This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-bugs] [Bug 338] Using more that 1 nic results in erratic behavior,

To: xen-bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-bugs] [Bug 338] Using more that 1 nic results in erratic behavior, broken broadcast domains etc..
From: bugzilla-daemon@xxxxxxxxxxxxxxxxxxx
Date: Mon, 24 Oct 2005 17:07:32 +0000
Delivery-date: Mon, 24 Oct 2005 17:07:36 +0000
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-bugs-request@lists.xensource.com?subject=help>
List-id: Xen Bugzilla <xen-bugs.lists.xensource.com>
List-post: <mailto:xen-bugs@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-bugs>, <mailto:xen-bugs-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-bugs>, <mailto:xen-bugs-request@lists.xensource.com?subject=unsubscribe>
Reply-to: bugs@xxxxxxxxxxxxxxxxxx
Sender: xen-bugs-bounces@xxxxxxxxxxxxxxxxxxx

------- Additional Comments From tedkaz@xxxxxxxxxxxxx  2005-10-24 17:07 -------
      --------------------------  subnet A eth1
         |                 |
     dom0-1              dom0-2
        |                  |              subnet B eth0
|   |   |   |   |   |   |   |   |   |    |   |   | 
1-1 1-2 1-3 1-4 1-5 1-6 1-7 1-8 1-9 1-10 2-1 2-2 2-3

This is the least problematic, ultimately one should be able to break out the
domU's aon mulitple nics. Assume from above subnet B * X. I have experimented
with many variations with three nics on domO's. If you think about the isolation
aspects such a scenario with domU to domU traffic isolated on Layer 2 this is
most appealing. Consider that this layer 2 link could can be provided with
encapsulation, 802.1q and or MPLS tags.

My bigger problem right now is bug #359 which keeps me from testing more.
I can use changeset 7396 but their are quite a few network changes after that
that invalidate any testing.

I do have a UP P4 running
xen_changeset : Wed Oct 19 06:53:00 2005 +0100 7425:7c951e3eb5a

It does appear to be behaving with the toplogy from above, but my testing
capabilities are impaired at this time but it does not show any obviuos problems
with repsect to the behavior of the bridge. I actually am tracking another issue
on that one where the dom0 blows up. Be nice to have a dedicated set of machines
for every issue one sees :-) 

Configure bugmail: 
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Xen-bugs mailing list

<Prev in Thread] Current Thread [Next in Thread>