Results 1 - 10
of
25
Detecting BGP Configuration Faults with Static Analysis
- in Proc. Networked Systems Design and Implementation
, 2005
"... The Internet is composed of many independent autonomous systems (ASes) that exchange reachability information to destinations using the Border Gateway Protocol (BGP). Network operators in each AS configure BGP routers to control the routes that are learned, selected, and announced to other routers. ..."
Abstract
-
Cited by 188 (15 self)
- Add to MetaCart
(Show Context)
The Internet is composed of many independent autonomous systems (ASes) that exchange reachability information to destinations using the Border Gateway Protocol (BGP). Network operators in each AS configure BGP routers to control the routes that are learned, selected, and announced to other routers. Faults in BGP configuration can cause forwarding loops, packet loss, and unintended paths between hosts, each of which constitutes a failure of the Internet routing infrastructure. This paper describes the design and implementation of rcc, the router configuration checker, a tool that finds faults in BGP configurations using static analysis. rcc detects faults by checking constraints that are based on a high-level correctness specification. rcc detects two broad classes of faults: route validity faults, where routers may learn routes that do not correspond to usable paths, and path visibility faults, where routers may fail to learn routes for paths that exist in the network. rcc enables network operators to test and debug configurations before deploying them in an operational network, improving on the status quo where most faults are detected only during operation. rcc has been downloaded by more than sixty-five network operators to date, some of whom have shared their configurations with us. We analyze network-wide configurations from 17 different ASes to detect a wide variety of faults and use these findings to motivate improvements to the Internet routing infrastructure. 1
Design and implementation of a routing control platform
- ACM/USENIX NSDI
, 2005
"... The routers in an Autonomous System (AS) must distribute the information they learn about how to reach external destinations. Unfortunately, today’s internal Border Gateway Protocol (iBGP) architectures have serious problems: a “full mesh ” iBGP configuration does not scale to large networks and “ro ..."
Abstract
-
Cited by 43 (6 self)
- Add to MetaCart
(Show Context)
The routers in an Autonomous System (AS) must distribute the information they learn about how to reach external destinations. Unfortunately, today’s internal Border Gateway Protocol (iBGP) architectures have serious problems: a “full mesh ” iBGP configuration does not scale to large networks and “route reflection ” can introduce problems such as protocol oscillations and persistent loops. Instead, we argue that a Routing Control Platform (RCP) should collect information about external destinations and internal topology and select the BGP routes for each router in an AS. RCP is a logicallycentralized platform, separate from the IP forwarding plane, that performs route selection on behalf of routers and communicates selected routes to the routers using the unmodified iBGP protocol. RCP provides scalability without sacrificing correctness. In this paper, we present the design and implementation of an RCP prototype on commodity hardware. Using traces of BGP and internal routing data from a Tier-1 backbone, we demonstrate that RCP is fast and reliable enough to drive the BGP routing decisions for a large network. We show that RCP assigns routes correctly, even when the functionality is replicated and distributed, and that networks using RCP can expect comparable convergence delays to those using today’s iBGP architectures. 1
Some foundational problems in Interdomain routing
- In HotNets, 2004. (Cited on
, 2004
"... The substantial complexity of interdomain routing in the Internet comes from the need to support flexible policies while scaling to a large number of Autonomous Systems. Despite impressive progress in characterizing the various ills of the Border Gateway Protocol (BGP), many problems remain unsolved ..."
Abstract
-
Cited by 32 (3 self)
- Add to MetaCart
(Show Context)
The substantial complexity of interdomain routing in the Internet comes from the need to support flexible policies while scaling to a large number of Autonomous Systems. Despite impressive progress in characterizing the various ills of the Border Gateway Protocol (BGP), many problems remain unsolved, and the behavior of the routing system is still poorly understood. This paper argues that we must understand interdomain routing in terms of: (1) intrinsic properties and design tradeoffs of policy-based routing, independent of the specific routing protocol and (2) properties that relate to artifacts in today’s protocol. We pose open questions for the research community that, if answered, should help us understand why BGP’s many problems are so difficult to fix. Understanding the fundamental properties of interdomain routing will help us decide how to make progress, be it making backward-compatible modifications to BGP or designing a radically different protocol. 1.
The Case for Separating Routing from Routers
- In ACM SIGCOMM Workshop on Future Directions in Network Architecture
, 2004
"... Over the past decade, the complexity of the Internet's routing infrastructure has increased dramatically. This complexity and the problems it causes stem not just from various new demands made of the routing infrastructure, but also from fundamental limitations in the ability of today's di ..."
Abstract
-
Cited by 29 (5 self)
- Add to MetaCart
(Show Context)
Over the past decade, the complexity of the Internet's routing infrastructure has increased dramatically. This complexity and the problems it causes stem not just from various new demands made of the routing infrastructure, but also from fundamental limitations in the ability of today's distributed infrastructure to scalably cope with new requirements.
Towards a Logic for Wide-Area Internet Routing
- IN ACM SIGCOMM WORKSHOP ON FUTURE DIRECTIONS IN NETWORK ARCHITECTURE
, 2003
"... Interdomain routing is a massive distributed computing task that propagates topological information for global reachability. Today's interdomain routing protocol, BGP4, is exceedingly complex because the wide variety of goals that it must meet---including fast convergence, failure resilience, s ..."
Abstract
-
Cited by 23 (7 self)
- Add to MetaCart
Interdomain routing is a massive distributed computing task that propagates topological information for global reachability. Today's interdomain routing protocol, BGP4, is exceedingly complex because the wide variety of goals that it must meet---including fast convergence, failure resilience, scalability, policy expression, and global reachability---are accomplished by mechanisms that have complicated interactions and unintended side effects. The complexity of wide-area routing configuration and protocol dynamics requires mechanisms for expressing wide-area routing that adhere to a set of logical rules. We propose a set of rules, called the routing logic, which can be used to determine whether a routing protocol satisfies various properties. We demonstrate how this logic can aid in analyzing the behavior of BGP4 under various configurations. We also speculate on how the logic can be used to analyze existing configuration in real-world networks, synthesize network-wide router configuration from a high-level policy language, and assist protocol designers in reasoning about new routing protocols.
Instability free routing: Beyond one protocol instance
- In Proc. ACM CoNEXT
, 2008
"... The copyright law of the United States (title 17, U.S. Code) governs the making of photocopies or other reproductions of copyrighted material. Any copying of this document without permission of its author may be prohibited by law. ..."
Abstract
-
Cited by 18 (12 self)
- Add to MetaCart
(Show Context)
The copyright law of the United States (title 17, U.S. Code) governs the making of photocopies or other reproductions of copyrighted material. Any copying of this document without permission of its author may be prohibited by law.
Using Forgetful Routing to Control BGP Table Size
- In Proc. of CoNext
, 2006
"... All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately. ..."
Abstract
-
Cited by 13 (2 self)
- Add to MetaCart
(Show Context)
All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
How to Construct a Correct and Scalable iBGP Configuration
- in IEEE INFOCOM
, 2005
"... Abstract — The Internet’s current interdomain routing ..."
Abstract
-
Cited by 9 (2 self)
- Add to MetaCart
(Show Context)
Abstract — The Internet’s current interdomain routing
Designing optimal iBGP route-reflection topologies
- In IFIP Networking
, 2008
"... Abstract. The Border Gateway Protocol (BGP) is used today by all Autonomous Systems (AS) in the Internet. Inside each AS, iBGP sessions distribute the external routes among the routers. In large ASs, relying on a full-mesh of iBGP sessions between routers is not scalable, so route-reflection is com ..."
Abstract
-
Cited by 9 (2 self)
- Add to MetaCart
(Show Context)
Abstract. The Border Gateway Protocol (BGP) is used today by all Autonomous Systems (AS) in the Internet. Inside each AS, iBGP sessions distribute the external routes among the routers. In large ASs, relying on a full-mesh of iBGP sessions between routers is not scalable, so route-reflection is commonly used. The scalability of route-reflection compared to an iBGP full-mesh comes at the cost of opacity in the choice of best routes by the routers inside the AS. This opacity induces problems like suboptimal route choices in terms of IGP cost, deflection and forwarding loops. In this work we propose a solution to design iBGP routereflection topologies which lead to the same routing as with an iBGP full-mesh and having a minimal number of iBGP sessions. Moreover we compute a robust topology even if a single node or link failure occurs. We apply our methodology on the network of a tier-1 ISP. Twice as many iBGP sessions are required to ensure robustness to single IGP failure. The number of required iBGP sessions in our robust topology is however not much larger than in the current iBGP topology used in the tier-1 ISP network.
BGP Scaling Techniques Revisited
- ACM Computer Communication Review
, 1999
"... This note adds definitions and clarification to “A Comparison of Scaling Techniques for BGP ” [1], corrects some minor errors and clarifies points which may not have been clear in the original pa-per. It also adds a new analysis of the scaling properties of route- ..."
Abstract
-
Cited by 6 (0 self)
- Add to MetaCart
(Show Context)
This note adds definitions and clarification to “A Comparison of Scaling Techniques for BGP ” [1], corrects some minor errors and clarifies points which may not have been clear in the original pa-per. It also adds a new analysis of the scaling properties of route-