On Thursday 12 December 2002 12:19, Abraham van der Merwe wrote:
> Hi Stef!
>
> > I did some small tests :
>
> [snip]
>
> > Conclusion : if you want to know how traffic will be shaped, you have to
> > follow some basic rules.
> > I will update docum.org tonight.
>
> You're right. I did some tests this weekend as well and realised that you
> have to stick with the basics. The problem was that it was not so clear to
> me how HTB behaves when the child's rates/ceil exceed the parent's.
>
> Some other questions I'd like to ask you while I'm at it:
>
> 1) if you have:
>
> 1:1
> / \
> / \
> 1:2 1:3
> / \ / \
> / \ / \
> 1:4 1:5 1:6 1:7
>
> If
>
> 1:2 = prio 1
> 1:3 = prio 2
>
> 1:4 = prio 10
> 1:5 = prio 11
>
> 1:6 = prio 5
> 1:7 = prio 6
>
> Is 1:4, 1:5 evaluated first or 1:6, 1:7? Iow, does HTB start at the root
> node, prioritize its children and, subprioritize their children, etc. or
> is only the leaf nodes' priorities important (i.e. 1:6, 1:7, 1:4, 1:5)
I don't know. But I think only the leaf nodes are important. I have some
work to do tonight, but I hope to test it this evening.
> 2) if a class is evaluated and there is some bandwidth available to borrow,
> is the first non-congested child given all bandwidth that is available and
> then the next is given the rest and so on or is each child given a quantum
> of bandwidth, then if there is some bandwidth left, a second round and so
> on?
If each class received his rate and the parent has some bandwidth left, the
bandwidth is distributed in a round robin so each class can send quantum
bytes.
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/
