The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2004-Oct> msg00042



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[mpls] For your review - Issues/errors/clarifications in RFC3 036

  • From: Eric Gray <ewgray2k@netscape.net>
  • Date: Fri, 08 Oct 2004 18:05:28 -0400
  • Cc: mpls@ietf.org
  • User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;rv:1.7.2) Gecko/20040804 Netscape/7.2 (ax)
  • X-AOL-IP: 24.61.197.198

Title:
Neil,

    What you ask should only be a concern if an implementation is either
sloppy in allocating labels or not particularly careful about re-allocating
them.

    There is a "window of opportunity" if an implementation re-issues labels
on a time scale similar to expected packet inter-arrival time for traffic using
that label (either before or after the re-issue). There is also a window of
opportunity if the same label is somehow issued in two separate contexts
and the LSR is not able to correctly recover context when it subsequently
receives  ambiguously labeled packets.

    The first window is avoidable by a robust implementation. The second
window only exists for a broken implementation.

    This is the reason why I feel we are making a big deal out of nothing
very much.

    :-)

--
Eric

neil.2.harrison@bt.com wrote:
Loa,

I was meaning in general, ie labelled pkts.  Just want to check nothing
is being overlooked here.

regards, Neil

  
-----Original Message-----
From: Loa Andersson [mailto:loa@pi.se] 
Sent: 08 October 2004 11:19
To: Harrison,N,Neil,IKR2 R
Cc: ewgray@GraIyMage.com; Nick.Weeds@dataconnection.com; mpls@ietf.org
Subject: Re: [mpls] For your review - 
Issues/errors/clarifications in RFC3 036


Neil,

are you talking about packets exchange over the TCP 
connection between two LDP peers, or (labeled) packets in genral?

/Loa

neil.2.harrison@bt.com wrote:
    
Eric,
<Snipped>
 

      
Not if the downstream router sends a label withdraw as an 
        
appropriate 
    
recovery response to getting a packet with a label that is invalid.
        
Is there not a possibility that packets might end up at the wrong 
destination for other (defect) reasons?  And in such a case 
      
sending a 
    
label withdraw seems the wrong action...comments?

regards, Neil


_______________________________________________
mpls mailing list
mpls@lists.ietf.org https://www1.ietf.org/mailman/listinfo/mpls


      
-- 
Loa Andersson

Principal Networking Architect
Acreo AB                           phone:  +46 8 632 77 14
Isafjordsgatan 22                  mobile: +46 739 81 21 64
Kista, Sweden                      email:  loa.andersson@acreo.se
                                            loa@pi.se

    

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls




  
_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls