← Prev in month ← Prev in thread
Next in thread → Next in month →

Re: [wsrp-interop] Re: Oracle producer throwing fault ongetServiceDescription

From
Rich Thompson <>
Date
2003-08-13T13:27:45+00:00
ID
Thread
Re: [wsrp-interop] Re: Oracle producer throwing fault ongetServiceDescription
I also agree that the reference Subbu
cites is an inconsistency. The key points are:

  Registration portType exposed
=> In-band registration supported

  requiresRegistration metadata
flag => states whether all other operations will require a valid registrationHandle

Rich Thompson

Subbu Allamaraju <>

08/11/2003 10:39 AM
        

        To:
       

        cc:
       

        Subject:
       Re: [wsrp-interop] Re: Oracle producer
throwing fault on getServiceDescription

David,

You convinced me, and I agree that the reference I mentioned is not 

consistent (although there is no conformance statement associated) with

the references you pointed out.

Regards,

Subbu

David Ward wrote:

> Subbu

> 

> This is indeed a misleading phrase in the spec, which seems to imply

> "!requiresRegistration => !(registration supported)"
[Read "A => B" as 

> "implies", i.e. "if A then B"]. However, this
is not the part of the 

> spec which specifies the meaning of the requiresRegistration flag,
so 

> shouldn't be taken too literally.

> 

>  From the definition of requiresRegistration:

> 

>     requiresRegistration: A boolean indicating whether or
not the

>     Producer requires Consumer registration. If requiresRegistration
is

>     set to “false” then it MUST be valid to not pass a

>     RegistrationContext parameter to all operations with
this parameter.

>     If requiresRegistration is set to “true” then the
Producer MUST

>     return a fault message when no RegistrationContext is
supplied to an

>     operation, other than getServiceDescription(), which
takes this field.

> 

> So we definitely have "requiresRegistration => registration
supported", 

> which is logically equivalent to "!(registraion supported) =>

> !requiresRegistration", but it still doesn't state the converse,
so it 

> is still valid for us to assume registration is supported when 

> !requiresRegistration.

> 

> Also, from the description of the registration port type

> 

>     A Producer that supports in-band registration of Consumers
exposes

>     the optional registration interface.

> 

> This also seems to imply that if the interface is exposed by the 

> producer, the producer supports registration (whether or not 

> requiresRegistration is set).

> 

> So could someone from the TC please clarify that it is wrong for a

> Consumer to assume "!requiresRegistration => !(registration
supported)" 

> and perhaps think about fixing the reference that Subbu mentions?

> 

> Dave

> 

> Subbu Allamaraju wrote:

> 

>> Could you point me to the spec where this is specified? From page
52, 

>> line 25,

>>

>> "the Producer’s metadata declares registration is not supported
(i.e. 25

>> requiresRegistration flag was set to “false”), ..."

>>

>> it is quite valid to interpret that the producer is indicating
that it 

>> does not support registration by setting this flag.

>>

>> Regards,

>>

>> Subbu

>>

>>

>> David Ward wrote:

>>

>>> If your producer has a registration port, our consumer will
call it. 

>>> Registration is still allowed by the spec when requiresRegistration
= 

>>> false. If you don't want our consumer to register you, you
will have 

>>> to remove the registration port from your WSDL.

>>>

>>> Regards

>>>

>>> Dave

>>>

>>> Subbu Allamaraju wrote:

>>>

>>>> Thanks.

>>>>

>>>> On a related note, we tried to register our producer with
your 

>>>> consumer. The producer we deployed does not support registration

>>>> (hence returns false for requiresRegistration) and throws
a fault 

>>>> when register() is called. But Oracle consumer is always
trying to 

>>>> call register() after getting service description.

>>>>

>>>> Regards

>>>>

>>>> Subbu

>>>>

>>>> David Ward wrote:

>>>>

>>>>> The schema doesn't define registrationHandle as nillable,
therefore 

>>>>> it is invalid to send xsi:nil for the registrationHandle.
The 

>>>>> registrationHandle should be ommitted altogether if
it is not set.

>>>>>

>>>>> I guess this is what is tripping up JAX-RPC.

>>>>>

>>>>> Regards

>>>>>

>>>>> David

>>>>>

>>>>> Subbu Allamaraju wrote:

>>>>>

>>>>>> David,

>>>>>>

>>>>>> I'm getting a server fault when our consumer is
trying to get 

>>>>>> service description. Details are below.

>>>>>>

>>>>>> Regards,

>>>>>>

>>>>>> Subbu

>>>>>>

>>>>>> Request >>>>>>>>>>>>>>>>

>>>>>>

>>>>>> Content-Length      369

>>>>>> User-Agent     BEA WLP 8.1

>>>>>> SOAPAction     urn:oasis:names:tc:wsrp:v1:types:getServiceDescription

>>>>>> Content-Type     application/x-www-form-urlencoded;
UTF-8

>>>>>> Accept     text/xml, application/xml,
*/*

>>>>>>     <soapenv:Envelope 

>>>>>> xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/">

>>>>>>    <soapenv:Body>

>>>>>>       <urn:getServiceDescription

>>>>>> xmlns:urn="urn:oasis:names:tc:wsrp:v1:types">

>>>>>>          <urn:registrationContext>

>>>>>>             <urn:registrationHandle
xsi:nil="true" 

>>>>>> xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"/>

>>>>>>          </urn:registrationContext>

>>>>>>       </urn:getServiceDescription>

>>>>>>    </soapenv:Body>

>>>>>> </soapenv:Envelope>

>>>>>>

>>>>>> Response <<<<<<<<<<<<<<<<

>>>>>>

>>>>>> <env:Envelope xmlns:env="http://schemas.xmlsoap.org/soap/envelope/">

>>>>>>    <env:Body>

>>>>>>       <env:Fault>

>>>>>>          <faultcode>env:Server</faultcode>

>>>>>>          <faultstring>Internal
server error</faultstring>

>>>>>>       </env:Fault>

>>>>>>    </env:Body>

>>>>>> </env:Envelope>

>>>>>>

>>>>>

>>>>> -- 

>>>>> ------------------------------------------------------------------------

>>>>>

>>>>>

>>>>> *David Ward*

>>>>> Principal Software Engineer

>>>>> Oracle Portal

>>>>>     Oracle European Development Centre

>>>>> 520 Oracle Parkway

>>>>> Thames Valley Park

>>>>> Reading

>>>>> Berkshire RG6 1RA

>>>>> UK    *Email:*     

>>>>> <mailto:>

>>>>> *Tel:*     +44 118 924 5079

>>>>> *Fax:*     +44 118 924 5005

>>>>>

>>>>>

>>>>>

>>>>

>>>>

>>>

>>> -- 

>>> ------------------------------------------------------------------------

>>>

>>> *David Ward*

>>> Principal Software Engineer

>>> Oracle Portal

>>>     Oracle European Development Centre

>>> 520 Oracle Parkway

>>> Thames Valley Park

>>> Reading

>>> Berkshire RG6 1RA

>>> UK    

>>> *Email:*      <mailto:>

>>> *Tel:*     +44 118 924 5079

>>> *Fax:*     +44 118 924 5005

>>>

>>>

>>>

>>

>>

> 

> -- 

> ------------------------------------------------------------------------

> 

> *David Ward*

> Principal Software Engineer

> Oracle Portal

>                  Oracle
European Development Centre

> 520 Oracle Parkway

> Thames Valley Park

> Reading

> Berkshire RG6 1RA

> UK                
 

> *Email:*              
    <mailto:>

> *Tel:*                
 +44 118 924 5079

> *Fax:*                
 +44 118 924 5005

> 

> 

> 

You may leave a Technical Committee at any time by visiting http://www.oasis-open.org/apps/org/workgroup/wsrp-interop/members/leave_workgroup.php
← Prev in month ← Prev in thread
Next in thread → Next in month →