> > > It may also be possible to remove the need for DLPI for IPv4, and > > incidentally simplify the code a fair amount, by adding a kernel > > option to allow transmission and reception of UDP packets over > > interfaces that are plumbed but not marked IFF_UP. This is left for > > future work. > > How aggressively do you think we should try to get the use of DLPI out > of dhcpagent?
This is on the list of things to do as part of Clearview, since we need it to allow DHCP to work with IPMP (which is long overdue and often requested). Jim's recent DAD work made a bit of headway by removing one of the two things the DHCP client used DLPI for. > We would need to revise how IP determines to bring up its DLPI provider > (DL_ATTACH_REQ, DL_BIND_REQ) which today doesn't happen until the > interface is both IFF_UP and has a non-zero IPv4 address. I don't think that's true. As long as there's an IFF_UP address -- regardless of if it's zero -- I believe you can send packets on the interface. If this doesn't work, then outbound load spreading is broken with IPMP when using only link-based failure detection and there are fewer data addresses than IP interfaces in the group. -- meem _______________________________________________ networking-discuss mailing list [email protected]
