On 08/03/2015 07:53 PM, Russell Bryant wrote: > On 08/03/2015 07:47 PM, Ben Pfaff wrote: >> On Mon, Aug 03, 2015 at 07:44:03PM -0400, Russell Bryant wrote: >>> On 08/03/2015 07:42 PM, Ben Pfaff wrote: >>>> Thank you for such a careful review! >>>> >>>> I incorporated all of these comments and applied this to master. >>> >>> Awesome work, Ben! >> >> Thanks. It really needs testing, though--I haven't done end-to-end >> testing of any kind, so there are probably bugs in some of the flows. > > OK. Now that it's in master OpenStack CI will be running it heavily. > That's all on a single host, though. > > I'll be sure to do some manual multi-node testing, as well. >
I've got things set up and am hitting a problem. My environment is OpenStack with two hypervisors. OpenStack attaches a DHCP server to the virtual networks. I noticed that VMs local to the DHCP server did not get a response. Remote VMs did get a response from the DHCP server. It seems the handling of broadcast packets skips output to any local ports. > table=17, > priority=100,metadata=0x1,dl_dst=01:00:00:00:00:00/01:00:00:00:00:00 > actions=set_field:0xffff->reg7,resubmit(,32) Note the resubmit only to table 32. Table 32 is for output to remote hypervisors. Output to local ports is table 33. It seems the fix is for this flow to also output to table 33. I'm still working on the patch, though. > table=33, priority=100,reg7=0xffff,metadata=0x1 > actions=set_field:0x1->reg7,resubmit(,34),set_field:0x4->reg7,resubmit(,34),set_field:0x3->reg7,resubmit(,34),set_field:0x2->reg7,resubmit(,34) -- Russell Bryant _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev