* [Cake] a start at vething cake for 200 subscribers
@ 2018-07-28 21:03 Dave Taht
0 siblings, 0 replies; only message in thread
From: Dave Taht @ 2018-07-28 21:03 UTC (permalink / raw)
To: Cake List
so I was tempted to try. I didn't enable namespaces. just coded this
little bit up...
a very quick test of this setup showed I wasn't routing packets through the
veth interfaces so I figure network namespaces, and... oh look! the sun's out!
#!/bin/bash
# Create the bridge
ip link add cable type bridge
ip link set cable up
# Create network namespaces and veth interfaces
# do I even need to use namespaces?
# do I need ips just to pass it through to the bridge?
ip route flush proto 51
for client in `seq 2 10`
do
c=c${client}
ip netns del $c
ip netns add $c
# probly all this needs to end up in namespaces
ip link add ${c}.l type veth peer name ${c}.r
ip link set ${c}.l up
ip link set ${c}.r up
ip addr add 10.10.0.1/24 dev ${c}.l
ip addr add 10.10.0.$client/24 dev ${c}.r
# fixme add ipv6 subnets
# I kind of wanted the route table to be global tho
ip route add 10.10.0.$client/32 dev ${c}.l proto 51
# fixme add default route back?
tc qdisc replace dev ${c}.l root cake bandwidth 20mbit flows
done
# fixme hook up to the bridge
# fixme hang actual clients off the bridge? or the namespace
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2018-07-28 21:03 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-07-28 21:03 [Cake] a start at vething cake for 200 subscribers Dave Taht
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox