RE: QoS for mVPN

From: Robert McCallum (rmccallum@thrupoint.net)
Date: Wed Aug 06 2008 - 12:43:10 ART


Layer 3 qos does not work on LAN cards. You need to use mls qos structures
to do this and you cannot see the bandwidth usage on such interfaces - hence
the reason why they are soooooo cheap.

> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On
> Behalf Of Reinhold Fischer
> Sent: 06 August 2008 15:53
> To: Harindha Fernando
> Cc: comserv@groupstudy.com; ccielab@groupstudy.com
> Subject: Re: QoS for mVPN
>
> Harin,
>
> you are assuming that QoS is not working, because your output
> policy does not show matching packets, right? Can you show
> the qos config of the outgoing interface? Since this is a LAN
> card, all the restrictions/limitations of cat6500 qos do apply ...
>
> reinhold
>
>
> On Wed, Aug 06, 2008 at 05:27:35PM +0300, Harindha Fernando wrote:
> > Hi,
> >
> > we are using SIP-400 with 5x1G SPA as the uplinks to MPLS,
> and 6548 as
> > the client facing interface.
> >
> > yes, unicast QoS is working fine. but does the GRE maps the
> inner QoS
> > to it's header or we have use qos pre-classify ?
> >
> > Rgds,
> > Harin
> >
> > On Wed, Aug 6, 2008 at 5:18 PM, Reinhold Fischer
> > <Reinhold.Fischer@gmx.net>wrote:
> >
> > > On Wed, Aug 06, 2008 at 04:14:11PM +0300, Harindha Fernando wrote:
> > > > Hi,
> > > >
> > > > I have been trying to set QoS for mVPN in MPLS QoS but
> I am still
> > > > unable
> > > to
> > > > achieve it. below is the scenario...
> > > >
> > > > Reciever1 -->PE1-->P1-->P2-->PE2--->Source
> > > > |
> > > > |
> > > > P3--->PE3--->Receiver2
> > > >
> > > > Core - PIM SSM
> > > > Customer mVPN - PIM-SSM with static SSM mapping
> > > >
> > > > mVPN is working fine, but I need to offer some QoS in the core.
> > > >
> > > > at PE2 customer interface I have a input policy-map setting all
> > > muliticast
> > > > traffic on that interface, set to DSCP AF41. the MPLS
> interface
> > > > towards core I have an output policy allocating BW
> based on dscp
> > > > or exp marking
> > > and
> > > > also I do set exp values. DSCP AF41 = Exp 4
> > > >
> > > > when I put a output policy in the PE1 and PE3 customer
> interfaces
> > > > to
> > > match
> > > > DSCP AF41 I do not see any matching packets.
> > > >
> > > > since mVPN traffic is going through GRE tunnels I don't see any
> > > possibility
> > > > of setting the DSCP values in the GRE header as this Tunnel
> > > > interfaces
> > > are
> > > > not configurable.
> > > >
> > > > can any body tell how can we achieve the QoS for mVPN ?
> > > >
> > > > I am using 7600s with RSP720 and 12.2(33)SRC1.
> > > >
> > >
> > > Harin,
> > >
> > > what kind of linecards are in use on the 7600s? Are you sure that
> > > your unicast QoS is OK and you have only issues with the
> multicast
> > > QoS? When the multicast traffic is encapsulated with the
> multipoint
> > > GRE header, it should copy the QoS information of the original
> > > packet into the new header. I did not verify this on 7600/SRC but
> > > with 6500/SXH2a. Just to make sure: You did consider that
> multicast
> > > is forwarded as IP and not MPLS and your QoS settings are
> adjusted to this?
> > >
> > > The cisco-nsp mailing list is also a good place to ask
> about such stuff....
> > >
> > > hth
> > >
> > > reinhold
>
> _____________________________________________________________________
> Subscription information: http://www.groupstudy.com/list/comserv.html
>

Note:The information contained in this message may be privileged and confidential and protected from disclosure . If the reader of this message is not the
intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any
dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us
immediately by replying to the message and deleting it from your computer. Thankyou. ThruPoint Ltd.

Blogs and organic groups at http://www.ccie.net



This archive was generated by hypermail 2.1.4 : Mon Sep 01 2008 - 08:15:29 ART