Hi Sebastian, > Le 17 nov. 2022 à 10:50, Sebastian Moeller <moell...@gmx.de> a écrit : > > Hi T. > > > so taking your proposa under consideration I canged the section that threw > you off course to read: > > > • Ethernet with Overhead: SQM can also account for the overhead imposed > by VDSL2 links - add 22 bytes of overhead (mpu 68). Cable Modems (DOCSIS) set > both up- and downstream overhead to 18 bytes (6 bytes source MAC, 6 bytes > destination MAC, 2 bytes ether-type, 4 bytes FCS), to allow for a possible 4 > byte VLAN tag it is recommended to set the overhead to 18 + 4 = 22 (mpu 64). > For FTTH the answer is less clear cut, since different underlaying > technologies have different relevant per-packet-overheads; however > underestimating the per-packet-overhead is considerably worse for > responsiveness than (gently) overestimating it, so for FTTH set the overhead > to 44 (mpu 84) unless there is more detailed information about the true > overhead on a link available. > • None: All shaping below the physical gross-rate of a link requires > correct per-packet overhead accounting to be precise, so None is only useful > if approximate shaping is sufficient, say if you want to clamp a guest > network to at best ~50% of the available capacity or similar tasks, but even > then configuring an approximate correct per-packet-overhead is recommended > (overhead 44 (mpu 84) is a decent default to pick). > > > I hope this is explicit enough.
Yes this looks a lot better, thank you. Although I must confess that it certainly feels counter-intuitive that for ethernet (and FTTH) we suggest a higher overhead than e.g. VDSL2/cable (which themselves run off an ethernet interface). I would also like to see some info about ppp vs ethernet interface in there (matching your previous email): unless you beat me to it I will add it. I also think the « details » page needs to be reformatted a bit, it’s very dense and relevant info is all over the place and not very well organized. I’ll try to get around improving that. Thanks _______________________________________________ Cake mailing list Cake@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/cake
