OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

kmip message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [kmip] KMIP :: Doubts



Hi Trinath, please see below.
Regards,
-Robert

Somanchi Trinath-B22327 <B22327@freescale.com> wrote on 09/29/2010 06:39:50 PM:
>
> RE: [kmip] KMIP :: Doubts

>
> Somanchi Trinath-B22327

>
> to:

>
> Robert Haas

>
> 09/29/2010 06:43 PM

>
> Cc:

>
> kmip

>
> Some issue with my mail client, So resending this mail again, Ignore
> if already received this email
>
> Thank you
>
>
> Hi Robert,
>
> Thank you for the reply.
>
> My 3rd doubt is still not cleared. Use case 3.1.3 from the use case
> document uses UUID in locate request. I think I was not clear in
> expressing my doubt.
>
> Here is my Doubt:
>
> In Use case 3.1.3, Locate request, UUID is given in the request
> payload in an Attribute Structure as:
>
>  Tag: Attribute (0x420008), Type: Structure (0x01), Data:
>         Tag: Attribute Name (0x42000A), Type: Text String (0x07),
> Data: Unique Identifier
>         Tag: Attribute Value (0x42000B), Type: Text String (0x07),
> Data: 1ed28ea5-2b31-4145-bcf2-36d0756d3890
>
> But in GET operation request, UUID is given in Unique identifier Tag. Like,
>
> Tag: Unique Identifier (0x420094), Type: Text String (0x07), Data:
> 1ed28ea5-2b31-4145-bcf2-36d0756d3890
>
> My doubt is that, why two types exists for UUID specification?


Thanks for helping me understand your question better.

Parameters to operations can be base objects (defined in section 2), parameters that correspond to attributes (defined in section 3), enumerations (defined in sections 9.1.3.2.*), or objects that are defined in the operation itself.

In addition to these parameters that must be listed explicitly in each operation request or response specification, a small set of operations allow to pass parameters that are NOT explicitly listed one by one. This is the case for Locate, Notify, Put, Get Attributes, Add/Modify/Delete Attribute. This is done with an attribute structure with name/index/value, that allows to pass multiple instances if needed.

In the example you raised, Unique Identifier is not treated differently than other such attribute parameters passed to Locate for matching purposes.

>
> I mean why can't UUID specified in Attribute structure (present in
> locate) be used in GET operation request also?


It's a decision in the specification to use the straight TTLV encoding for parameters that are explicitly specified (whether they are attributes, base objects, or enumerations), as opposed to using the heavier attribute name/index/value structure used only in the cases I mention above.

>
> Is there any specific reason behind the framing of request in this way?
>
> Please help me understand this.
>
>
> Best Regards,
> --------------------
> Trinath Somanchi.
> email : trinath.somanchi@gmail.com
> Mobile: +91 9866 235 130
>
>
>
> -----Original Message-----
> From: Robert Haas [
mailto:rha@zurich.ibm.com]
> Sent: Wed 29-Sep-10 9:35 PM
> To: Somanchi Trinath-B22327
> Cc: kmip@lists.oasis-open.org
> Subject: Re: [kmip] KMIP :: Doubts
>
>
> Hi Trinath, please find answers to your questions below.
> Regards,
> -Robert
>
> Somanchi Trinath-B22327 <B22327@freescale.com> wrote on 09/28/2010
> 06:52:31 AM:
> >
> > [kmip] KMIP :: Doubts
> >
> > Somanchi Trinath-B22327
> >
> > to:
> >
> > kmip
> >
> > 09/28/2010 06:54 AM
> >
> > Hi-
> >  
> > I have the following doubts regarding KMIP.  Please review and help
> > me understand them.
> >  
> > Doubts:
> >  
> > [1] An SHA-256 hash is created by the KMIP server when a Symmetric
> > Key is created. How to generate some other kind of Hash say,
> > SHA-512? I mean do client need to send add attribute request with
> > name value pairs of variables present in Digest structure or how
> > this can be achieved?
>
> It is up to a server implementation to choose to have digests beyond
> the mandatory one. The client cannot request this.
>
> >  
> > [2] With respect to ADD attribute operation, the spec says, READ-
> > ONLY attributes cannot be added using this operation. READ-ONLY
> > attribute is defined as 'Attribute that shall not be modified by
> > either server or client and that shall not be deleted by client'. My
> > reading through spec document gave me the following NON-READ-ONLY
> > attributes which can be used in ADD attribute operation. The listing
> > is as follows:
>
> Digest should be removed from your list.
>
> > Name
> > o   Name Value
> > o   Name Type
> > Cryptographic Parameters
> > o   Block Cipher Mode
> > o   Padding Method
> > o   Hashing Algorithm
> > o   Key Role Type
> > Usage Limits
> > o   Usage Limits Total
> > o   Usage Limits Count
> > o   Usage Limits Unit
> > Object Group
> > Link
> > o   Link Type
> > o   Linked Object Identifier
> > Contact Information
> > Custom Attribute
> > Activation Date
> > Process Start Date
> > Protect Stop Date
> > Deactivation Date
> > Digest
> > Operational Policy
> > Application Specific Information
> > o   Application Namespace
> > o   Application Data
> > Please comment if I missed any attribute.
> >  
> > [3] With respect to the Usage Guide, I'm comparing two requests.
> > LOCATE and GET operation Requests.
> >  
> > In the LOCATE operation request, the UUID is placed in an ATTRIBUTE
> > BLOCK which is common for all requests. like,
> >  
> > Locate
> > In: uuidKey
> >  
> > Tag: Request Message (0x420078), Type: Structure (0x01), Data:
> >   Tag: Request Header (0x420077), Type: Structure (0x01), Data:
> >     Tag: Protocol Version (0x420069), Type: Structure (0x01), Data:
> >       Tag: Protocol Version Major (0x42006A), Type: Integer (0x02),
> > Data: 0x00000001 (1)
> >       Tag: Protocol Version Minor (0x42006B), Type: Integer (0x02),
> > Data: 0x00000000 (0)
> >     Tag: Batch Count (0x42000D), Type: Integer (0x02), Data: 0x00000001 (1)
> >   Tag: Batch Item (0x42000F), Type: Structure (0x01), Data:
> >     Tag: Operation (0x42005C), Type: Enumeration (0x05), Data:
> > 0x00000008 (Locate)
> >     Tag: Request Payload (0x420079), Type: Structure (0x01), Data:
> >       Tag: Attribute (0x420008), Type: Structure (0x01), Data:
> >         Tag: Attribute Name (0x42000A), Type: Text String (0x07),
> > Data: Unique Identifier
> >         Tag: Attribute Value (0x42000B), Type: Text String (0x07),
> > Data: 1ed28ea5-2b31-4145-bcf2-36d0756d3890
> >  
> > But in GET operation request, the UUID is not placed in the ATTRIBUTE block.
> >  
> >  
> > Tag: Request Message (0x420078), Type: Structure (0x01), Data:
> >   Tag: Request Header (0x420077), Type: Structure (0x01), Data:
> >     Tag: Protocol Version (0x420069), Type: Structure (0x01), Data:
> >       Tag: Protocol Version Major (0x42006A), Type: Integer (0x02),
> > Data: 0x00000001 (1)
> >       Tag: Protocol Version Minor (0x42006B), Type: Integer (0x02),
> > Data: 0x00000000 (0)
> >     Tag: Batch Count (0x42000D), Type: Integer (0x02), Data: 0x00000001 (1)
> >   Tag: Batch Item (0x42000F), Type: Structure (0x01), Data:
> >     Tag: Operation (0x42005C), Type: Enumeration (0x05), Data:
> > 0x0000000A (Get)
> >     Tag: Request Payload (0x420079), Type: Structure (0x01), Data:
> >       Tag: Unique Identifier (0x420094), Type: Text String (0x07),
> > Data: 1ed28ea5-2b31-4145-bcf2-36d0756d3890
> >  
> > Why this difference is maintained? Will this differentiation helps
> > improving KMIP request/response? Please comment on this.
>
> Locate returns the Unique Identifier(s) of object(s) matching a
> given search criteria. So you should not need to specify Unique
> Identifier in a Locate operation request.
>
> >  
> > [4] With respect to Use case Document, ADD Attribute batch item
> > request, each batch item has UUID. This facility is provided to
> > handle adding of attributes to TWO different objects at a time?
> >  
>
> Yes. The use case described repeats the same Unique Identifier in
> the operation of each batch item, but they could have been different
> Unique Identifiers as well.
>
> > Please help me understand these.
> >  
> > Thanking You,
> >  
> > Trinath Somanchi,
> > trinath.somanchi@freescale.com
> >


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]