On Wed, Mar 24, 2021 at 9:22 PM Ben Pfaff <b...@ovn.org> wrote: > > On Tue, Mar 23, 2021 at 09:22:43PM -0400, Ihar Hrachyshka wrote: > > When vlan-passthru is true for LS, for untagged localnet ports, tagged > > VLAN traffic originating from VIFs should be passed through intact since > > the VLAN header belongs to VIF user, not the localnet port fabric. > > > > This patch adds multinode test cases to cover scenarios where packets > > traverse through fabric layer, for both tagged and untagged (tag=0) > > localnet ports. > > > > Signed-off-by: Ihar Hrachyshka <ihrac...@redhat.com> > > Thanks for adding ddlog support. It looks correct to me.
Hi Ihar, The patch LGTM. I just have one comment. I noticed that there is slight difference in the way ovn-northd and ovn-northd-ddlog sets the vlan-passthru in the options column of Port_Binding. If the logical switch's other_config:vlan-passthru is set to false i.e ovn-nbctl set logical_switch ls other_config:vlan-passthru=false then, ovn-northd doesn't set "vlan-passthru=false" in the options column of port binding. It will clear the option if it was set earlier. Whereas ovn-northd-ddlog, sets "vlan-passthru=false". Technically this will not have any effect on the functionality. Although I'd suggest making it consistent. Either make ovn-northd to set "vlan-passthru=false" or make ovn-northd-ddlog to clear vlan-passthru. I'm fine either way. Thanks Numan > _______________________________________________ > dev mailing list > d...@openvswitch.org > https://mail.openvswitch.org/mailman/listinfo/ovs-dev > _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev