From: Brian McGahan (bmcgahan@internetworkexpert.com)
Date: Thu Feb 19 2004 - 12:38:33 GMT-3
Allan,
There really is no "first interval". Each interval is only
significant to the interval directly before and directly after. Sending of
Be can occur in any interval as long as there is credit. Be only
accumulates if there was unused Bc in previous intervals.
HTH,
Brian McGahan, CCIE #8593
bmcgahan@internetworkexpert.com
Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987 x 705
Outside US: 775-826-4344 x 705
> -----Original Message-----
> From: Allan Wells [mailto:wellse@bigpond.net.au]
> Sent: Wednesday, February 18, 2004 11:51 PM
> To: Brian McGahan
> Subject: Re: Limiting ICMP using CAR
>
> Thanks Brian,
>
> ?
> S0/0 with multiple pvc's line rate 1024k
>
> With FRTS say a pvc has mincir =128,000 and we want to burst to line
> which
> is 1024k
>
> Their is excess traffic on this pvc Does the burst "BE" happen every tc or
> only once at the beggining of each second.
>
> I just remember spemding some time a while ago reading that you only burst
> in the first interval of each second.
>
> Is their any substance to this just a figment of my imagination.
>
>
> Thanks again.
>
>
> Allan
>
>
> ----- Original Message -----
> From: "Brian McGahan" <bmcgahan@internetworkexpert.com>
> To: "'Allan Wells'" <wellse@bigpond.net.au>; <ccielab@groupstudy.com>
> Sent: Thursday, February 19, 2004 10:55 AM
> Subject: RE: Limiting ICMP using CAR
>
>
> > Allan,
> >
> > Tc is indirectly set by configuring Bc. Tc conforms to the formula:
> >
> > Tc = Bc/CIR * 1000
> >
> > The maximum Tc is 125ms, and the minimum is 10ms. Incorrect
> > calculation will result in the default Tc being used for that CIR.
> >
> > Tc is in milliseconds, Bc and Be are in bytes in CAR and in bits in
> > traffic-shaping. CIR/target-rate are always in bits per second.
> >
> > Bc is the amount of data that must be send/received per Tc in order
> > to achieve the CIR/target rate over the interface. This is simply the
> > shaping rate over a smaller interval. The Be is the excess burst per
> > interval that can be sent based on previously accumulated credit.
> >
> > If Bc is consistently used it is permissible that you will never be
> > able to burst in excess (Be).
> >
> >
> > HTH,
> >
> > Brian McGahan, CCIE #8593
> > bmcgahan@internetworkexpert.com
> >
> > Internetwork Expert, Inc.
> > http://www.InternetworkExpert.com
> > Toll Free: 877-224-8987 x 705
> > Outside US: 775-826-4344 x 705
> >
> >
> > > -----Original Message-----
> > > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
> Of
> > > Allan Wells
> > > Sent: Wednesday, February 18, 2004 6:01 PM
> > > To: Brian McGahan; ccielab@groupstudy.com
> > > Subject: Re: Limiting ICMP using CAR
> > >
> > > Hi Brian,
> > >
> > > I have some questions relating to below TS.
> > >
> > > Getting confused : |
> > >
> > > In CAR if you set Bc to 4000 and Be to
> > > > 8000, it means your excess burst is actually 4000.(bytes or bits)
> > >
> > >
> > > Assuming TC is 1/8 =125ms CAR 128kbps
> > >
> > > rate-limit output access-group 100 128000 4000 8000 conform-action
> (are
> bc
> > > and be in bytes here ?)I understand tc is in bits but if bc in this
> > > statement is bytes that would mean TC = 250ms as 4000bytes x 8
> =32000bps
> > > x4
> > > = 128kbps ?
> > >
> > >
> > >
> > > My understanding and therfor confusion is that I thought burst was
> only
> > > happening once per second not every TC in this example 125ms.Does that
> > > ever
> > > happen wherby you could only burst once per second ?
> > >
> > >
> > >
> > >
> > > ----- Original Message -----
> > > From: "Brian McGahan" <bmcgahan@internetworkexpert.com>
> > > To: "'Yasser Aly'" <blackyeyes00@hotmail.com>; <ccie2b@hotmail.com>;
> > > <ccielab@groupstudy.com>
> > > Sent: Thursday, February 19, 2004 1:43 AM
> > > Subject: RE: Limiting ICMP using CAR
> > >
> > >
> > > > > >So, 128,000 kbps dvided by 8 = 16,000 = bits per interval, Tc.
> To
> > > allow
> > > > > >for a burst of 1/4 of this rate, divide by 4. 16,000 / 4 = 4000.
> > > >
> > > > Yes and no. The interval is determined by what you configure. Your
> > > > target rate is 128000, which is entered in bits per second. The
> burst
> > > > committed value, which comes next, is in bytes per interval. Since
> it
> > > is
> > > > specified that the burst should be a quarter of the target, the rate
> is
> > > > (128000/8)/4 = 4000.
> > > >
> > > > The reason that the Be is 4000 is that there is no excess burst.
> > > > When Be = Bc in CAR, there is no burst. This is different from GTS
> and
> > > FRTS
> > > > where the configured Be is above Bc.
> > >
> > >
> > > >
> > > >
> > > > HTH,
> > > >
> > > > Brian McGahan, CCIE #8593
> > > > bmcgahan@internetworkexpert.com
> > > >
> > > > Internetwork Expert, Inc.
> > > > http://www.InternetworkExpert.com
> > > > Toll Free: 877-224-8987 x 705
> > > > Outside US: 775-826-4344 x 705
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On
> Behalf
> > > Of
> > > > > Yasser Aly
> > > > > Sent: Tuesday, February 17, 2004 7:15 PM
> > > > > To: ccie2b@hotmail.com; ccielab@groupstudy.com
> > > > > Subject: RE: Limiting ICMP using CAR
> > > > >
> > > > > Hi Packet Man,
> > > > >
> > > > > Bc and Be are configured in Bytes not bits. I am not sure about
> > > whether
> > > > > it
> > > > > is per Tc or not.
> > > > >
> > > > > Needs clarification.
> > > > >
> > > > > Regards,
> > > > > Yasser
> > > > >
> > > > > >From: "Packet Man" <ccie2b@hotmail.com>
> > > > > >To: blackyeyes00@hotmail.com, ccielab@groupstudy.com
> > > > > >Subject: RE: Limiting ICMP using CAR
> > > > > >Date: Tue, 17 Feb 2004 18:27:11 -0500
> > > > > >
> > > > > >Yasser,
> > > > > >
> > > > > >I think the issue you might be having is due to the different
> units
> > > used
> > > > > >for the different parameters. Remember that the 1st parameter,
> cir,
> > > is
> > > > > in
> > > > > >bits per SECONDS while the Bc and Be are in bits per Tc which is
> 1/8
> > > of
> > > > > a
> > > > > >second.
> > > > > >
> > > > > >So, 128,000 kbps dvided by 8 = 16,000 = bits per interval, Tc.
> To
> > > allow
> > > > > >for a burst of 1/4 of this rate, divide by 4. 16,000 / 4 = 4000.
> > > > > >
> > > > > >Now, given the requirements you stated, I think the answer that
> was
> > > > > >provided is wrong. I think it should be
> > > > > > CIR Bc
> > > Be
> > > > > >rate-limit output access-group 100 128000 16000 4000 conform-
> action
> > > > > >transmit exceed-action drop
> > > > > >
> > > > > >Hopefully, if I'm mistaken, someone will immediately say so and
> point
> > > out
> > > > > >my mistake. (It wouldn't be the 1st time)
> > > > > >
> > > > > >HTH
> > > > > >
> > > > > >
> > > > > >>From: "Yasser Aly" <blackyeyes00@hotmail.com>
> > > > > >>Reply-To: "Yasser Aly" <blackyeyes00@hotmail.com>
> > > > > >>To: ccielab@groupstudy.com
> > > > > >>Subject: Limiting ICMP using CAR
> > > > > >>Date: Tue, 17 Feb 2004 18:07:34 -0500
> > > > > >>
> > > > > >>Hi group,
> > > > > >>
> > > > > >> The task is as follows: Configure E0/0 such that ICMP traffic
> is
> > > > > limited
> > > > > >>to 128Kbps.
> > > > > >>Allow for a burst of 1/4th of this rate.
> > > > > >>
> > > > > >>
> > > > > >>The answer for this question was as follows:
> > > > > >>
> > > > > >>acess-group 101 permit icmp any any
> > > > > >>!
> > > > > >>interface Ethernet0/0
> > > > > >>rate-limit output access-group 100 128000 4000 4000 conform-
> action
> > > > > >>transmit exceed-action drop
> > > > > >>
> > > > > >>I am confused why Bc and Be were set this way. Would you kindly
> > > explain
> > > > > >>and give more detail on general concepts used when configuring
> Bc
> > > and
> > > Be
> > > > > >>while using CAR.
> > > > > >>
> > > > > >>Regards,
> > > > > >>Yasser
> > > > > >>
> > > > >
> >>_________________________________________________________________
> > > > > >>The new MSN 8: smart spam protection and 2 months FREE*
> > > > > >>http://join.msn.com/?page=features/junkmail
> > > > > >>http://join.msn.com/?page=dept/bcomm&pgmarket=en-
> > > > >
> > >
> ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarke
> > > > > t%3den-ca
> > > > > >>
> > > > >
> > >
> >>_______________________________________________________________________
> > > > > >>Please help support GroupStudy by purchasing your study
> materials
> > > from:
> > > > > >>http://shop.groupstudy.com
> > > > > >>
> > > > > >>Subscription information may be found at:
> > > > > >>http://www.groupstudy.com/list/CCIELab.html
> > > > > >
> > > > >
> > > > > _________________________________________________________________
> > > > > STOP MORE SPAM with the new MSN 8 and get 2 months FREE*
> > > > > http://join.msn.com/?page=dept/bcomm&pgmarket=en-
> > > > >
> > >
> ca&RU=http%3a%2f%2fjoin.msn.com%2f%3fpage%3dmisc%2fspecialoffers%26pgmarke
> > > > > t%3den-ca
> > > > >
> > > > >
> > >
> _______________________________________________________________________
> > > > > Please help support GroupStudy by purchasing your study materials
> > > from:
> > > > > http://shop.groupstudy.com
> > > > >
> > > > > Subscription information may be found at:
> > > > > http://www.groupstudy.com/list/CCIELab.html
> > > >
> > > >
> _______________________________________________________________________
> > > > Please help support GroupStudy by purchasing your study materials
> from:
> > > > http://shop.groupstudy.com
> > > >
> > > > Subscription information may be found at:
> > > > http://www.groupstudy.com/list/CCIELab.html
> > >
> > >
> _______________________________________________________________________
> > > Please help support GroupStudy by purchasing your study materials
> from:
> > > http://shop.groupstudy.com
> > >
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Fri Mar 05 2004 - 07:13:51 GMT-3