Charles, thanks for this info. Just to confirm, this behavior is not a result of RRP downtime and is the normal OpenSRS server operation. However, your indication that nothing had changed caused me to dig deeper into the cause of the problem. Here are the results so other developers can use the results. I did determine that I can "sort of" trap this event... There is a "return_code" of 465 'invalid data supplied...' Here is what is returned in the hash %register_results when this return_code is returned: Domain = somedomain.com _response_code = 200 = _response_text = Registration successful _is_success = 1 id = 802437 registration_code = 465 What's confusing is that the published 'return_codes' are sometime used as _resonse_code values and sometimes used as other undocumented values (like registration_code in this example). You will never see them used in a field named 'result_code'. For example, the return_code 487 'domain not transferable' is issued when a domain cannot be transfered. However, in this case, it will be used as the _response_code and the _response_text is set to the text 'domain not transferable'. For an invalid name server, you will note that there is no text returned to indicate what the nature of the 'invalid data' is so there is no way to definitively trap this particular error. Is there anyone at OpenSRS working on documentation for returned values? For example, for each named field (_response_code, registration_code, etc.) developers need to know what are all the possible values each of these fields can contain AND when (under what circumstances) they can contain them. Rich Shockney RS Marketing -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Charles Daminato Sent: Sunday, July 30, 2000 10:37 AM To: [EMAIL PROTECTED] Cc: Dev-List@Opensrs. Org Subject: Re: FW: Problem with server response to invalid name servers To the best of my knowledge the methodology has not changed - however, when there is an outage at the RRP level (such as last night, and a subsequent outage this morning of the test system) - our system will Pend orders to allow the gathering of information. Customers can then inform the RRP of the order number for followup. Charles Daminato OpenSRS Support Manager [EMAIL PROTECTED] On Sun, 30 Jul 2000 [EMAIL PROTECTED] wrote: > I have also sent the following to [EMAIL PROTECTED] However, I thought I > would seek advise/comment here from other developers. > > Are my observations (noted below) regarding the change in server-side > response accurate? > > Am I correct that no error return codes are generated when invalid name > servers are supplied or am I missing something? > > How are others handling situations where the name is pended and the customer > thinks the name has been successfully registered to the name servers they > supplied, but in fact are not? > > Rich > > ==================================== > > I'm encountering a problem with the action taken by the server-side script > when invalid name servers are specified. > > During our testing (on both the test server and the production server) the > action taken when an invalid name server was specified was to return an > error. However, now when an invalid name server is specified, the following > happens: > > (1) The domain is placed in the "Pending" area. > (2) The customer is provided a domain confirmation registration number! > (3) The customer receives a confirmation email informing him or her that > their domain was successfully registered. > (4) The client-side script receives a response indicating that the domain > has been successfully registered. > > Questions: > > Is my observation correct that the action taken by the server with respect > to invalid name server entries has changed? > > If the action has changed, what was the rationale for making the change? Why > would a change be made that would remove from the reseller any and all > control over what action is taken with respect to invalid name server > entries? > > Can the server-side response be modified so the client-side script can > "know" that the domain has been pended? > > Thanks. > Rich Shockney > RS Marketing > >
