Re: contact:disclose clarifications / best practices
Bernhard Reutner-Fischer <[email protected]>
| Newsgroups | gmane.ietf.provreg |
|---|---|
| Message-ID | <CAC1BbcShnVCxtw2PyEP4qWp-t++iEtv_vh4+NU8Mkyov36mtmA@mail.gmail.com> |
On 23 January 2013 13:20, Keith Gaughan <[email protected]> wrote: > On Wed, Jan 23, 2013 at 11:26:31AM +0100, Bernhard Reutner-Fischer wrote: > >> I have questions about contact:disclose. >> >> 1) How do i correctly handle non-disclosure of all fields of a contact:info >> command from a third party via EPP, especially name, email, addr (city, cc >> therein) etc? > > Are you talking about a <contact:info> request being made against a contact by a > registrar other than the owning registrar, and is a <authInfo> element being > provided? In my example i do not have an authinfo set in the contact, no. > >> 2) what dummy values should i use for the required fields, what is best/common >> practice? > > If you omit them, the default registry policy is applied [SS2.9]: > > A server operator announces a default disclosure policy when establishing a > session with a client. [...] When an object is created or updated, the client > can specify contact attributes that require exceptional disclosure handling > using an OPTIONAL <contact:disclose> element. > > [SS2.9] http://tools.ietf.org/html/rfc5733#section-2.9 > > Thus, a good default is to comit the <contact:disclose> block entirely unless > there are specific fields you wish not to be disclosed. right, but that was not my question. I was assuming that a contact:info sent in from a non-sponsoring registrar for a contact that wants not to disclose it's name, email, addr is considered to be from a third party so i would have had to omit those affected fields from the reponse. Since the schema neither allows to omit e.g. the email nor to have it empty, i would have had to fill in some dummy data (see "n/a" in the second response in my initial mail). > >> 3) is it ok to disclose the contact:disclose itself to third-parties? >> I guess it is, also to alleviate determining what fields contain real >> data instead of workaround fill junk? > > I can't see any harm in doing so. It's not disallowed. Again, we need to know if > this command is being issued with or without a <contact:authInfo> element. > >> Long story: >> Ideally i would have expected to be able to return: >> >> <?xml version="1.0" encoding="UTF-8" standalone="no"?> >> <epp xmlns="urn:ietf:params:xml:ns:epp-1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> >> <response> >> <result code="1000"> >> <msg>Command completed successfully</msg> >> </result> >> <resData> >> <contact:infData xmlns="urn:ietf:params:xml:ns:contact-1.0" xmlns:contact="urn:ietf:params:xml:ns:contact-1.0"> >> <contact:id>dnd018360624626</contact:id> >> <contact:roid>C0000000704-BOX</contact:roid> >> <contact:status s="ok"/> >> <contact:postalInfo type="int"> >> <contact:addr/> >> </contact:postalInfo> >> <contact:clID>primary</contact:clID> >> <contact:crID>primary</contact:crID> >> <contact:crDate>2013-01-22T18:50:12.160307Z</contact:crDate> >> <contact:disclose flag="0"> >> <contact:name type="int"/> >> <contact:org type="int"/> >> <contact:addr type="int"/> >> <contact:voice/> >> <contact:fax/> >> <contact:email/> >> </contact:disclose> >> </contact:infData> >> </resData> >> <trID> >> <clTRID>8DE21C7A-64C4-11E2-BE5E-5148DCA3D1ED</clTRID> >> <svTRID>20130122195013136028C3-secondary-box</svTRID> >> </trID> >> </response> >> </epp> >> >> But that complains about >> Element '{urn:ietf:params:xml:ns:contact-1.0}clID': This element is not expected. Expected is one of ( {urn:ietf:params:xml:ns:contact-1.0}voice, {urn:ietf:params:xml:ns:contact-1.0}fax, {urn:ietf:params:xml:ns:contact-1.0}email > I assume the it's a client parser that's giving these errors. If a contact is > being queried with an <authInfo> block, I don't think that disclosure flags > actually apply: if you've got a domain's authorisation code, that implies full > access to linked objects for the purpose of issuing <info> commands. yes, agree. > While this is an assumption on my part, there's good reason for it: transfer. > Say you have the email field set for nondisclosure and somebody was transferring > the domain away from you. The gaining registrar needs to be able to get at the > email so that they can contact the registrant for confirmation, so if the the > email field was set for nondisclosure, they wouldn't be able to deal with the > transfer. > >> i.e. fill in name, city, email with an arbitrary token (which one?!) >> and fill in cc with an arbitrary, invalid ccType. > > I think the disclosure fields deal more with the likes of what's published in > WHOIS and the likes. That is one way to see it, agree. thanks, _______________________________________________ provreg mailing list [email protected] https://www.ietf.org/mailman/listinfo/provreg