<div dir="ltr">>I gave this a shot, it doesn't route the packets back trhough the<br>>underlying interface...<br>>perhaps policy routing?<br><div><br></div><div>tried it with policy routing too, packets get dropped when i point them to </div><div>the ifb interfaces fast/slow. tried both dummy and ifb.</div><div><br></div><div>ip route add $custip dev fast table 10</div><div>ip route add $custip dev slow table 11</div><div>ip rule add from $cacheip to $custip lookup 10 priority 10</div><div>ip rule add from all to $custip lookup 11 priority 11</div><div><br></div><div>im not sure if we can route it to ifb this way,( maybe only works with tc redirection?) as i feel that i have just set up a routing loop.</div><div><br></div><div>i did not have time to recompile the kernel for imq, as jonathan mentioned but will try that too,</div><div>did not try the ifb method mentioned by jonathan yet. </div><div><br></div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Mar 30, 2016 at 5:46 AM, Dave Taht <span dir="ltr"><<a href="mailto:dave.taht@gmail.com" target="_blank">dave.taht@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I gave this a shot, it doesn't route the packets back trhough the<br>
underlying interface...<br>
perhaps policy routing?<br>
<br>
TC=/home/d/git/tc-adv/tc/tc<br>
IFACE=eno1<br>
<br>
F=2601:640:4103:56c0:2ab4:103a:39c5:a43a<br>
S=2601:640:4103:56c0:120d:7fff:0:647<br>
<br>
ip link add $IFACE name fast type dummy # maybe ifb?<br>
ip link add $IFACE name slow type dummy # maybe ifb?<br>
<br>
ip link set fast up<br>
ip link set slow up<br>
<br>
$TC qdisc add dev fast root cake bandwidth 20Mbit<br>
$TC qdisc add dev slow root cake bandwidth 5Mbit<br>
$TC qdisc add dev $IFACE root cake bandwidth 20Mbit<br>
<br>
ip route add $F dev fast metric 1<br>
ip route add $S dev slow metric 1<br>
Dave Täht<br>
Let's go make home routers and wifi faster! With better software!<br>
<a href="https://www.gofundme.com/savewifi" rel="noreferrer" target="_blank">https://www.gofundme.com/savewifi</a><br>
<div class="HOEnZb"><div class="h5"><br>
<br>
On Tue, Mar 29, 2016 at 4:31 PM, Dave Taht <<a href="mailto:dave.taht@gmail.com">dave.taht@gmail.com</a>> wrote:<br>
> An overall ISP in tc need exposed by this discussion is some means of<br>
> mapping multiple ipv4 and ipv6 addresses and netmasks into something<br>
> that will return a (key,value) pair. This would work something like<br>
> ipset does, although what you would return is not "present or not" but<br>
> present and a value<br>
><br>
> insert customers 1.1.1.1,1<br>
> insert customers 2001:db1::/64,1<br>
> insert customers 2.2.2.2,2<br>
> insert customers 2001:db2::/64,1<br>
><br>
> then on the relevant path you'd set up the qdisc hierarchy and do a<br>
> lookup into that to get the right number to go to the right cake<br>
> instance. You'd also have to do a longest prefix match into the above,<br>
> so a 1x1 hash won't do.<br>
><br>
> the massive tc filter option discussed already does not scale to a<br>
> random number of customers with randomly different numbers of ip<br>
> addresses, types, and netmasks. Code like this must exist in dedicated<br>
> devices already, CMTSes, BRASes, deep packet inspection devices, etc.<br>
><br>
> Secondly, in the case of cake the hierarchy could just be a bunch of<br>
> somewhat unassociated queues, not htb or drr, letting cake do the<br>
> scheduling of deliveries.<br>
><br>
> Regrettably I know of few ISPs that are actively using linux in any<br>
> way they have sources to. I suspect a few dslams are linux based, but<br>
> nobody's talking.<br>
><br>
> ...<br>
><br>
> Another way to maybe get there is to use the ip route functionality<br>
> instead and send stuff to virtual devices layered on top of the real<br>
> device.<br>
><br>
> ip route add from :: to <a href="http://1.1.1.0/24" rel="noreferrer" target="_blank">1.1.1.0/24</a> dev dev1<br>
> ip -6 route add from :: to 2001:db1::/64 dev dev1<br>
> ip -6 route add from :: to 2001:db2::/64 dev dev2<br>
> ip route add from :: to <a href="http://2.2.2.0/24" rel="noreferrer" target="_blank">2.2.2.0/24</a> dev dev2<br>
><br>
> Then the reverse would be out one of two devices, one device<br>
> configured for the "fast, local cache server", the other for the<br>
> regular internet.<br>
><br>
> solution too long to fit in the margins of this email.<br>
_______________________________________________<br>
Cake mailing list<br>
<a href="mailto:Cake@lists.bufferbloat.net">Cake@lists.bufferbloat.net</a><br>
<a href="https://lists.bufferbloat.net/listinfo/cake" rel="noreferrer" target="_blank">https://lists.bufferbloat.net/listinfo/cake</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr">Thanx and regd's.<br><br>Allan.<br><br></div></div>
</div>