<html> <head> <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> @Gao see the other thread Managing eths in vm from hv too. We weren't able to order based on vnets, though I agree its seems intuitive. But our test wasn't thorough. Eric says: "Libvirt passes the devices to the qemu command line in the same order that they appear in the XML. It is not libvirt's fault if qemu then takes the command line arguments and presents it to the guest in such a way that the guest can initialize the devices in the order in which a bus scan finds them, rather than in the order they were presented on the command line." as far as I can tell the bridges are always setup first, then the direct VF connections. Whether this is caused by our hardware or Qemu or PCI I have no idea. Now, I am trying the idea of editing the /etc/udev/rules.d/70* file to move the MAC addresses around. Results are mixed. very puzzling. John <div class="moz-cite-prefix">On 12/08/2012 08:21 AM, Gao Yongwei wrote: </div> <blockquote cite="mid:CAAHBLWGHWihHY1WCQPatRpNn_Dg27OwDW1Wbyq2WA7ZGXKMfag@mail.gmail.com" type="cite"> <div class="gmail_quote">2012/12/7 john fisher <<a moz-do-not-send="true" href="mailto:john@jpfisher.net" target="_blank">john@jpfisher.net</a>> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <target dev='vnet0'/></blockquote> </div> I think the vnet* can affect the order in vm. </blockquote> <pre class="moz-signature" cols="120">-- John Fisher</pre> </body> </html>