Re: [Xen-devel] [patch] xen udev rule interfering with openvpn

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Wed, 2012-04-25 at 14:16 +0100, Roger Pau Monne wrote:
> Ian Campbell escribiÃ:
> > On Wed, 2012-04-25 at 11:14 +0100, Ian Campbell wrote:
> >> On Wed, 2012-04-25 at 11:11 +0100, Ian Jackson wrote:
> >>> Ian Campbell writes ("Re: [Xen-devel] [patch] xen udev rule interfering 
> >>> with openvpn"):
> >>>> So for vifname="foo" it is a no-brainer to call the vif "foo" and the
> >>>> emulated tap device "foo-emu".
> >>>>
> >>>> But what about the case where no vifname is given, in that case vif is
> >>>> named "vif<DOM>.<DEV>". But what to call the tap? Previously I was
> >>>> changing the name from tap<DOM>.<DEV>  to xentap<DOM>.<DEV>  but perhaps
> >>>> now "vif<DOM>.<DEV>-emu" makes more sense/is more consistent?
> >>> I think either is fine.  While we're changing it it probably makes
> >>> sense to use "vif..." as indeed it is more consistent.
> >> OK, vifX.Y and vifX.Y-emu it is...
> >
> > This turned out to be a bit more complex than I was expecting, mostly
> > because the use of "vifname" with tap devices was already broken. I
> > think this does the right things with each type of device.
> >
> > Roger, not sure what knock on effect this has on your series. The main
> > thing is likely to be that you needn't find the vifname since you
> > actually want the standard name not the user supplied once on most
> > occasions.
> 
>  From what I see, I have to pass the name returned by
> libxl__device_nic_devname to the hotplug scripts, is that right?

AFAICT pretty much, yes.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel