[Nsi-wg] Open dynamicKL NSA
John MacAuley
john.macauley at surfnet.nl
Wed Sep 7 13:55:50 CDT 2011
All,
I generated a request through OpenDRAC and here is the format of the reservation message. I have compared it to Henrik's message and there is definitely an issue with the "<ns6:reservation>" in my message and the "<ns0:reservation>" in his message since they are from two separate namespaces. If you look in the file ogf_nsi_connection_interface_v1_0.wsdl you will see that ReservationRequestType is referencing the reservation message definition (ref="types:reservation") from the type namespace. This means that <ns6:reservation> shown below is correct. Henrik's implementation seems to be using the interface namespace. Looks like there is a bug in his SOAP stack. Now we can get around it but I will need to modify the WSDL definition to not reference elements, but use imported types instead. This will require everyone to make changes to their implementations. I assume this is too risky for Rio...
John.
<?xml version='1.0' encoding='utf-8'?>
<S:Envelope xmlns:S="http://schemas.xmlsoap.org/soap/envelope/">
<S:Body>
<ns5:reservationRequest xmlns:ns6="http://schemas.ogf.org/nsi/2011/07/connection/types"
xmlns:ns5="http://schemas.ogf.org/nsi/2011/07/connection/interface"
xmlns:ns4="urn:oasis:names:tc:SAML:2.0:assertion"
xmlns:ns3="http://www.w3.org/2001/04/xmlenc#"
xmlns:ns2="http://www.w3.org/2000/09/xmldsig#">
<ns5:correlationId>urn:uuid:5d2a2bc0-a51a-4418-b5d4-322807247593</ns5:correlationId>
<ns5:replyTo>http://localhost:8084/nsi-v1/ConnectionServiceRequester</ns5:replyTo>
<ns6:reservation>
<requesterNSA>urn:ogf:network:nsa:ferb.surfnet.nl</requesterNSA>
<providerNSA>urn:ogf:network:nsa:phineas.surfnet.nl</providerNSA>
<sessionSecurityAttr>
<ns4:Attribute NameFormat="urn:oasis:names:tc:SAML:2.0:attrname-format:basic"
Name="globalUserName">
<ns4:AttributeValue xmlns:xs="http://www.w3.org/2001/XMLSchema"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:type="xs:string">jrv at internet2.edu</ns4:AttributeValue>
</ns4:Attribute>
</sessionSecurityAttr>
<reservation>
<globalReservationId>urn:ogf:network:service:1ae919be-bff0-4053-bf1f-3b629f694197</globalReservationId>
<description>This is a test schedule connecting Aiden to Ashley</description>
<connectionId>urn:uuid:5d2a2bc0-a51a-4418-b5d4-322807247593</connectionId>
<serviceParameters>
<schedule>
<startTime>2011-09-08T00:00:00.000-04:00</startTime>
<endTime>2011-09-09T00:00:00.000-04:00</endTime>
</schedule>
<bandwidth>
<desired>200</desired>
</bandwidth>
</serviceParameters>
<path>
<directionality>Bidirectional</directionality>
<sourceSTP>
<stpId>urn:ogf:network:stp:Aruba:Aiden</stpId>
</sourceSTP>
<destSTP>
<stpId>urn:ogf:network:stp:Aruba:Ashley</stpId>
</destSTP>
</path>
</reservation>
</ns6:reservation>
</ns5:reservationRequest>
</S:Body>
</S:Envelope>
----- Original Message -----
> From: "Henrik Thostrup Jensen" <htj at nordu.net>
> To: "jeonghoon moon" <otello90 at gmail.com>
> Cc: "Henrik Thostrup Jensen" <htj at ndgf.org>, "???(YoungWook Cha)" <ywcha at andong.ac.kr>, "Jerry Sobieski"
> <jerry at nordu.net>, "JongUk Kong" <jonguk.kong at gmail.com>, "John MacAuley" <john.macauley at surfnet.nl>, "NSI Working
> Group" <nsi-wg at ogf.org>
> Sent: Tuesday, September 6, 2011 9:26:39 AM
> Subject: Re: [Nsi-wg] Open dynamicKL NSA
>
> Hi again
>
> On Tue, 6 Sep 2011, jeonghoon moon wrote:
>
> > I am very sorry for our inconvenient log file and ambiguous
> > indication about name space problem.
>
> We are all trying to figure out what is going on here :-).
>
> > We developed our web services with JAX-WS2.0 based Metro engine,
> > which supports WSDL to Java generation tool.
> >
> > To confirm the usage of name spaces in NSI messages, we also used
> > AXIS2 and CFX engines to generate two more Java
> > source codes from the WSDL document.
> >
> > All three engines generated the same reservationRequest message
> > which has two name spaces as attributes of the
> > reservationRequest element and does not have prefixes for all
> > children elements(e.g., requesterNSA, providerNSA,
> > …):
>
> Mm... okay. I am using SUDS. So far I have found two bugs in SUDS. It
> is
> quite likely that there are more.
>
>
> > <ns5:reservationRequest
> > xmlns:ns2="urn:oasis:names:tc:SAML:2.0:assertion"
> > xmlns:ns3="http://www.w3.org/2001/04/xmlenc#"
> > xmlns:ns4="http://www.w3.org/2000/09/xmldsig#"
> > xmlns:ns5="http://schemas.ogf.org/nsi/2011/07/connection/interface"
> > xmlns:ns6="http://schemas.ogf.org/nsi/2011/07/connection/types">
> > <ns5:correlationId>78729d97-8264-4a4b-ad53-8fec5ad529fb</ns5:correlationId>
> > <ns5:replyTo>http://220.69.240.218:2000/kreonet/ConnectionServiceRequester</ns5:replyTo>
> > <ns6:reservation>
> > <requesterNSA>urn:ogf:network:NSnetwork:kreonet</requesterNSA>
> > <providerNSA>urn:ogf:network:NSnetwork:dynamicKL</providerNSA>
> > <reservation>
>
> This strikes me as incorrect. I would expect a default namespace
> declaration or an explicit namespace in front of the requesterNSA,
> providerNSA, and reservation elements.
>
> But I could very well be wrong.
>
> > Our implementation utilizes Metro engine to parse NSI messages.
> >
> > For your latest trail with ns2 name space for all children elements
> > of reservation element,
> >
> > Metro engine returns null value when our implementation asks it to
> > parse requesterNSA element of your
> > reservationRequest message.
> >
> > Syntax error was marked in our log because of null value.
> >
> > We also have great difficulty to understand why Metro engine
> > returns null value.
> >
> > When we use AXIS2(Jaxb) to parse your message, it returns an
> > exception with unexpected element cause.
>
> Since you have tried two different implementations it does point to
> the
> SOAP stack I'm using. However my impression from reading the WSDL
> manually
> is that everything under the <ns6:reservation> should be under the
> xmlns:ns6="http://schemas.ogf.org/nsi/2011/07/connection/types
> namespace
> as well.
>
> John, can you comment here?
>
> > To fix out this problem, we are studying unprefixed attribute
> > (http://www.rpbourret.com/xml/NamespaceMyths.htm#myth4), but we
> > felt deeply our limitation about WSDL and name
> > space knowledge.
>
> This is my first venture into WSDL as well, so I am puzzled as well.
>
> > We sincerely ask someone to give us some comments about this
> > problem.
>
> +1 from here as well.
>
>
> Best regards, Henrik
>
> Henrik Thostrup Jensen <htj at ndgf.org>
> NORDUnet / Nordic Data Grid Facility.
More information about the nsi-wg
mailing list