Open dtaht opened 10 years ago
magnific0
commented
6 years ago @dtaht Indeed you are correct, this should be implemented and actually shouldn't be a lot of work. I imagine that only the tc filter add commands need to be updated. However, I want to investigate and test this thoroughly. So for now work in progress.
dtaht
commented
6 years ago I kind of gave up on wondershaper and worked on the sqm-scripts. Currently twiddling with sch_cake.
kostasapostolopoulos
commented
6 years ago It will be nice to have support for IPv6, since it becomes more and more relevant. All Google domains, use IPv6 by default for example.
ghost
commented
6 years ago Have a look at http://lartc.org/howto/lartc.adv-filter.ipv6.html and https://home.regit.org/2013/09/using-tc-with-ipv6-and-ipv4/ I will test this next weeks.
jakajancar
commented
5 years ago Just removing all occurrences of "protocol ip" seems to work for me.
dtaht
commented
5 years ago Jaka Jančar notifications@github.com writes:
Just removing all occurrences of "protocol ip" seems to work for me.
Can't help but continue to encourage y'all to try cake, fq_codel, and the sqm-scripts instead of wondershaper.
Do some benchmarks...
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.
jakajancar
commented
5 years ago No time for benchmarks (and the images on bufferbloat are all broken).
Checked sqm-scripts... Don't want to install a crapton of files on my Debian box in an unmanaged way (without package manager). Also contains some hardcoded eth0 references, relies on ifupdown (and I've moved on to systemd-networkd).
"good enough"
jakajancar
commented
5 years ago Pull request: https://github.com/magnific0/wondershaper/pull/28
dtaht
commented
5 years ago Good point. Most of our work predates systemd. I bug noted it.
how about
tc qdisc add dev whatever root cake bandwidth XXMbit
biggest flaw of wondershaper is that ipv6 traffic, native or not, is not correctly handled.