GroupStudy.com GroupStudy.com - A virtual community of network engineers
 Home  BookStore  StudyNotes  Links  Archives  StudyRooms  HelpWanted  Discounts  Login
Re: IEWB Lab1 4.7 ppp multilink posted 08/11/2004
[Chronological Index] [Thread Index] [Top] [Date Prev][Date Next] [Thread Prev][Thread Next]


I stand corrected.  Thank you Scott Morris.

This is why Group Study is so great.

I'm always learning something new.

----- Original Message ----- 
From: "Scott Morris" <swm@xxxxxxxxxx>
To: "'ccie2be'" <ccie2be@xxxxxxxxxx>; "'Carlos G Mendioroz'"
<tron@xxxxxxxxxxx>; "'Group Study'" <ccielab@xxxxxxxxxxxxxx>
Sent: Wednesday, August 11, 2004 8:03 AM
Subject: RE: IEWB Lab1 4.7 ppp multilink


> I think the confusion comes with some of Cisco's descriptions.  See
>
http://www.cisco.com/en/US/products/sw/iosswrel/ps1835/products_configuratio
> n_guide_chapter09186a00800b75d2.html
>
> The basic description is that LFI is responsible to sequence and
reassemble
> packets properly when balancing over multiple links (implying, of course,
> that more than one link is necessary).
>
> If you look a little further down though, you do see the
> exception/restriction which falls into the specific category you are
talking
> about!  "Interleaving on MLP allows large packets to be multilink
> encapsulated and fragmented into a small enough size to satisfy the delay
> requirements of real-time traffic;  small real-time packets are not
> multilink encapsulated and are sent between fragments of the large
packets."
>
> Interestingly enough, none of the examples shown make any requirement for
> minimum links, which carries the logic that interleaving and fragmentation
> will work just as well with only one particular link coming up, because
when
> 'ppp multilink' is used, you can still have a a bundle of one.
>
> Another good document to view (also containing the table of fragmentation
> recommendations that I used to have and 'lost'!!!) is
>
http://www.cisco.com/en/US/tech/tk652/tk698/technologies_tech_note09186a0080
> 094660.shtml which shows this using a multilink interface although it is
> only applied to a single interface (hence bundle of 1).
>
> HTH,
>
>
> Scott Morris, CCIE4 (R&S/ISP-Dial/Security/Service Provider) #4713, CISSP,
> JNCIP, et al.
> IPExpert CCIE Program Manager
> IPExpert Sr. Technical Instructor
> swm@xxxxxxxxxx/smorris@xxxxxxxxxxxx
> http://www.ipexpert.net
>
>
>
> -----Original Message-----
> From: nobody@xxxxxxxxxxxxxx [mailto:nobody@xxxxxxxxxxxxxx] On Behalf Of
> ccie2be
> Sent: Wednesday, August 11, 2004 7:20 AM
> To: Carlos G Mendioroz; Group Study
> Subject: Re: IEWB Lab1 4.7 ppp multilink
>
> Carlos,
>
> I have to disagree with you on this.
>
> If only one bri is used, how can multilink fragment and interleave over
> multple paths?
>
> For fragmentation to take place, there needs to be at least 2 paths over
> which traffic is sent.
>
> I think where you're getting confused is with the command, ppp multilink.
> True, this command is required before ppp will fragment packets, but by
> itself, it's not sufficient.  There also has to be multiple paths over
which
> to send traffic.
>
> That's where the 2nd command comes into play.  By adding the command,  ppp
> multilink links min 2, you're insuring that there will always be 2 bri's.
> Now, the ppp multilink command can actually fragment packets and split the
> fragments across the 2 bri's.
>
> HTH, Tim
>
> ----- Original Message -----
> From: "Carlos G Mendioroz" <tron@xxxxxxxxxxx>
> To: "Group Study" <ccielab@xxxxxxxxxxxxxx>
> Sent: Wednesday, August 11, 2004 5:37 AM
> Subject: IEWB Lab1 4.7 ppp multilink
>
>
> > Requirement is to have LFI (Link Fragmentation and interleaving)
> > no matter what load on the first channel.
> > Proposed answer is to have multilink with minimum 2.
> >
> >  From where should I get the info that I have to dial with two channels
> > ? LFI is up as soon as I do multilink AFAIK, so 4.7 does not imply load
> > threshold... so for me ppp multilink would be enough.
> >
> > This is where one gets very picky as to doing the thing they want you to
> > do and not what they ask you to do...
> > Or am I missing something ?
> >
> > -- 
> > Carlos G Mendioroz  <tron@xxxxxxxxxxx>  LW7 EQI  Argentina
> >
> > _______________________________________________________________________
> > 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
>
> _______________________________________________________________________
> 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