NEOHAPSIS - Peace of Mind Through Integrity and Insight

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com

Neohapsis > Archives > MS DCOM Discussion> 2005-q3

Re: Forcing DCOM to use one brach of a redundant network

From: Anders Karlsson (anders.karlssonSG.ABB.COM)
Date: Thu Sep 01 2005 - 18:56:52 CDT

Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Hi,

Here is one approach, not sure if it is good but:

On each server have a small program(service) broadcasting its existence.
Then on the client side it knows the ip addresses where the COM servers
reside.

Once having the ip addresses, why not connect to both servers but having
only one of them active - so that when the active fails the client can
switch over without interruption.

hth
anders.

---

   Anders J Karlsson ATPA/IMS/COE

        Lead Engineer ABB Industry Pte. Ltd

                         No 2 Ayer Rajah Crescent

                         139935 Singapore



  MS Messenger/Email: anders.karlssonsg.abb.com

         Office / HP: +65 6773 8779 / +65 9863
                         8829

I have noticed that the people who are late are often so much jollier than
the people who have to wait for them.
-- E.V.Lucas.

Message from Stephen Sauls <SSAULSOPTO22.COM>DISCUSS.MICROSOFT.COM
received on 09/02/2005 07:45 AM


                09/02/2005 Stephen Sauls <SSAULSOPTO22.COM>DISCUSS.MICROSOFT.COM
                 07:45 AM
                                  Please respond to Distributed COM-Based Code <DCOMDISCUSS.MICROSOFT.COM>
                                  Sent by Distributed COM-Based Code <DCOMDISCUSS.MICROSOFT.COM>

                               To: DCOMDISCUSS.MICROSOFT.COM
                               cc:
                               Subject: Forcing DCOM to use one brach of a redundant network

We want to build into our client application the ability to choose which
branch of a redundant network to use to activate a COM server. The
architecture is as follows:

Client machine with 2 NICs, each NIC connected to a separate network
Server machine with 2 NICs, each NIC connected to a separate network

                   CLIENT
                 / \
                / \
NETWORKA----switch switch----NETWORKB
                \ /
                 \ /
                   SERVER

The design is meant to handle fault tolerance of the network. Normally the
client would be connected to the server over just one of the networks. If
that network fails (or needs maintenance) the client can switch to using
the
other network.

It is typical to use the Machine Name of the server in the remote
activation. Is there a way to control which network gets used in the
CoCreateInstanceEx( ) call when the client tries to activate the server? We
tried specifying the ip address of each NIC in the server machine as the
Machine Name and that works. But what about the situation where the ip
addresses are assigned using DHCP? Can the ip addresses be determined at
runtime, or is there a better way to handle it?

Thank you very much,
Steve Sauls

mailto:ssaulsopto22.com

----------------------------------------------------------------
Users Guide http://discuss.microsoft.com/archives/mailfaq.html
contains important info. Save time, search the archives at
http://discuss.microsoft.com/archives/index.html .
To unsubscribe, mailto:DCOM-signoff-requestDISCUSS.MICROSOFT.COM

Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]