RE: ICMP Query!!!

From: Tyson Scott <tscott_at_ipexpert.com>
Date: Sun, 14 Nov 2010 15:05:10 -0500

Dale,

I agree. My PAK argument doesn't hold water after I think about it further
as well ;). I have thought a lot about this the last day and I think there
is room for debate each way. But if you read Yusuf Bhaji's Network Security
Technologies book his simple statement on control plane is that it consists
of protocols that help to "glue the network together". As a network can
fundamentally function without the use of ICMP anywhere, meaning I could
block all ICMP traffic and everything will still work, I consider it to be
out of scope. That although ICMP traffic may come to the control plane for
one reason or another, like ICMP redirect to give better route information
or ICMP unreachable in the event of an unknown network or TTL expiration for
traceroute, ICMP is not required to run the network. Whereas other things
like IGMP, as Paul pointed out below is required for multicast to work.

Fundamentally the Control Plane is traffic generated or accepted by the
router that are necessary for the network to perform functions, i.e. routing
protocols, multicast, IOS firewall (transit control plane). ICMP doesn't
fall under any of those categories. Read Yusuf's book, it is probably one
of the best clarifications on this topic out there. I also have the slides
from his internal presentation on the topic.

Now in what I have stated I will clarify that ICMP should be considered in
CoPP Policy because it is a protocol that can affect the performance and
security of the router. Just as undesirable traffic is also considered
something you should protect the control plane from or undesirable IP
options. So ICMP falls under the category of a protocol that Control Plane
Protection is used to prevent from affecting the router not a protocol that
is necessary for the operation of the control plane.

Regards,
 
Tyson Scott - CCIE #13513 R&S, Security, and SP
Managing Partner / Sr. Instructor - IPexpert, Inc.
Mailto: tscott_at_ipexpert.com

-----Original Message-----
From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of Dale
Shaw
Sent: Sunday, November 14, 2010 2:09 PM
To: Marko Milivojevic
Cc: ron wilkerson; Tyson Scott; negron.paul_at_gmail.com; tron_at_huapi.ba.ar;
eliteccie_at_gmail.com; ccielab_at_groupstudy.com
Subject: Re: ICMP Query!!!

I concur with Sadiq's assessment.

In terms of the way ICMP traffic is classified (control vs data) from
the perspective of a router, it depends entirely on whether it is
destined for that router or simply being passed through it.

I can't think of a ICMP packet type that triggers a response by an
intermediate/transit system, aside from oddball things like IDS/IPS.
Can anyone else? By design, It's all about signalling to/from
endpoints.

And the PAK priority thing doesn't hold water either as last time I
checked (12.4(15)T), IKE traffic - very much a control plane protocol
when a SA is terminated on the local router - wasn't given any special
treatment (no PAK priority, no DSCP marking).

Happy to stand corrected.

Cheers
Dale
(apologies for top-post and lack of quoting and trimming)

On Monday, November 15, 2010, Marko Milivojevic <markom_at_ipexpert.com> wrote:
> Of course, based on the assumption that Cisco documentation is never
> wrong, a very wise choice.
>
> --
> Marko Milivojevic - CCIE #18427
> Senior Technical Instructor - IPexpert
>
> FREE CCIE training: http://bit.ly/vLecture
>
> Mailto: markom_at_ipexpert.com
> Telephone: +1.810.326.1444
> Web: http://www.ipexpert.com/
>
> On Sun, Nov 14, 2010 at 02:13, ron wilkerson <ron.wilkerson_at_gmail.com>
wrote:
>> that's fine..
>> i'll stick to what the cisco doc states and others can stick to what they
>> believe is correct...agree to disagree.
>>
>> another item to be added to the list where everyone won't agree due to
>> various reasons.
>>
>>
>>
>> On Sat, Nov 13, 2010 at 7:10 PM, Tyson Scott
>> <tyson.scott_at_advtechracks.com>wrote:
>>
>>> As I said ICMP unreachables will go to the CEF exception sub interface
>>>
>>> Regards,
>>>
>>> Tyson Scott
>>> CCIE # 13513 (R&S, Security, SP)
>>>
>>> Managing Partner/Technical Instructor - IPexpert Inc.
>>> tscott_at_ipexpert.com
>>>
>>>
>>> ----- Reply message -----
>>> From: "ron wilkerson" <ron.wilkerson_at_gmail.com>
>>> Date: Sat, Nov 13, 2010 3:30 pm
>>>
>>> Subject: ICMP Query!!!
>>> To: "Tyson Scott" <tyson.scott_at_advtechracks.com>
>>> Cc: "negron.paul_at_gmail.com" <negron.paul_at_gmail.com>, "tron_at_huapi.ba.ar"
<
>>> tron_at_huapi.ba.ar>, "eliteccie_at_gmail.com" <eliteccie_at_gmail.com>, "
>>> ccielab_at_groupstudy.com" <ccielab_at_groupstudy.com>
>>>
>>>
>>>
http://www.cisco.com/en/US/prod/collateral/iosswrel/ps6537/ps6586/ps6642/pro
d_white_paper0900aecd805ffde8.html
>>>
>>> read the 4th paragraph.
>>>
>>> On Sat, Nov 13, 2010 at 3:24 PM, Tyson Scott
<tyson.scott_at_advtechracks.com
>>> > wrote:
>>>
>>>> ICMP is not control plane traffic. B ICMP unreachables go to the CEF
>>>> exception for example. B Consider the control plane as protocols that
>>>> glue the network together. B ICMP traffic to the router go to the host
>>>> control plane because of being directed to the device thus it must
>>>> handle it. B ICMP is data traffic that may be used for management
>>>> purposes
>>>>
>>>> Regards,
>>>>
>>>> Tyson Scott
>>>> CCIE # 13513 (R&amp;S, Security, SP)
>>>> Managing Partner/Technical Instructor - IPexpert Inc.
>>>> tscott_at_ipexpert.com
>>>>
>>>>
>>>> ----- Reply message -----
>>>> From: "Paul Negron" <negron.paul_at_gmail.com>
>>>> Date: Sat, Nov 13, 2010 2:10 pm
>>>> Subject: ICMP Query!!!
>>>> To: "ron.wilkerson_at_gmail.com" <ron.wilkerson_at_gmail.com>, "Carlos G
>>>> Mendioroz" <tron_at_huapi.ba.ar>
>>>> Cc: "CCIE KID" <eliteccie_at_gmail.com>, "Cisco certification"
>>>> <ccielab_at_groupstudy.com>
>>>>
>>>>
>>>> Very Interesting Response.
>>>>
>>>> I guess I primarily viewed ICMP as testing the Control Plane/ Data
Plane
>>>> with the Majority of ICMP Query types:
>>>>
>>>> * 0 = Echo Reply (3ping response2)
>>>> * 8 = Echo Request (3ping query2)
>>>> * 9 = Router Advertisement (RFC 1256)
>>>> * 10 = Router Solicitation (RFC 1256)
>>>> * 13 = Time Stamp Request
>>>> * 14 = Time Stamp Reply
>>>> * 17 = Address Mask Request
>>>> * 18 = Address Mask Reply
>>>>
>>>> I know my definition is a little Narrow but it does help differentiate
>>>> ICMP
>>>> from protocols like RSVP, PIM, EIGRP that strictly represent Control
Plane
>>>> from a Routing Switching perspective.
>>>>
>>>> As far as the view that because ICMP uses the CPU being a CLEAR
>>>> definition,
>>>> this I would disagree with. What would Process Switching be then?
Control
>>>> Plane or Data Plane activity?
>>>>
>>>> Carlos and Ron do make a good point to expand my Narrow definition
though.
>>>> :-)
>>>>
>>>> Paul
>>>>
>>>>
>>>> --
>>>> Paul Negron
>>>> CCIE# 14856 CCSI# 22752
>>>> Senior Technical Instructor
>>>> www.micronicstraining.com
>>>>
>>>>
>>>>
>>>> > From: <ron.wilkerson_at_gmail.com>
>>>> > Reply-To: <ron.wilkerson_at_gmail.com>
>>>> > Date: Fri, 12 Nov 2010 23:58:17 +0000
>>>> > To: Paul Negron <negron.paul_at_gmail.com>, Carlos G Mendioroz <
>>>> tron_at_huapi.ba.ar>
>>>>
>>>> > Cc: CCIE KID <eliteccie_at_gmail.com>, Cisco certification
>>>> > <ccielab_at_groupstudy.com>
>>>> > Subject: Re: ICMP Query!!!
>>>> >
>>>> > Agree with carlos...
>>>> > I've always thought of control plane as anything that the cpu has to
>>>> look at.
>>>> > Some icmp packets require the cpu, so I'd classify those icmp as
control
>>>> plane
>>>> > packets.
>>>> >
>>>> >
>>>> > Sent from my Verizon Wireless BlackBerry
>>>> >
>>>> > -----Original Message-----
>>>> > From: Paul Negron <negron.paul_at_gmail.com>
>>>> > Sender: nobody_at_groupstudy.com
>>>> > Date: Fri, 12 Nov 2010 16:39:10
>>>> > To: Carlos G Mendioroz<>>> >>> I apologize, B I meant to state:
>>>> >>>
>>>> >>>> IGMP packets are used to create state on the Router that receives
>>>> them.
>>>> >>>> Since it is used to create state, it is a part of the Control
Plane
>>>> >>>> process.
>>>> >>>> It joins so that trees can be built, Although it is PIM that
builds
>>>> them.
>>>> >>>>
>>>> >>>> ICMP is generating traffic and is not associated with building
>>>> ANYTHING. It
>>>> >>>> is considered Data Plane traffic. It uses paths that have already
>>>> been
>>>> >>>> setup
>>>> >>>> by a Control Plane Protocol, like OSPF or EIGRP or PIM for that
>>>> matter.
>>>> >>>
>>>> >>> I accidentally stated ICMP twice.
>>>> >>>
>>>> >>> Paul
>>>> >>
>>>> >> --
>>>> >> Carlos G Mendioroz B <tron_at_huapi.ba.ar> B LW7 EQI B Argentina
>>>> >
>>>> >
>>>> > Blogs and organic groups at http://www.ccie.net
>>>> >
>>>> >
Received on Sun Nov 14 2010 - 15:05:10 ART

This archive was generated by hypermail 2.2.0 : Sun Dec 05 2010 - 22:14:56 ART