COmanage Developers List

[Scott Koranda <>]

Hi,

You can tell Tom that the main application hosted by the SP (COmanage)
now detects a missing ePPN and properly directs the user to the
InCommon EHS.

Cheers,

Scott

On Thu, Oct 25, 2012 at 10:57 AM, Heather Flanagan
<>
 wrote:
> And to add to the question about the EHS stuff...
>
> From Tom this morning:
>> Regarding the Error Handling Service, we do intend to improve error
>> handling and may use the InCommon EHS, but that is going to happen
>> sometime in the first half of next year.
>
> Unfortunately, this will concern some TAC members since you display all 
> IdPs on your discovery interface. What happens exactly if insufficient 
> attributes are released? Can you provide a URL so we can test this with an 
> arbitrary IdP?
>
> ----- Original Message -----
> From: "Scott Koranda" 
> <>
> To: "Benn Oshrin" 
> <>
> Cc: 
> 
> Sent: Thursday, October 18, 2012 7:56:38 PM
> Subject: Re: [comanage-dev] Two questions for the R&S application
>
> Hi,
>
> On Thu, Oct 18, 2012 at 9:53 PM, Benn Oshrin 
> <>
>  wrote:
>> On 10/18/12 10:41 PM, Scott Koranda wrote:
>>>
>>> On Thu, Oct 18, 2012 at 6:49 PM, Benn Oshrin 
>>> <>
>>>  wrote:
>>>>
>>>> On 10/18/12 12:19 PM, Scott Koranda wrote:
>>>>>>
>>>>>>
>>>>>> Do you know about the Federated Error Handling Service?
>>>>>
>>>>>
>>>>> Yes.
>>>>>
>>>>> We are aiming to move in that direction and to add proper error
>>>>> handling to the application (COmanage).
>>>>
>>>>
>>>>
>>>> We are? Is there a ticket for this? :)
>>>>
>>>
>>> There is:
>>>
>>> https://bugs.internet2.edu/jira/browse/CO-486
>>>
>>> That guy Benn opened it...
>>
>>
>> The scope of CO-486 is different: Registry generates a message that you're
>> not a valid (to Registry) user, rather than random menu rendering. ie: a 
>> bug
>> fix.
>>
>> Integrating with InCommon's FEHS would be a different task (ie: an
>> enhancement), and warrants a separate ticket.
>>
>
> By "we are aiming to move in that direction" I did not mean
> integration with the InCommon FEHS. I meant provide proper error
> handling when REMOTE_USER Is not populated.
>
> If we do that at the application layer then I don't think we need to
> integrate with the FEHS. I see that more for situations where one
> cannot get a good error message out of the application.
>
> So I don't think we need another ticket.
>
> Cheers,
>
> Scott