From: "Serguei I. Ivantsov" <admin@gsc-game.kiev.ua>
To: lartc@vger.kernel.org
Subject: Re[2]: [LARTC] Why SFQ does not drops packets
Date: Fri, 01 Jun 2001 19:57:27 +0000 [thread overview]
Message-ID: <marc-lartc-99142706118046@msgid-missing> (raw)
In-Reply-To: <marc-lartc-99138592919119@msgid-missing>
Hello Wingtung.Leung,
Friday, June 01, 2001, 6:58:44 PM, you wrote:
WL> On Fri, 1 Jun 2001, Serguei I. Ivantsov wrote:
>> Please explain me why my class 10:300 does not drops packets?
>> My class is bounded and isolated - but 8kbit!\x1783bps!!!
>>
>>
>> class cbq 10:300 parent 10:1 leaf 801b: rate 8Kbit (bounded,isolated) prio 5
>> Sent 470737 bytes 388 pkts (dropped 0, overlimits 7465)
>> rate 1783bps 1pps backlog 40p
>> borrowed 0 overactions 290 avgidle -786633 undertime 612289
WL> I'm not sure what 1783bps means. If it is "bytes/sec" more packets should
WL> be dropped, if it is "bits/sec", everything seems to be ok.
Looking at the MRTG graph I can say that bps="bytes per second"
WL> How do you get that output? Is it documented somewhere, so I can try it?
To get the list of classes with details use "tc -s class ls dev eth0"
To get the list of qdiscs with details use "tc -s qdisc ls dev eth0"
If you want to check bandwidth consumed with every class use estimator
in class definition "est 1sec 4sec"
WL> tc class add dev eth0 parent 10:1 classid 10:300 est 1sec 4sec \
WL> cbq bandwidth 64Kbit rate \
WL> 8Kbit allot 1514 weight 1Kbit prio 5 maxburst 1 avpkt 1000 \
WL> bounded isolated
After number of tests I come to conclusion that the only queueing
discipline which can DROP packets and thereby throttle bandwith is
TBF. But it does not allow to throttle bandwith dynamically, does not
allow to relocate bandwith between classes:(
Please put me right if I wrong.
--
Best regards,
Serguei mailto:admin@gsc-game.kiev.ua
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/2.4Routing/
next prev parent reply other threads:[~2001-06-01 19:57 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-06-01 8:58 [LARTC] Why SFQ does not drops packets Serguei I. Ivantsov
2001-06-01 15:58 ` Wingtung.Leung
2001-06-01 17:41 ` Ethy H. Brito
2001-06-01 17:41 ` Ramin Alidousti
2001-06-01 19:57 ` Serguei I. Ivantsov [this message]
2001-06-01 20:23 ` Re[2]: " Serguei I. Ivantsov
2001-06-01 20:35 ` Ramin Alidousti
-- strict thread matches above, loose matches on Subject: below --
2001-06-02 0:40 Re[2]: " Wingtung.Leung
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=marc-lartc-99142706118046@msgid-missing \
--to=admin@gsc-game.kiev.ua \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox