On 04/26/2016 08:49 AM, Cole Robinson wrote: > If 'virsh change-media --live --config ...' is requested, the > inactive XML is used as the template for the media change/eject. > If the runtime XML has diverged from the inactive XML, say via > other update operations, those changes may be undone since the > command will sync all non-path XML with the inactive config. > > To 'fix' this would involve two passes of XML editing and calls > to UpdateDevice, which is kind of a pain for a case that won't > affect most users. So let's just document it instead > > https://bugzilla.redhat.com/show_bug.cgi?id=1243342 > --- > tools/virsh.pod | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/tools/virsh.pod b/tools/virsh.pod > index 90ab47d..4c1b7fc 100644 > --- a/tools/virsh.pod > +++ b/tools/virsh.pod > @@ -2794,6 +2794,9 @@ the hypervisor's implementation. > Both I<--live> and I<--config> flags may be given, but I<--current> is > exclusive. If no flag is specified, behavior is different depending > on hypervisor. > +If both I<--live> and I<--config> are specified, the inactive device XML > +is used as the template for the media change. If your I<--live> device XML > +has received runtime changes, they may be removed. > If I<--print-xml> is specified, the XML that would be used to change media is > printed instead of changing the media. > > Actually I'm dropping this. I see that Nitesh has patches to properly fix just this type of issue for detach-interface, so it's better to extend that pattern to change-media and detach-disk too - Cole -- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list