Bill,
This should be doable. The reason that Muse
uses IP addresses is that the receiving machine may not be able to resolve a
given host name. I’m thinking this should be a configurable option (both
in COSMOS and in Muse). I’m a committer on both projects – if we
think this is something we should do in Muse then I can open the bug there, and
we’ll just need to log-roll the IP for the next release of Muse into
COSMOS.
Cheers,
Joel
-----Original Message-----
From:
cosmos-dev-bounces@xxxxxxxxxxx [mailto:cosmos-dev-bounces@xxxxxxxxxxx] On Behalf Of Muldoon, William H
Sent: Wednesday, August 29, 2007
9:56 AM
To: Ebright, Don; Cosmos Dev
Subject: RE: [cosmos-dev] IP v6
compliance for COSMOS DC
Don
I suggest that as a general rule the COSMOS DC should avoid using IP addresses
whenever possible. For example, the COSMOS DC should replace the IP addresses
in the muse WSDM EPRs with hostnames:
INFO: [ID = 'ResourceInitialized'] The resource at
'/org.eclipse.cosmos.dc.runtime.osgi.OSGiRuntimeHost' has been initialized.
Bound<wsa:EndpointReference
xmlns:wsa="http://www.w3.org/2005/08/addressing">
<wsa:Address>http://192.168.0.103:8080/cosmos/services/org.eclipse.cosmos.dc.runtime.osgi.OSGiRuntimeHost</wsa:Address>
<wsa:ReferenceParameters>
<muse-wsa:ResourceId
xmlns:muse-wsa="http://ws.apache.org/muse/addressing">RuntimeIdentifierValue</muse-wsa:ResourceId>
</wsa:ReferenceParameters>
</wsa:EndpointReference>
Hostnames are more
“stable” than IP addresses. This is important because the EPRs may
be registered in various components across the network (like the Brokers) and
the IP address may change after a machine reboot. Also IPv6 support introduces
multiple addresses and formats for the same physical machine, potentially
causing a problem maintaining unique EPRs in the registries.
Regards
Bill
From: Ebright,
Don [mailto:Don.Ebright@xxxxxxxxxxxxx]
Sent: Tuesday, August 28, 2007 10:16
AM
To: Cosmos Dev
Cc: Muldoon, William H
Subject: RE: [cosmos-dev] IP v6
compliance for COSMOS DC
Jimmy,
It would be good if we
can get specific enough to be able to understand the
configurations that we would need to test. I assume that there is no
requirement to support a v6-only network so I would expect most things
to just work by encapsulating v4 traffic.
Don
From:
cosmos-dev-bounces@xxxxxxxxxxx [mailto:cosmos-dev-bounces@xxxxxxxxxxx] On Behalf Of Mohsin, Jimmy
Sent: Tuesday, August 28, 2007
9:31 AM
To: Cosmos Dev
Cc: Muldoon, William H
Subject: RE: [cosmos-dev] IP v6
compliance for COSMOS DC
Don,
Even though I probably
will not be on, can we please cover this tomorrow (even if briefly)? If I
cannot make it, Mr. Muldoon will represent me. To ++ what you say, If we
can, I would STRONGLY prefer the DC to stay independent of this issue.
The IP v6 compliance has extensive legal implications at our end.
Thanks,
Jimmy Mohsin
+1-609-635-1703
From: cosmos-dev-bounces@xxxxxxxxxxx
[mailto:cosmos-dev-bounces@xxxxxxxxxxx] On
Behalf Of Ebright, Don
Sent: Tuesday, August 28, 2007
7:31 AM
To: Cosmos Dev
Subject: RE: [cosmos-dev] IP v6
compliance for COSMOS DC
I hope that we can
avoid storing or processing IP addresses as such so we can remain
relatively independent of this issue. That said, if we need
to support configurations that employ IP v6, we will need to test
them.
Perhaps we could discuss
this on tomorrow's architecture call.
Don
The contents of this e-mail are intended for the named addressee only.
It contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose it
to anyone else. If you received it in error please notify us immediately and
then destroy it.
From:
cosmos-dev-bounces@xxxxxxxxxxx [mailto:cosmos-dev-bounces@xxxxxxxxxxx] On Behalf Of Mohsin, Jimmy
Sent: Monday, August 27, 2007 6:18
PM
To: Cosmos Dev
Subject: [cosmos-dev] IP v6
compliance for COSMOS DC
All,
Do you see any IP v6 compliance
issues for the COSMOS DC?
Thanks,
Jimmy Mohsin
+1-609-635-1703