From: Dave Temkin (dave@ordinaryworld.com)
Date: Mon Oct 10 2005 - 17:00:49 GMT-3
I know this was addressed before, but I've got a quick opinion call for
everyone as to whether or not they think this is appropriate..
Given a question regarding anything with FRTS, is it appropriate to use
the MQC? Do you think when the proctor is evaluating the final config,
they'd check for an alternate solution aside from configuring it directly
on the interface?
ie, I just tested the following two scenarios:
1) set the DE bit on any traffic greater than 512 bytes for traffic on
DLCI 102:
class-map match-all DLCI102and512
match fr-dlci 102
match packet length min 512
policy-map SETFRDE
class DLCI102and512
set fr-de
interface Serial0/0/0:0
service-policy output DLCI102and512
This seems to work perfectly fine (note the out DE pkts):
DLCI = 102, DLCI USAGE = LOCAL, PVC STATUS = ACTIVE, INTERFACE =
Serial0/0/0:0
input pkts 638 output pkts 620 in bytes 122240
out bytes 120780 dropped pkts 0 in pkts dropped 0
out pkts dropped 0 out bytes dropped 0
in FECN pkts 0 in BECN pkts 0 out FECN pkts 0
out BECN pkts 0 in DE pkts 0 out DE pkts 93
2) On DLCI 103, the CIR is 1024K. Ensure all traffic conforms to this
with a BC of 16000 and a be of 8000
Police all ICMP traffic to 512K and set
fr-de on all other ICMP traffic, and set the DE bit on all FTP traffic.
class-map match-all DLCI103ICMP
match fr-dlci 103
match protocol icmp
class-map match-all FTPOVER256K
match protocol ftp
policy-map PARENT
class class-default
police cir 1024000 bc 16000 be 8000
service-policy DOSTUFF
policy-map DOSTUFF
class DLCI103ICMP
police cir 512000
conform-action set-frde-transmit
exceed-action drop
class FTPOVER256K
set fr-de
interface Serial0/0/0:0
service-policy output PARENT
Can anyone think of any reasons why this approach wouldn't work for most
FRTS questions?
Thanks,
-Dave
This archive was generated by hypermail 2.1.4 : Sun Nov 06 2005 - 22:00:50 GMT-3