Wasn’t thinking Quantum Networking would pop up this way… The “marginal link” can’t simply be measurable packet loss. There are L2 networks, wired/wireless that can have measurable packet loss “as properly designed”. Not good, and you could avoid it, but it is expected behavior. Reasonable sized 10Base5 networks were fun.
Hopefully the whole Homenet effort is to keep the Home; with multiple MAC types, speeds, wired and wireless links in a large L2 bridge domain from being a “marginal link”. On 2/19/15, 12:54 PM, "Juliusz Chroboczek" <[email protected]> wrote: >> Seems like ³marginal link² is associated with the IGP behavior > >No. A marginal link is simply one that has a measurable amount of packet >loss. > >On a properly functioning wired network, you don't have any marginal >links: a link is either up or down, and a link that is up has negligible >amount of packet loss (the link, not the switches). A marginal link is >one that's neither up or down -- it's somewhere in between. > >This is made somewhat more complicated by layer 2 ARQ (for example, in >802.11b a link that is marginal at the physical layer will usually appear >as merely slow for unicast traffic, but clearly marginal for multicast; >weirder things are possible in 802.11a/g/n). > >> Is this close to what a ³marginal link² is? > >No. Being marginal is a property of a link, it has nothing to do with the >IGP. > >-- Juliusz > >_______________________________________________ >homenet mailing list >[email protected] >https://www.ietf.org/mailman/listinfo/homenet > _______________________________________________ homenet mailing list [email protected] https://www.ietf.org/mailman/listinfo/homenet
