OASIS Key Management Interoperability Protocol (KMIP) TC

 View Only
  • 1.  Groups - Cryptographic Length of Certificates Proposal (KMIPCertificateLengthProposal.doc) uploaded

    Posted 07-24-2011 20:30
    The document named Cryptographic Length of Certificates Proposal (KMIPCertificateLengthProposal.doc) has been submitted by Ms. Judith Furlong to the OASIS Key Management Interoperability Protocol (KMIP) TC document repository. Document Description: Proposal for changing the KMIP specification wording to clarify what value should be used for the cryptographic length of a certificate.

    This addresses topic 3 of the open PKI itmes. View Document Details: http://www.oasis-open.org/committees/document.php?document_id=42985 Download Document: http://www.oasis-open.org/committees/download.php/42985/KMIPCertificateLengthProposal.doc PLEASE NOTE: If the above links do not work for you, your email application may be breaking the link into two pieces. You may be able to copy and paste the entire link address into the address field of your web browser. -OASIS Open Administration


  • 2.  Re: [kmip] Groups - Cryptographic Length of Certificates Proposal (KMIPCertificateLengthProposal.doc) uploaded

    Posted 07-24-2011 22:14
    X-NONE X-NONE MicrosoftInternetExplorer4 For keys, Cryptographic Length is the length in bits of the clear-text cryptographic key material of the Managed Cryptographic Object. For certificates, Cryptographic Length is the length in bits of the encoded Certificate Managed Cryptographic Object. There are three issues with this: - one the definition for keys is actually interpreted differently by different implementations for DES keys - does this include or not include the parity bits?   (I know this isn't part of Judy's change - but it is something to get sorted out if we are changing this section) - what is the purpose of making this length be specified in bits? - what purpose does it serve to have this just specified as the length of the encoded certificate - that is not cryptographic - and the certificate value has a length in its encoding. I can see a use in this matching Cryptographic Length of the public key contained with in the certificate. Tim.


  • 3.  RE: [kmip] Groups - Cryptographic Length of Certificates Proposal (KMIPCertificateLengthProposal.doc) uploaded

    Posted 07-25-2011 19:58
    The alternative proposal from what I posted yesterday would be to define a new attribute -- Certificate Length -- and have that carry the length of the certificate.   Then Cryptographic Length would then be restricted to only apply to keys.      So there would be no Cryptographic Length associated with the Certificate, but there would be a Cryptographic Length associated with the Public Key associated (linked) with the Certificate.   I went with the simpler approach of just trying to repurpose the Cryptographic Length for both Certificates and Keys, but can submit the alternative proposal if that is what the TC wants.   We can discuss this on the Thursday call.   Other responses to your comments below ?¦.   Judy   From: Tim Hudson [mailto:tjh@cryptsoft.com] Sent: Sunday, July 24, 2011 6:14 PM To: Furlong, Judith Cc: kmip@lists.oasis-open.org Subject: Re: [kmip] Groups - Cryptographic Length of Certificates Proposal (KMIPCertificateLengthProposal.doc) uploaded   " For keys, Cryptographic Length is the length in bits of the clear-text cryptographic key material of the Managed Cryptographic Object. For certificates, Cryptographic Length is the length in bits of the encoded Certificate Managed Cryptographic Object. " There are three issues with this: - one the definition for keys is actually interpreted differently by different implementations for DES keys - does this include or not include the parity bits?   (I know this isn't part of Judy's change - but it is something to get sorted out if we are changing this section) [JAF] Ok sounds like we need additional clarification of Cryptographic Length - what is the purpose of making this length be specified in bits? [JAF] What other unit of length would you suggest?   Bytes? Or leaving the Length unit unspecified? - what purpose does it serve to have this just specified as the length of the encoded certificate - that is not "cryptographic" - and the certificate value has a length in its encoding. [JAF] KMIP defines Certificates as Cryptographic Managed Objects.   I do agree that the certificate contains more than cryptographic materials (e.g. the public key and the signature) but I wouldn't say that a certificate is not cryptographic at all. Yes the certificate has a length within the encoding but KMIP doesn ??t not rely upon info inside the encoded managed object --- it has be to parsed are pulled out an placed in an attribute.   So if we want the length of the certificate they we need to pull that out and specify it. I can see a use in this matching "Cryptographic Length" of the public key contained with in the certificate. [JAF] The Public Key will be linked to the Certificate and that will have the Cryptographic Length of the key Tim.


  • 4.  Re: [kmip] Groups - Cryptographic Length of Certificates Proposal(KMIPCertificateLengthProposal.doc) uploaded

    Posted 07-25-2011 22:03
    > We can discuss this on the Thursday call. That's why I sent the feedback to the list - so TC members have an opportunity to consider the topic and respond to the list prior to the call ... or discuss on the call. >> I can see a use in this matching Cryptographic Length of the public key contained with in the certificate. > [JAF] The Public Key will be linked to the Certificate and that will have the Cryptographic Length of the key The public key may or may not be present as a managed object so it cannot be relied on being present. If we want to support the concept of operations like locate-all-certificates-that-use-2048-bit-public-keys (not an unreasonable request to expect to be able to perform) then the information needs to be represented directly as an attribute on the Certificate managed object. Tim