Also somehow related to #2
Right now, the algorithm is a simple token bucket, which just drop packages. This is obviously not right since, ideally, we should queue them. Also, the limits are not super accurate.
Investigate a better way of doing this (directly with or something involving eBPF), which can also depend on the approach we end up doing in #2 .
Some stuff for reference:
https://github.com/torvalds/linux/blob/master/samples/bpf/hbm.c
https://man7.org/linux/man-pages/man8/tc.8.html
https://wiki.linuxfoundation.org/networking/ifb
https://netdevconf.info/0x14/pub/slides/55/slides.pdf
https://github.com/xdp-project/bpf-examples/blob/main/traffic-pacing-edt/edt_pacer_vlan.c
https://netfilter.org/projects/libnetfilter_queue/doxygen/html/
Also somehow related to #2
Right now, the algorithm is a simple token bucket, which just drop packages. This is obviously not right since, ideally, we should queue them. Also, the limits are not super accurate.
Investigate a better way of doing this (directly with or something involving eBPF), which can also depend on the approach we end up doing in #2 .
Some stuff for reference:
https://github.com/torvalds/linux/blob/master/samples/bpf/hbm.c
https://man7.org/linux/man-pages/man8/tc.8.html
https://wiki.linuxfoundation.org/networking/ifb
https://netdevconf.info/0x14/pub/slides/55/slides.pdf
https://github.com/xdp-project/bpf-examples/blob/main/traffic-pacing-edt/edt_pacer_vlan.c
https://netfilter.org/projects/libnetfilter_queue/doxygen/html/