From: Bob Sinclair (bsinclair@netmasterclass.net)
Date: Fri Apr 02 2004 - 14:44:51 GMT-3
William,
I labbed up a scenario that appears to answer your first question in the
affirmative.
Flow of traffic is generator to CAT1 to CAT2 to R1. All Links are trunks.
I marked traffic inbound to CAT1 with AF11. When it came inbound the trunk
between CAT1 and CAT2, I trusted the CoS. On R1 I got hits on an
access-list statement matching on DSCP 8.
This is just what the maps suggest should happen. AF11 gets mapped to CoS
1. CoS 1 gets mapped to DSCP 8 (CS1). So without any particular
configuration, AF11 got mapped to CoS 1 on the link between Cat1 and Cat2.
Regarding your second question: I have not labbed up a proof, but I would
bet a cup of coffee that the dscp-cos map would be controlling. I don't
think there is a way for the switch to "remember" the original CoS.
HTH,
Bob Sinclair
CCIE #10427, CISSP, MCSE
www.netmasterclass.net
----- Original Message -----
From: "William Chen" <kwchen@netvigator.com>
To: <ccielab@groupstudy.com>
Sent: Friday, April 02, 2004 11:30 AM
Subject: Catalyst QoS - Cos.
> Dear all,
>
> I understand that in Catalyst 3550 QoS. The egress queue of the packets
> will depends on the DSCP-COS map. If the egress port is a trunk port, will
> the packets actually be marked with COS value according to the DSCP-COS
map?
>
> Moreover, if I use "mls qos tust cos pass-thru dscp" at the ingress, I
> understand that the COS-DSCP map is ignored and the internal DSCP will use
> the DSCP value of the received packets. How about the egress queue? The
> packets go into the egress queue according to the COS or DSCP-COS map?
>
> Best Regards,
> William Chen
>
> _______________________________________________________________________
> 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 : Mon May 03 2004 - 19:48:41 GMT-3