From: ccie2be (ccie2be@nyc.rr.com)
Date: Wed Jun 08 2005 - 15:38:45 GMT-3
John,
I just got up to this part of the lab.
1st of all, this ecn thing is a new feature listed under in the new features
section of QoS in 12.3.
I had to read that doc a few times before it really sunk in.
But, here's the essence of it.
BECN's and FECN's might not get through the network when the network is
congested. And, it's the network itself that determines whether to indicate
congestion by setting these indicators.
However, if a sending or receiving router can see congestion by monitoring
it's queues, wouldn't it be better for the router to indicate congestion to
other side by setting an indicator itself.
When wred is configured on an interface, the router is monitoring it's
queues because if the queue gets to filled, it starts dropping packets. So,
if it does that maybe it should tell the other side, it's congested before
it's queues fill up and actually dropping packets.
Since this feature is new, not all devices support it which is why the task
stipulated "if supported".
ECN bits are set in such a way routers know if the other side supports it.
This is what ecn does.
WRED takes advantage of the fact that when packets are lost, TCP slows down.
By using ECN, the transmission rate can be slowed down before packets are
lost.
HTH, Tim
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of John
Matus
Sent: Tuesday, June 07, 2005 3:01 PM
To: bmcgahan@internetworkexpert.com; ccielab@groupstudy.com
Subject: RE: IE LAB7 VOL 2 QOS comparison
well sh@t brian :-)
does my config satify the requirements of the task tho?
>From: "Brian McGahan" <bmcgahan@internetworkexpert.com>
>To: "John Matus" <john_matus@hotmail.com>,<ccielab@groupstudy.com>
>Subject: RE: IE LAB7 VOL 2 QOS comparison
>Date: Tue, 7 Jun 2005 13:39:49 -0400
>
>John,
>
> No they are not the same. BECN is a flag in the layer 2 frame
>relay header. ECN is a flag in the layer 3 IP header. See this for
>more info:
>
>http://www.internetworkexpert.com/rfc/rfc3168.txt
>
>
>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
>24/7 Support: http://forum.internetworkexpert.com
>Live Chat: http://www.internetworkexpert.com/chat/
>
>
> > -----Original Message-----
> > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
>Of
> > John Matus
> > Sent: Tuesday, June 07, 2005 11:59 AM
> > To: ccielab@groupstudy.com
> > Subject: IE LAB7 VOL 2 QOS comparison
> >
> > question regarding the actual solution and what i came up with.
> >
> > the actual solution guide had the following
> >
> > policy-map X
> > class class-default
> > fair-queue
> > random-detect
> > random-detect dscp based
> > random-detect ecn
> >
> > int s0/0
> > service policy X out
> >
> > my solution was:
> >
> > int s0/0
> > fair-queue
> > random-detect
> > random-detect dscp-based
> > frame-relay class frts501
> >
> > map-class frame-relay frts501
> > frame-relay adaptive becn
> >
> > are the 2 policies comparable? the seem to be to me anyhow!
> >
> > _________________________________________________________________
> > On the road to retirement? Check out MSN Life Events for advice on how
>to
> > get there! http://lifeevents.msn.com/category.aspx?cid=Retirement
> >
> >
>_______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Wed Jul 06 2005 - 14:43:41 GMT-3