Sent from my iPhone

On Dec 6, 2010, at 6:21 PM, Marc Petit-Huguenin <[email protected]> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 12/06/2010 02:34 PM, jc wrote:
>> 
>> 
>> On Dec 6, 2010, at 5:12 PM, Marc Petit-Huguenin <[email protected]
>> <mailto:[email protected]>> wrote:
>> 
>> More questions, comments and nits:
>> 
> 
> [...]
> 
>> 
>> A.15. Section 5.3.4 "HashAlgorithm hash_alg;"
>> 
>> What are the algorithms that should be supported?
>> Is the hash only used to identify the certificate in
>> SecurityBlock.certificates?
>> 
>>> No, it is used for AOR storage among other things.
>> 
> 
> My question was about Signature in SecurityBlock, not in StoredData, so let me
> rephrase it:
> 
> What is the certificate_hash value in SecurityBlock used for?
> 
> [...]
> 
>> A.20. Section 5.7, "If the message is not fragmented, then both the
>> first and
>> last fragment are set to 1..."
>> 
>> If the first fragment bit is set to 1 when the message is fragmented
>> and is also
>> set to one when the message is not fragmented, then it is always set
>> to 1, so
>> what is the point of having it in the first place?

Because it IS the first and last fragment.

>> 
>>> I think you misread. There are two fragment flags, begin and end.
> 
> Here's the quotes:
> 
> Section 5.3.2: "If the high bit (0x80000000) is set, it indicates that the
> message is fragmented."
> Section 5.7: "If the message is not fragmented, then both the first and last
> fragment bits are set to 1..."
> 
> [...]
> 
>> A.28. Section 10.1.1 last paragraph "such an XML configuration file
>> sent over
>> email."
>> 
>> Because the signatures on the XML document are done on exact byte
>> string and
>> because emails servers are known to mess with end of lines, we will see
>> configuration documents that cannot be verified after been sent by
>> email (what
>> was wrong with using XML-sig anyway?).
>> 
>> 
>>> Why would an email server alter an "attachment"? Don't send them in the
>>> body.
> 
> Some email servers do alter attachments, it's a fact (and the reason why pcap
> files have to be gzipped before been sent by email), but in fact I just wanted
> to express that basing a signature on an XML fragment without canonicalization
> was perhaps not a good idea.
> 
> [...]
> 
> - -- 
> Marc Petit-Huguenin
> Personal email: [email protected]
> Professional email: [email protected]
> Blog: http://blog.marc.petit-huguenin.org
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> 
> iEYEARECAAYFAkz9b+4ACgkQ9RoMZyVa61emFQCfaHnoKXXtbBmgVcEBDmUYFytD
> CFwAoIKbTDa7BjW1twZRTVDt+m3OEGYJ
> =Kz5R
> -----END PGP SIGNATURE-----
_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

Reply via email to