All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stephen Lally <stlally@cs.tcd.ie>
To: lartc@vger.kernel.org
Subject: [LARTC] simulating RED using tcsim question
Date: Wed, 04 Aug 2004 18:05:59 +0000	[thread overview]
Message-ID: <41112587.8010709@cs.tcd.ie> (raw)

I'm trying to setup a bottlekneck scenario so I can see the effects of 
the RED qdisc on a congested interface.
My example simulation code is at the bottom of this email if someone 
wants to see it.

The following is a snippet form TCSIMs output.

0.602000 E : 0x80bc080 40 : inputB_eth0: 45000028 00000000 40060000 ...
0.602000 D : 0x80bc080 40 : inputB_eth0: 45000028 00000000 40060000 ...
0.602003 E : 0x80bbfd8 40 : router_eth2: 45000028 00000000 40060000 ...
0.602003 * : 0x80bbfd8 40 : router_eth2: enqueue returns CN (2)
0.602003 E : 0x80bc080 40 : router_eth2: 45000028 00000000 40060000 ...
0.602003 * : 0x80bc080 40 : router_eth2: enqueue returns CN (2)
0.604000 E : 0x80bbfd8 40 : inputA_eth0: 45000028 00000000 40060000 .

As I hoped, packets seem to get dropped as marked by '*' when the 
interface gets congested.

On closer inspection (with -v) tcsim output is:

0.182000 D : 0x80bc0a0 40 : inputB_eth0: 45000028 00000000 40060000 ...
0.182003 E : 0x80bbff8 40 : router_eth2: 45000028 00000000 40060000 ...
0.182003 e : 0x80bbff8 40 : <1> red (2:0) returns CN (2)
0.182003 e : 0x80bbff8 40 : <0> dsmark (1:0) returns CN (2)
0.182003 * : 0x80bbff8 40 : router_eth2: enqueue returns CN (2)
0.182003 E : 0x80bc0a0 40 : router_eth2: 45000028 00000000 40060000 ...
0.182003 e : 0x80bc0a0 40 : <1> red (2:0) returns CN (2)
0.182003 e : 0x80bc0a0 40 : <0> dsmark (1:0) returns CN (2)
0.182003 * : 0x80bc0a0 40 : router_eth2: enqueue returns CN (2)
0.184000 E : 0x80bbff8 40 : inputA_eth0: 45000028 00000000 40060000 ...

I was exspecting the '*' to now print as 'x' to indicate a dropped 
packet rather than some error occuring!

The network setup looks something like this

|inputA_eth0|--100 Mbs------]
                            ]----|router_eth2|---10 Bs------|output_eth1|  
|inputB_eth0|--100 Mbs------]
     
    RED is added router_eth2, router_eth2 is set as a 10 Bs interface to 
force some congestion.


Anyone know what's going on? What does "router_eth2: enqueue returns CN 
(2)" mean?
Any help is much appreciated.

Stephen.


bottleneck.tcsim
-----------------
//inputA to router
host {
        dev inputA_eth0 100 Mbps
}

//inputB to router
host {
        dev inputB_eth0 100 Mbps
}


//the router
host {
        dev router_eth0 100 Mbps
        dev router_eth1 100 Mbps
        dev router_eth2 10 //the bottleneck
        {
                egress {
                        red( min 150 B,
                             max 450 B,
                             burst 200 B,
                             limit 600 B,
                             bandwidth 100 bps,
                             probability 0.4,
                             avpkt 80 B );
                }
        }

        route 10.0.0.0 netmask 255.0.0.0 router_eth2

}

//output destination from router
host {
        dev output_eth0 10 Mbps
        dev output_eth1 100 Mbps
        route 10.0.0.2 netmask 255.0.0.0 output_eth1
}

connect inputA_eth0 router_eth0
connect inputB_eth0 router_eth1
connect router_eth2 output_eth0


every 0.002s until 5s send inputA_eth0 TCP_PCK($tcp_sport = PORT_HTTP);

every 0.002s until 5s send inputB_eth0 TCP_PCK($tcp_sport = PORT_SSH);

time 6s
end



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

                 reply	other threads:[~2004-08-04 18:05 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=41112587.8010709@cs.tcd.ie \
    --to=stlally@cs.tcd.ie \
    --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.