[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [1/3] [XEN] gnttab: Add new op unmap_and_replace


  • To: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
  • From: Keir Fraser <keir@xxxxxxxxxxxxx>
  • Date: Mon, 04 Jun 2007 14:15:20 +0100
  • Cc: Xen Development Mailing List <xen-devel@xxxxxxxxxxxxxxxxxxx>, "Daniel P. Berrange" <berrange@xxxxxxxxxx>
  • Delivery-date: Mon, 04 Jun 2007 06:13:32 -0700
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: AcemqmajpVke8BKdEdyEowAX8io7RQ==
  • Thread-topic: [Xen-devel] [1/3] [XEN] gnttab: Add new op unmap_and_replace

On 4/6/07 13:39, "Herbert Xu" <herbert@xxxxxxxxxxxxxxxxxxx> wrote:

> That's weird.
> 
> There is only one call to 'brctl addbr' and that's via network-bridge.
> So if xend's only been started once then I can't see how you could end
> up with two bridges.
> 
> Can you add a set -x to xen-network-common.sh to see who's creating
> two bridges?

It's also invoked by create_bridge() inside xend. So I had to blow away
/var/lib/xend/state and that's fixed the problem.

Anyway, all seems better now except for some hotplug slowness on my test
machine for the first VM I create after rebooting the host (possibly a stale
lock file that I'm having to wait to time out).

However, there *are* still a lot of references to xenbr0 in the tools/
directory, including in our example configs. Should we get rid of these
references, or change to 'bridge=eth0'?

 -- Keir


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.