[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: EF and BE traffic--unexplained behaviour





On Sat, 5 Aug 2000, Aditya Chugh wrote:

> Hi Jamal , Muneyb....and friends,
> 
> Could you please suggest the correct parameter values to be used with RED
> Your help is sincerely appreciated.
> Thanks a lot,
> Aditya
> 

Actually, i dont even know if it is RED; however, you copied this from the
efcbq example and changed those RED parameters, so RED was the first thing
i noticed (some of them like the burst seem to make sense but some
like the threshold values are baffling). So go back to the drawing
board: start again with those parameters from efcbq example and change one
at a time and repeat your tests. When you run into trouble post more
complete information like the stats outputs for one (tc -s, ifconfig etc)

cheers,
jamal

PS:- those example scripts are _just examples_ not religious 
definitions. It's good to start from there, but depending on your
requirements you should really start from scratch. It _really_ helps for
you to understand the meanings of all the associated parameters you are
changing before you change them or you associate them with values.
the diffserv parameters are well documented in the draft.


> jamal wrote:
> 
> > The RED parameters look weird....
> >
> > On Fri, 4 Aug 2000, Muneyb Minhazuddin wrote:
> >
> > >
> > > Can I ask what type of traffic is it?
> > >
> > > tcp/udp
> > >
> > > What is the value of the tos byte that you mark the EF traffic with?
> > >
> > > 0xb8/0x2e
> > >
> > > Did you monitor the tcpdump o/p to see if the packets are correctly
> > > marked? Use ttt or tcptrace or any other tcpdump analysing package to plot
> > > the graphs for you.
> > >
> > > I am mystified by your throughput results, you should atleast get 3 to
> > > 3.5Mpbs for both EF and BE. Can you give me an idea of your network, is it
> > > just 2 machines connected back to back?
> > >
> > > Cheers
> > > Muneyb
> > >
> > > On Thu, 3 Aug 2000, Aditya Chugh wrote:
> > >
> > > > Hello Everybody,
> > > >
> > > > I am using my modified version of TTCP (with which I can set the TOS
> > > > byte of my packets ) to test the throuput and the time taken to traverse
> > > >
> > > > the path from A to B.
> > > > A and B are both linux machines  connected to the same local LAN. The
> > > > kernel of A
> > > > is diffserv enabled,while B has plain vanilla Linux.
> > > > I run a modified version of "EFCBQ" - the standard script to set the
> > > > diffserv queues for EF and BE.
> > > > I assign both EF and BE  4 Mbit of rate ( out of total b/w of 10Mbit ) I
> > > >
> > > > make both the queues "bounded and isolated" ...so they can neither
> > > > borrow nor lend b/w. The weight assigned to each is 1.  (Script attached
> > > >
> > > > below)
> > > >
> > > > When I send both EF and BE packets simultaneously from A to B , I get EF
> > > >
> > > > b/w about 467 Kb/sec
> > > > and BE bandwidth about 340Kb/sec---which is OK according to me.
> > > > When I send only EF packets from A to B the b/w is reduced to around 260
> > > >
> > > > Kb/sec
> > > > When I send only BE packets the b/w I get is around 336 Kb/sec ?
> > > > These results are getting repeated consistently. (measurements have been
> > > >
> > > > taken 8-10 times for each case)
> > > > Isn't this strange ??  How do we explain it ?
> > > > I am totally stumped ....and  am unable to come up with a coherent
> > > > explanation
> > > >
> > > > Any help/advice/explanation from you will be sincerely appreciated.
> > > > Thanks in advance.Please reply soon.
> > > > regards,
> > > > Aditya
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > The script I used:---
> > > >
> > > > #!/usr/bin/perl
> > > > #
> > > > $TC = "/usr/local/iproute2/tc/tc" ;
> > > > $DEV = "dev eth0";
> > > > print "$TC qdisc add $DEV handle 1:0 root dsmark indices 64
> > > > set_tc_index\n";
> > > > print "$TC filter add $DEV parent 1:0 protocol ip prio 1 tcindex ".
> > > >   "mask 0xfc shift 2\n";
> > > > print "$TC qdisc add $DEV parent 1:0 handle 2:0 cbq bandwidth ".
> > > >  "10Mbit cell 8 avpkt 1000 mpu 64\n";
> > > > #
> > > > # EF class
> > > > #
> > > > print "$TC class add $DEV parent 2:0 classid 2:1 cbq bandwidth ".
> > > >  "10Mbit rate 4Mbit avpkt 1000 prio 1 bounded isolated ".
> > > >  "allot 1514 weight 1 maxburst 4 \n";
> > > > # packet fifo for EF?
> > > > print "$TC qdisc add $DEV parent 2:1 pfifo limit 4\n";
> > > > print "$TC filter add $DEV parent 2:0 protocol ip prio 1 ".
> > > >    "handle 0x2e tcindex classid 2:1 pass_on\n";
> > > > #
> > > > # BE class
> > > > #
> > > > print "#BE class(2:2) \n";
> > > > print "$TC class add $DEV parent 2:0 classid 2:2 cbq bandwidth ".
> > > >  "10Mbit rate 4Mbit avpkt 1000 prio 7 bounded isolated allot 1514 weight
> > > >
> > > > 1 ".
> > > >  "maxburst 4  split 2:0 defmap 0xffff \n";
> > > > print "$TC qdisc add $DEV parent 2:2 red limit 4KB ".
> > > >    "min 20kBit max 4KB burst 4 avpkt 1000 bandwidth 10Mbit ".
> > > >    "probability 0.02\n";
> > > > print "$TC filter add $DEV parent 2:0 protocol ip prio 2 ".
> > > >    "handle 0 tcindex mask 0 classid 2:2 pass_on\n";
> > > >
> > > >
> > > >
> > > >
> > >
> > > _________________________________________________________
> > > Muneyb Minhazuddin - Telecommunications Research Engineer
> > > CSIRO Telecommunication and Industrial Physics
> > > Marsfield, NSW, Australia.
> > >
> > > Phone no. : 61 2 9372 4113
> > > FAX       : 61 2 9372 4490
> > > e-mail    : mminhazu@tip.csiro.au
> > > Home Page : http://www-networks.tip.csiro.au/~mminhazu
> > > ---------------------------------------------------------
> > >
> > >
>