From: Lee Sanders <lartc-337@ccp.com.au>
To: lartc@vger.kernel.org
Subject: Re: [LARTC] Consider r2q change
Date: Thu, 20 Oct 2005 14:32:39 +0000 [thread overview]
Message-ID: <200510202232.39935.lartc-337@ccp.com.au> (raw)
In-Reply-To: <001801c5d580$46ec2b20$0700000a@d3lta>
You really should search the list archives. This question has been asked a
dozen times in the last year.
Google also turned this up quite quickly.
http://www.docum.org/docum.org/faq/cache/31.html
To copy and paste some previous answers from the list. You will need to sort
these yourself.
It is documented in HTB manual. The complaint was not displayed
before and sharing was silently wrong.
Quantum of leaf should be between 1500 and 60000 and it is computed
as rate/r2q (or can be supplied independly).
So that if your maximal leaf rate is 1Mbit (120000Bps) then r2q
should be 3. If max rate is 10kbit (1200Bps) then r2q should be 1.
It is warning only and will not affect functionality, only precision.
devik
Smallest rate : 16kbit = 2 kilobyt / r2q (\x10) = 200. And this is < 1500. So
you get warnings.
Biggest rate : 100mbit = 12.5 mbyte / r2q = 1.2 Mbyte > 60.000. So you get
warnings.
If you do
tc qdisc add dev eth0 root handle 1: htb default 10 r2q 1
Smallest rate : 16kbit = 2kilobyte / r2k = 2000. And this is > 1500. So no
warnings.
Biggest rate : 100mbit = 12.5 mbyte / r2q = 12.5 Mbyte > 60.000. So you get
warnings. But you can overrule the quantum :
tc class add dev eth0 parent 1:1 classid 1:11 htb rate 128kbit burst 2k
quantum 60000
Quantum is used when 2 classes are getting more bandwidth then the rate. So
it's only important for sharing the remaining bandwidth. In that case, each
class may send quantum bytes.
If I understand the source, that message occours when you have a leaf
class whose calculated quantum (rate/r2q) is less than 1000 byte and the
kernel set the quantum to 1000 byte.
I've read that the quantum should be bigger than the MTU so I don't
understand why the kernel don't use a minimum quantum value bigger than
1500. Someone has a clue?
From kernel-source-2.6.10/net/sched/sch_htb.c
if (!hopt->quantum && cl->un.leaf.quantum < 1000) {
printk(KERN_WARNING "HTB: quantum of class %X is small. Consider r2q
change.\n", cl->classid);
cl->un.leaf.quantum = 1000;
I hope this helps.
Regards,
Lee
On Thu, 20 Oct 2005 10:12 pm, James Lista wrote:
> nsider r2q change.
> HTB: quantum of class 10002 is big. Consider r2q change.
> HTB: quantum of class 10030 is small. Consider r2q change.
> HTB: quantum of class 10031 is small. Consider r2q change.
_______________________________________________
LARTC mailing list
LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
prev parent reply other threads:[~2005-10-20 14:32 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-10-20 14:12 [LARTC] Consider r2q change James Lista
2005-10-20 14:23 ` m.innocenti
2005-10-20 14:32 ` Lee Sanders [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200510202232.39935.lartc-337@ccp.com.au \
--to=lartc-337@ccp.com.au \
--cc=lartc@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.