Seams you missed last Ganeticon 2018 in Düsseldorf... Google released last 2.16.0 stable and handed code over to the community. Ganeti might become a Debian project/fork in near future.
Is this patch enough or is the one from Adrew (first post) still needed? Am Donnerstag, 11. Oktober 2018 15:15:54 UTC+2 schrieb George Diamantopoulos: > > Turns out the solution is a lot simpler: one needs to provide a name for > the tap device when adding additional queues. From > https://www.kernel.org/doc/Documentation/networking/tuntap.txt: > > The device allocation is the same as before, and if user wants to create > multiple queues, TUNSETIFF with the same device name must be called many > times with IFF_MULTI_QUEUE flag. > > So we just need to set "name" after the first iteration of the loop found > in the function, if it's not already set: > > --- a/lib/hypervisor/hv_kvm/netdev.py > +++ b/lib/hypervisor/hv_kvm/netdev.py > @@ -189,6 +189,9 @@ > > tapfds.append(tapfd) > > + if ( _ == 0 and name == "" ): > + name = struct.unpack("16sh", res)[0].strip("\x00") > + > # Get the interface name from the ioctl > ifname = struct.unpack("16sh", res)[0].strip("\x00") > > I'd create a pull request, but I doubt anyone is looking at them these > days... > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <https://github.com/ganeti/ganeti/issues/1268#issuecomment-428950828>, or > mute > the thread > <https://github.com/notifications/unsubscribe-auth/ADQ7IgVkfXMB2EujCeP097zLqLm11w27ks5uj0SIgaJpZM4PXTkz> > . >
