All of lore.kernel.org
 help / color / mirror / Atom feed
* [LARTC] interface level b/w capping using HTB
@ 2003-02-05  9:21 hm
  2003-02-05 21:17 ` Stef Coene
  0 siblings, 1 reply; 2+ messages in thread
From: hm @ 2003-02-05  9:21 UTC (permalink / raw)
  To: lartc

[-- Attachment #1: Type: text/plain, Size: 1032 bytes --]

hi all

           the following is the setup i have done.

eth0    globals ip
eth1 internal ip 10.1.1.0/24  --->>>shd be capped to max 256Kbit
linux 8.0 with htb 3.6

tc qdisc add dev eth1 root handle 1:0 htb default 2
tc class add dev eth1 parent 1:0 classid 1:1 htb rate 250Kbit ceil 250Kbit
tc class add dev eth1 parent 1:1 classid 1:2 htb rate 128Kbit ceil 250Kbit

then i created leaves from 1:5 to 1:50 each getting 16Kbit and ceil 20Kbit
a range of 50 ipes
e.g. 
tc class add dev eth1 parent 1:2 classid 1:5 htb rate 16Kbit ceil 20Kbit
tc filter add dev eth1 parent 1:0 protocol ip pref 5 u32 match dst 10.1.1.5/32 flowid 1:5

like that....

although for sometime this setup works fine but suddenly it goes totally unbouded.
i have gone thru the pages at docum.org and found out that if sum of child calsses exceed parent rate then parent rate is not respected.

then what shd i do if i have to cap outgoing b/w on eth1 stricktly to 256Kbit only.
hope to get an early response.


regds
hm




[-- Attachment #2: Type: text/html, Size: 2795 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [LARTC] interface level b/w capping using HTB
  2003-02-05  9:21 [LARTC] interface level b/w capping using HTB hm
@ 2003-02-05 21:17 ` Stef Coene
  0 siblings, 0 replies; 2+ messages in thread
From: Stef Coene @ 2003-02-05 21:17 UTC (permalink / raw)
  To: lartc

On Wednesday 05 February 2003 10:09, hm wrote:
> hi all
>
>            the following is the setup i have done.
>
> eth0    globals ip
> eth1 internal ip 10.1.1.0/24  --->>>shd be capped to max 256Kbit
> linux 8.0 with htb 3.6
Cool.  Linux 8.0.  Where can I download it?  I thought they are working on 
getting 2.5.x stable and now they already is 8.0 :)

> tc qdisc add dev eth1 root handle 1:0 htb default 2
> tc class add dev eth1 parent 1:0 classid 1:1 htb rate 250Kbit ceil 250Kbit
> tc class add dev eth1 parent 1:1 classid 1:2 htb rate 128Kbit ceil 250Kbit
Why rate = 158 Kbit ??

> then i created leaves from 1:5 to 1:50 each getting 16Kbit and ceil 20Kbit
> a range of 50 ipes
> e.g.
> tc class add dev eth1 parent 1:2 classid 1:5 htb rate 16Kbit ceil 20Kbit
> tc filter add dev eth1 parent 1:0 protocol ip pref 5 u32 match dst
> 10.1.1.5/32 flowid 1:5
>
> like that....
>
> although for sometime this setup works fine but suddenly it goes totally
> unbouded. i have gone thru the pages at docum.org and found out that if sum
> of child calsses exceed parent rate then parent rate is not respected.
>
> then what shd i do if i have to cap outgoing b/w on eth1 stricktly to
> 256Kbit only. hope to get an early response.
Solution is easy.  Make sure sum of child rates = parent rates.  
So if you have 45 classes (from 1:5 to 1:50) and a total rate of 256kbit, then 
use rate = 256kbit/45.

Stef

-- 

stef.coene@docum.org
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net


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

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2003-02-05 21:17 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-02-05  9:21 [LARTC] interface level b/w capping using HTB hm
2003-02-05 21:17 ` Stef Coene

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.