion Technology Group
Birlik Mah. 435. Cad. 403. Sok. No:3/3
Çankaya, Ankara - Turkey

 <tel:%2B90%20312%20496%2033%2016> +90 312 496 3316 (office)
 <tel:%2B90%20533%20556%203333> +90 533 556 3333 (mobile)

 

NB-REM implementation in Turkey in a tiny nutshell: In the typical REM implementation in Turkey, an original message is required to be signed with the sender’s Qualified Electronic Signature, two out of the currently three REMSPs provide tools for constructing and signing the sender’s S/MIME structure on the client, receive it over https (or over a web service for enterprise integration instances) to the REMSP servers. The third REMSP constructs the MIME message on its own server, hashes it an sends it for a CAdES signature to the sender’s client. Evidence is produced by REMSPs during the mail acceptance, delivery and retrieval milestones and is expected to be stored in WORM storage or to be archived, XadES-A for evidence, CAdES-A for all system logs.

_______________________________________________
smime mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/smime