Re: [Handle-info] is rfc3651 still up to date?

Robert

On Mar 18, 2021, at 11:15 AM, Robert R Tupelo-Schneck <schneck@cnri.reston.va.us> wrote:

A new specification to replace the Handle RFCs is a project of current interest to both CNRI and DONA. I can't currently give any accurate time estimates on when that will be available, however.

Tibor's advice to use the Handle.Net source code is useful. Also, please feel free to ask any and all questions here, or by emailing me directly.

We would also be interested to hear your motivation -- simple curiosity, or are you developing some new software to make use of Handle?

This short list of errata for the HS_SITE section of RFC 3651 should help:

- The HS_SITE version is 1, not 0.
- The first bit of PrimaryMask is PrimarySite and the second bit is MultiPrimary.
- IPv4 addresses are encoded not as ::FFFF:xxxx:xxxx but as ::xxxx:xxxx.
- In ServiceType, 1 is admin, 2 is query, 3 is both.
- In TransmissionProtocol, 0 is UDP, 1 is TCP, 2 is HTTP.
- CERT.X509 was never implemented.
- Hashing uses only the last (least significant) four bytes of the 16-byte MD5 hash.

Best,
Robert

On Mar 17, 2021, at 9:30 PM, Gerhard Gonter <ggonter@gmail.com> wrote:

The venerable rfc3651[1] describes how data in Handle entries are
encoded, so I tried to make sense of the HS_SITE element of our own
0.NA Handle entry [2]. I know, the rfc describes the binary structure
of the Handle protocol data unit while the web server presents
something interpreted. I assume, it is a hex dump of what would be
expected as the original element. However, the structure of the
HS_SITE element of [2] seems not to be matching what is described in
the rfc [1]. For instance, the IP address of the Handle server can be
found in there as "00000000000000000000000083820432" which should
represent the IPv6 address of the machine while the rfc states it
*should* be encoded there as "00000000000000000000FFFF83820432" :

-- 8< ---
<Address>
        The 16-byte IPv6 [15, 16] address of the handle server. Any
        IPv4 address should be presented as :::::FFFF:xxxx:xxxx (where
        xxxx:xxxx can be any 4-byte IPv4 address).
--- >8 ---

I know, the "should" here is not a "MUST" as defined in rfc 2119 [3].
However, it does not match. There are other parts of the HS_SITE
element that do not match (IMHO) the description in rfc3651, but if
anyone can point me to an up-to-date documentation of the data
structures (or other documentation relevant for the Handle System), I
will try to make more sense of what i can find in live data.

* [1] https://tools.ietf.org/html/rfc3651#page-17
* [2] http://hdl.handle.net/0.NA/11353
* [3] https://tools.ietf.org/html/rfc2119

regards, Gerhard Gonter
_______________________________________________
Handle-Info mailing list
Handle-Info@cnri.reston.va.us
http://www.handle.net/mailman/listinfo/handle-info

_______________________________________________
Handle-Info mailing list
Handle-Info@cnri.reston.va.us
http://www.handle.net/mailman/listinfo/handle-info