Re: [Megaco] ALL in Topology descriptor and CHOOSE in command

[ckaushik@hss.hns.com]

Hi Christian/List,

Extending this discussion further pls consider the following scenarios:

**C1, C2 and C3 are three commands in the same action as the Topology
Descriptor**
1. Topology={T1,T2,isolate,T1,T3,oneway,T1,T4,bothway},C1,C2,C3

As per the discussions held on mailing list earlier:
    (a) Topology descriptor will be executed in the beginning and before any of
the commands C1,C2 or C3 are
        executed.

    (b) If any error occurs in the execution of the Topology descriptor, a
Context Level Error is sent in the response     and non of the commands in the
action i.e. C1, C2 or C3 get executed.

2. Topology={T1,T2,isolate,$,T1,oneway,T1,T3,oneway},C1,C2,Add=${Media...,
Event..., Signal...}

    (a) Topologies T1,T2,isolate and T1,T3,oneway will be applied before
execution of commands C1, C2 and Add=$.              Topology $,T1,oneway gets
executed once the execution of Add=$ is complete.

    (b) If execution of Topology $,T1,oneway fails a Command Level response is
sent at the same level where the command Add=$ is present.

I have the following queries regarding the above scenarios: -

1. Does the execution of all the Topology triples determine the final result of
the execution of Topology Descriptor OR    Are the Topology Triples considered
to be independent of each other and the result of execution of one does not
affect the other?

2. In the second bullet above, if execution of Topology $,T1,oneway fails, what
happens to the commands C1, C2 and Topologies T1,T2,isolate and T1,T3,oneway.
These commands and Topologies have already been executed before the failure of
Topology $,T1,oneway. Does the MG send a Success response for commands C1, C2
and Topologies T1,T2,isolate and T1,T3,oneway and an error response for Topology
$,T1,oneway OR Is the MG expected to do a rollback for commands C1, C2 and
topologies already executed and send an error response for the same.

Pls clarify.

Regards,
Chintan...
www.hssworld.com








shbansal@hss.hns.com on 01/24/2003 02:36:53 PM

To:   Christian Groves <Christian.Groves@ericsson.com.au>
cc:   manfred.volz@nokia.com, megaco@ietf.org (bcc: Chintan Kaushik/HSS)

Subject:  Re: [Megaco] ALL in Topology descriptor and CHOOSE in command







Hi Chris,

In earlier mails it was stated that

Section 7.1.18 means that you must consider the topology descriptor before
addressing the
commands. You should then use the information that is in the topology descriptor
when you process the commands in the message.

Now with this implementation the final result should be
T1 is isolated from T2 and
T1 is isolated from T3 and
T2 is bothway connected to T3

Correct me if I have some wrong interpretation

Thanks and regards
Shobhit





Christian Groves <Christian.Groves@ericsson.com.au> on 01/22/2003 05:36:36 AM

To:   manfred.volz@nokia.com
cc:   megaco@ietf.org (bcc: Shobhit Bansal/HSS)

Subject:  Re: [Megaco] ALL in Topology descriptor and CHOOSE in command




Hello Manfred,

In accordance to 7.1.18 "The Topology descriptor occurs before the commands in
an action." it would be b).

Christian

manfred.volz@NOKIA.COM wrote:
>
> Hi all,
>
> Let's assume we have 2 terminations T1 and T2 which are oneway connected.
> Now we add a new termination T3 with the following topology settings.
>
> Topology (T1,ALL,isolate)
>   ADD $
>
> Which of the following results is correct?
>
> a.) T1 is isolated from T2 and
>     T1 is isolated from T3 and
>     T2 is bothway connected to T3
> b.) T1 is isolated from T2 and
>     T1 is bothway connected to T3 and
>     T2 is bothway connected to T3
>
> Thanks for your help.
>
> best regards Manfred
>
> _______________________________________________
> Megaco mailing list
> Megaco@ietf.org
> https://www1.ietf.org/mailman/listinfo/megaco


_______________________________________________
Megaco mailing list
Megaco@ietf.org
https://www1.ietf.org/mailman/listinfo/megaco




_______________________________________________
Megaco mailing list
Megaco@ietf.org
https://www1.ietf.org/mailman/listinfo/megaco




_______________________________________________
Megaco mailing list
Megaco@ietf.org
https://www1.ietf.org/mailman/listinfo/megaco