[ppml] Policy Proposal 2006-2: Micro-allocations for Internal Infrastructure

Harold Ritter (hritter) hritter at cisco.com
Wed Apr 12 22:56:25 EDT 2006


Greetings,
 
Yesterday, I attempted to propose a solution to the problem described in policy proposal 2006-2. I realize that the explanation I gave was somewhat confusing and ipv4 biased, so let me try it again but in ipv6 terms this time.
 
The problem leading to the BGP convergence issue:
 
When the /128 route, used to resolve the BGP next-hop of the current BGP best path, is removed from the RIB because the BGP neighbor has gone down, the BGP next-hop is still resolvable via a less specific route, an aggregate route for instance, which causes the staled BGP paths to remain in the routing table for up to 3 minutes (BGP holdtime) even though one or many other paths are available. This obviously results in traffic being blackholed.
 
The solution:
 
Modify BGP implementations so they only consider host routes (/128) as valid routes to resolve the BGP next-hop. This way, if a router goes down and its loopback interface address disappears from the IGP, its BGP neighbors will immediately be able to react by invalidating all BGP prefixes having this address as their BGP next-hop, which leads to a new BGP best path being selected. This should also be configurable to accept a less specific prefix to cope with cases where /128 prefixes are not injected in the IGP under normal conditions.
 
This solution as been applied and tested with ipv4 and is also applicable to ipv6.
 
The point I would like to get across is that in my view there is ways to handle this issue from a routing protocol standpoint. I hope this clears out the confusion I created when I explained at the conference. 
 
Please let me know if you have any questions or comments,
 
BTW: I really enjoyed my first ARIN meeting!!
 
Harold Ritter, CCIE 4168 (R&S / SP)
Cisco Advanced Services
1501 McGill College, 6ième étage
Montréal, Québec H3A 3M8 Canada
Téléphone: 514 847-6856

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20060412/b50eb948/attachment.htm>


More information about the ARIN-PPML mailing list