Re: ROID how useful?
"Gould, James" <[email protected]>
| Newsgroups | gmane.ietf.provreg |
|---|---|
| Message-ID | <C41D7AF7FCECBE44940E9477E8E70D7A0D7329FD@BRN1WNEXMBX01.vcorp.ad.vrsn.com> |
We used the ROID in the post-create commands for the Defensive Registration and NameWatch object mappings in DotName since there could be multiple registrations for the same string. There are cases where the registration name is not a unique identifier and there are cases where the name can be changed like with hosts. A history command for objects that support changing name strings can show all of the transactions for the object based on the ROID. To note, the launch phase extension defines an application identifier separate from the ROID, since the allocation of an application to a domain might or might not result in a different ROID. -- JG James Gould Principal Software Engineer [email protected] 703-948-3271 (Office) 12061 Bluemont Way Reston, VA 20190 VerisignInc.com On 12/10/12 6:35 AM, "Klaus Malorny" <[email protected]> wrote: >On 09/12/12 20:33, Francisco Obispo wrote: >> [...] >>> In contrast, roids or some equivalent will prove necessary whenever the >>> natural key is NOT unique. >>> >> >> Exactly, but in case of domain names, contact ids, and host names, they >>all >> have natural keys. >> [...] > >Hi, > >all three are not unique over time. If you delete a domain and re-create >it some >time later, you have two different objects. A use of the ROIDs could have >been >to address registry objects in a generic way, but EPP does not make use >thereof >(except in the mentioned auth-infos). As such, it is indeed questionable >why >ROIDs are exposed in EPP. > >Regards, > >Klaus > > >_______________________________________________ >provreg mailing list >[email protected] >https://www.ietf.org/mailman/listinfo/provreg _______________________________________________ provreg mailing list [email protected] https://www.ietf.org/mailman/listinfo/provreg