All of lore.kernel.org
 help / color / mirror / Atom feed
* [LARTC] htb and iptables marked packets question
@ 2004-12-27 15:27 Felipe Ceglia
  2004-12-28 19:31 ` Stef Coene
  2004-12-28 22:28 ` Francisco Pereira
  0 siblings, 2 replies; 3+ messages in thread
From: Felipe Ceglia @ 2004-12-27 15:27 UTC (permalink / raw)
  To: lartc

Hi folks,

I have a strange situation. When I add branches to the tree, everything 
goes to the default class.
The error might be obvious, but I cannot find it. I would really appreciate 
your help.


this works, nothing goes to "1:9999":
#############################################################################
/sbin/iptables -F -t mangle
/sbin/tc qdisc del dev eth1 root > /dev/null 2>&1
/sbin/tc qdisc add dev eth1 root handle 1: htb default 9999
/sbin/tc class add dev eth1 parent 1: classid 1:1 htb rate 100Mbit ceil 100Mbit
/sbin/tc class add dev eth1 parent 1:1 classid 1:9999 htb rate 128kbit
/sbin/tc class add dev eth1 parent 1:1 classid 1:4 htb rate 100Mbit
/sbin/tc class add dev eth1 parent 1:4 classid 1:28 htb rate 510kbit
#
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.0.0/16 -j MARK 
--set-mark 4
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 4 fw 
flowid 1:4
#
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.200.0/24 -j MARK 
--set-mark 28
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 28 fw 
flowid 1:28
#############################################################################



when I add subclasses to "1:28", traffic goes to "1:34" and "1:35", but 
exceeding traffic goes to "1:9999" instead of being placed at "1:28":
#############################################################################
/sbin/iptables -F -t mangle

/sbin/tc qdisc del dev eth1 root > /dev/null 2>&1
/sbin/tc qdisc add dev eth1 root handle 1: htb default 9999
/sbin/tc class add dev eth1 parent 1: classid 1:1 htb rate 100Mbit ceil 100Mbit
/sbin/tc class add dev eth1 parent 1:1 classid 1:9999 htb rate 128kbit
/sbin/tc class add dev eth1 parent 1:1 classid 1:4 htb rate 100Mbit
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.0.0/16 -j MARK 
--set-mark 4
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 4 fw 
flowid 1:4
#
/sbin/tc class add dev eth1 parent 1:4 classid 1:28 htb rate 510kbit
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.200.0/24 -j MARK 
--set-mark 28
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 28 fw 
flowid 1:28
#
/sbin/tc class add dev eth1 parent 1:28 classid 1:34 htb rate 64kbit
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.200.8/32 -j MARK 
--set-mark 34
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 34 fw 
flowid 1:34
#
/sbin/tc class add dev eth1 parent 1:28 classid 1:35 htb rate 64kbit
/sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.200.14/32 -j 
MARK --set-mark 35
/sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 35 fw 
flowid 1:35
#############################################################################



iptables says:
iptables -L -n -v -t mangle
Chain PREROUTING (policy ACCEPT 184K packets, 105M bytes)
  pkts bytes target     prot opt 
in     out     source               destination

Chain INPUT (policy ACCEPT 115K packets, 63M bytes)
  pkts bytes target     prot opt 
in     out     source               destination

Chain FORWARD (policy ACCEPT 68617 packets, 42M bytes)
  pkts bytes target     prot opt 
in     out     source               destination

Chain OUTPUT (policy ACCEPT 123K packets, 63M bytes)
  pkts bytes target     prot opt 
in     out     source               destination

Chain POSTROUTING (policy ACCEPT 191K packets, 105M bytes)
  pkts bytes target     prot opt 
in     out     source               destination
  1827 1990K 
MARK       all  --  *      eth1    0.0.0.0/0            192.168.0.0/16 
MARK set 0x4
  1827 1990K 
MARK       all  --  *      eth1    0.0.0.0/0            192.168.200.0/24 
MARK set 0x1c
   174  245K 
MARK       all  --  *      eth1    0.0.0.0/0            192.168.200.8 
MARK set 0x22
    30 10519 
MARK       all  --  *      eth1    0.0.0.0/0            192.168.200.14 
MARK set 0x23

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [LARTC] htb and iptables marked packets question
  2004-12-27 15:27 [LARTC] htb and iptables marked packets question Felipe Ceglia
@ 2004-12-28 19:31 ` Stef Coene
  2004-12-28 22:28 ` Francisco Pereira
  1 sibling, 0 replies; 3+ messages in thread
From: Stef Coene @ 2004-12-28 19:31 UTC (permalink / raw)
  To: lartc

On Monday 27 December 2004 16:27, Felipe Ceglia wrote:
> Hi folks,
>
> I have a strange situation. When I add branches to the tree, everything
> goes to the default class.
> The error might be obvious, but I cannot find it. I would really appreciate
> your help.
<snip>

> MARK set 0x1c
This is in hex.  Put 0x before the mark in the iptables and the match in the 
tc command to make sure they are all in hex.


Stef
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [LARTC] htb and iptables marked packets question
  2004-12-27 15:27 [LARTC] htb and iptables marked packets question Felipe Ceglia
  2004-12-28 19:31 ` Stef Coene
@ 2004-12-28 22:28 ` Francisco Pereira
  1 sibling, 0 replies; 3+ messages in thread
From: Francisco Pereira @ 2004-12-28 22:28 UTC (permalink / raw)
  To: lartc

Hi.

Felipe Ceglia wrote:
> Hi folks,
> 
> I have a strange situation. When I add branches to the tree, everything 
> goes to the default class.
> The error might be obvious, but I cannot find it. I would really 
> appreciate your help.

I think it's the expected behaviour.

> 
> this works, nothing goes to "1:9999":

perhaps all the traffic you generate to test was marked by the iptables 
rules and then classified by that.
btw in this order the rule for 192.168.200.0/24 never matches, because 
the rule for 192.168.0.0/16 matches always first.

> ############################################################################# 

> /sbin/iptables -F -t mangle
> /sbin/tc qdisc del dev eth1 root > /dev/null 2>&1
> /sbin/tc qdisc add dev eth1 root handle 1: htb default 9999

That's whats the default parameter is for                ^^^^
Traffic not classified goes to the default.

> /sbin/tc class add dev eth1 parent 1: classid 1:1 htb rate 100Mbit ceil 100Mbit
> /sbin/tc class add dev eth1 parent 1:1 classid 1:9999 htb rate 128kbit
> /sbin/tc class add dev eth1 parent 1:1 classid 1:4 htb rate 100Mbit
> /sbin/tc class add dev eth1 parent 1:4 classid 1:28 htb rate 510kbit
> #
> /sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.0.0/16 -j MARK --set-mark 4
> /sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 4 fw flowid 1:4
> #
> /sbin/iptables -A POSTROUTING -o eth1 -t mangle -d 192.168.200.0/24 -j MARK --set-mark 28
> /sbin/tc filter add dev eth1 protocol ip parent 1:0 prio 3 handle 28 fw flowid 1:28
> ############################################################################# 
> 
> 
> 
> 
> when I add subclasses to "1:28", traffic goes to "1:34" and "1:35", but 
> exceeding traffic goes to "1:9999" instead of being placed at "1:28":
> ############################################################################# 
> 
> /sbin/tc qdisc add dev eth1 root handle 1: htb default 9999

That's whats the default parameter is for                ^^^^

Regards,
Francisco.
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2004-12-28 22:28 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-12-27 15:27 [LARTC] htb and iptables marked packets question Felipe Ceglia
2004-12-28 19:31 ` Stef Coene
2004-12-28 22:28 ` Francisco Pereira

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.