Re: Some more test on ingress, ifb, fwmark

Linux Advanced Routing and Traffic Control

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Marco Gaiarin wrote:
Mandi! Andy Furniss
   In chel di` si favelave...

Not sure really - need more info on the rest of the setup like what
qdiscs/queue lengths (if any) you set on the leafs. The nature and
speed of the wan.

My complete IFB setup; egress, indeed, is roughly the same.
Related to an 3M/2M WiMAX link.

I have no experience of WiMAX, but it may be different enough from "normal" links to cause problems.

If it's rated 3M down then the 3mbit ceils and rate on parent will be too high.


# I setup the interface, matching reduced MTU with the real interface
ifconfig ifb2 mtu 1476 txqueuelen 600 up

I wouldn't bother setting mtu here (though the test I just did looked like it didn't do anything anyway). The txqueuelen seems long, but as you add sfq you'll get the default for that (128).


# Then i setup htb on egress for ifb2
/sbin/tc qdisc del dev ifb2 root
/sbin/tc qdisc add dev ifb2 root handle 1: htb
/sbin/tc class add dev ifb2 parent 1: classid 1:1 htb rate 3000kbit ceil 3000kbit burst 16k cburst 8k mtu 1476

I also wouldn't set mtu here it's not the same meaning as doing it on a real if - IIRC it also needs 14 adding for traffic on/from eth)

/sbin/tc class add dev ifb2 parent 1:1 classid 1:10 htb rate 1200kbit ceil 3000kbit burst 6k cburst 8k quantum 2952 prio 1
/sbin/tc qdisc add dev ifb2 parent 1:10 handle 10: sfq perturb 10
/sbin/tc class add dev ifb2 parent 1:1 classid 1:20 htb rate 600kbit ceil 3000kbit burst 3k cburst 8k quantum 2952 prio 2
/sbin/tc qdisc add dev ifb2 parent 1:20 handle 20: sfq perturb 10
/sbin/tc class add dev ifb2 parent 1:1 classid 1:30 htb rate 300kbit ceil 2400kbit burst 1k cburst 6k quantum 7380 prio 3
/sbin/tc qdisc add dev ifb2 parent 1:30 handle 30: sfq perturb 10
/sbin/tc class add dev ifb2 parent 1:1 classid 1:40 htb rate 300kbit ceil 1500kbit burst 1k cburst 4k quantum 7380 prio 4
/sbin/tc qdisc add dev ifb2 parent 1:40 handle 40: sfq perturb 10
/sbin/tc class add dev ifb2 parent 1:1 classid 1:50 htb rate 300kbit ceil 600kbit burst 1k cburst 1k quantum 2952 prio 5
/sbin/tc qdisc add dev ifb2 parent 1:50 handle 50: sfq perturb 10

Trying tu understand your sentence on previous email, eg:

but it is flawed because the ${UPLINK}kbit and $[9*$UPLINK/10]kbit on
the leafs will not be constrained by the parent - so go over limit.

you say that does not work because 'ceil' misses from original
wondershaper?
Eg, using 'ceil' (on parent and childeren, indeed) my setup works?
I've understood right?

Your set up looks fine - the rates on the child classes add up to <= the rate on the parent.

The lartc example has rate on childs that add up to more than the parent and the rate on the parent won't stop them getting their full rates (parent rate only limits when they are borrowing)

Anyway, go further:

# Then i setup ingress on ifb2 (probably not useful at all, it was a try...)
/sbin/tc qdisc del dev ifb2 ingress
/sbin/tc qdisc add dev ifb2 handle ffff: ingress
/sbin/tc filter add dev ifb2 parent ffff: protocol ip prio 50 u32 match ip src 0.0.0.0/0 police rate 3000kbit burst 16k drop flowid :1

ingress doesn't work on ifb so I would get rid of this.

--
To unsubscribe from this list: send the line "unsubscribe lartc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [LARTC Home Page]     [Netfilter]     [Netfilter Development]     [Network Development]     [Bugtraq]     [GCC Help]     [Yosemite News]     [Linux Kernel]     [Fedora Users]
  Powered by Linux