RFC 7705 PROPOSED STANDARD BGP · ROUTING · OPERATIONS
BGP AS Migration: Autonomous System Migration Mechanisms and Their Effects on the BGP AS_PATH Attribute
The router tricks ISPs use to switch to a new AS number without making every customer reconfigure at once.
RFC EDITOR
George, W. and S. Amante, "Autonomous System Migration Mechanisms and Their Effects on the BGP AS_PATH Attribute", RFC 7705, DOI 10.17487/RFC7705, November 2015, <https://www.rfc-editor.org/info/rfc7705>.
BIBTEX
@misc{rfc7705,
series = {Request for Comments},
number = 7705,
howpublished = {RFC 7705},
publisher = {RFC Editor},
doi = {10.17487/RFC7705},
url = {https://www.rfc-editor.org/info/rfc7705},
author = {W. George and S. Amante},
title = {{Autonomous System Migration Mechanisms and Their Effects on the BGP AS_PATH Attribute}},
pagetotal = 16,
year = 2015,
month = nov,
} LINEAGE · FROM THE RFC INDEXOpen full graph →
RFC text
Contents
Internet Engineering Task Force (IETF) W. George
Request for Comments: 7705 Time Warner Cable
Updates: 4271 S. Amante
Category: Standards Track Apple, Inc.
ISSN: 2070-1721 November 2015
Autonomous System Migration Mechanisms
and Their Effects on the BGP AS_PATH Attribute
Abstract
This document discusses some existing commonly used BGP mechanisms
for Autonomous System Number (ASN) migration that are not formally
part of the BGP4 protocol specification. It is necessary to document
these de facto standards to ensure that they are properly supported
in future BGP protocol work.
Status of This Memo
This is an Internet Standards Track document.
This document is a product of the Internet Engineering Task Force
(IETF). It represents the consensus of the IETF community. It has
received public review and has been approved for publication by the
Internet Engineering Steering Group (IESG). Further information on
Internet Standards is available in Section 2 of RFC 5741.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
http://www.rfc-editor.org/info/rfc7705.
Copyright Notice
Copyright (c) 2015 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3
1.2. Documentation Note . . . . . . . . . . . . . . . . . . . 3
2. ASN Migration Scenario Overview . . . . . . . . . . . . . . . 3
3. External BGP Autonomous System Migration Mechanisms . . . . . 5
3.1. Modify Inbound BGP AS_PATH Attribute . . . . . . . . . . 5
3.2. Modify Outbound BGP AS_PATH Attribute . . . . . . . . . . 7
3.3. Implementation . . . . . . . . . . . . . . . . . . . . . 8
4. Internal BGP Autonomous System Migration Mechanisms . . . . . 9
4.1. Internal BGP AS Migration . . . . . . . . . . . . . . . . 10
4.2. Implementation . . . . . . . . . . . . . . . . . . . . . 12
5. Additional Operational Considerations . . . . . . . . . . . . 13
6. Security Considerations . . . . . . . . . . . . . . . . . . . 14
7. References . . . . . . . . . . . . . . . . . . . . . . . . . 14
7.1. Normative References . . . . . . . . . . . . . . . . . . 14
7.2. Informative References . . . . . . . . . . . . . . . . . 14
Appendix A. Implementation Report . . . . . . . . . . . . . . . 16
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . 16
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 16
1. Introduction
This document discusses some existing commonly used BGP mechanisms
for Autonomous System Number (ASN) migration that are not formally
part of the BGP4 [RFC4271] protocol specification. These mechanisms
are local to a given BGP speaker and do not require negotiation with
or cooperation of BGP neighbors. The deployment of these mechanisms
do not need to interwork with one another to accomplish the desired
results, so slight variations between existing vendor implementations
exist and will not necessarily be harmonized due to this document.
However, it is necessary to document these de facto standards to
ensure that new implementations can be successful, and any future
protocol enhancements to BGP that propose to read, copy, manipulate,
or compare the AS_PATH attribute can do so without inhibiting the use
of these very widely used ASN migration mechanisms.
The migration mechanisms discussed here are useful to ISPs and
organizations of all sizes, but it is important to understand the
business need for these mechanisms and illustrate why they are so
critical for ISPs' operations. During a merger, acquisition, or
divestiture involving two organizations it is necessary to seamlessly
migrate both internal and external BGP speakers from one ASN to a
second ASN. The overall goal in doing so is to simplify operations
through consistent configurations across all BGP speakers in the
combined network. In addition, given that the BGP Path Selection
algorithm selects routes with the shortest AS_PATH attribute, it is
critical that the ISP does not increase AS_PATH length during or
after ASN migration, because an increased AS_PATH length would likely
result in sudden, undesirable changes in traffic-patterns in the
network.
By default, the BGP protocol requires an operator to configure a
router to use a single remote ASN for the BGP neighbor, and the ASN
must match on both ends of the peering in order to successfully
negotiate and establish a BGP session. Prior to the existence of
these migration mechanisms, it would have required an ISP to
coordinate an ASN change with, in some cases, tens of thousands of
customers. In particular, as each router is migrated to the new ASN,
to avoid an outage due to ASN mismatch, the ISP would have to force
all customers on that router to change their router configurations to
use the new ASN immediately after the ASN change. Thus, it was
critical to define a more asymmetric migration method so that the ISP
could seamlessly change the ASN within its network(s) but allow the
customers to gradually migrate to the ISP's new ASN when convenient.
These customer migrations could occur either by coordinating
individual session reconfigurations or, to allow for truly asymmetric
migration, by accepting sessions using either the old or new ASN.
1.1. Requirements Language
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [RFC2119].
1.2. Documentation Note
This document uses Autonomous System Numbers (ASNs) from the range
reserved for documentation as described in RFC 5398 [RFC5398]. In
the examples used here, they are intended to represent Globally
Unique ASNs, not private use ASNs as documented in Section 5 of RFC
6996 [RFC6996].
2. ASN Migration Scenario Overview
The use case being discussed here is an ISP merging two or more ASNs,
where eventually one ASN subsumes the other(s). In this use case, we
will assume the most common case where there are two ISPs, A and B,
that prior to the ASN migration use AS 64500 and 64510, respectively.
AS 64500 will be the permanently retained ASN used across the
consolidated set of both ISPs network equipment, and AS 64510 will be
retired. Thus, at the conclusion of the ASN migration, there will be
a single ISP A' with all internal BGP speakers configured to use AS
64500. To all external BGP speakers, the AS_PATH length will not be
increased.
In this same scenario, AS 64496 and AS 64499 represent two separate
customer networks: C and D, respectively. Originally, customer C (AS
64496) is attached to ISP B, which will undergo ASN migration from AS
64510 to AS 64500. Furthermore, customer D (AS 64499) is attached to
ISP A, which does not undergo ASN migration since the ASN for ISP A
will remain constant, (AS 64500). Although this example refers to AS
64496 and 64499 as customer networks, either or both may be
settlement-free or other types of peers. In this use case, they are
referred to as "customers" merely for convenience.
------ ------
/ ISP A \ / ISP B \
| AS 64500 | | AS 64510 |
\ / \ /
------- -------
| |
| |
------------ -------------
| Cust D | | Cust C |
| AS 64499 | | AS 64496 |
------------ -------------
Figure 1: Before Migration
---------------
/ \
| ISP A' |
| AS 64500 |
\ /
---------------
/ \
/ \
| |
------------ -------------
| Cust D | | Cust C |
| AS 64499 | | AS 64496 |
------------ -------------
Figure 2: After Migration
The general order of operations, typically carried out in a single
maintenance window by the network undergoing ASN migration (ISP B),
is as follows. First, ISP B will change the global BGP ASN used by a
Provider Edge (PE) router, from ASN 64510 to 64500. At this point,
the router will no longer be able to establish External BGP (eBGP)
sessions toward the existing Customer Edge (CE) devices that are
attached to it and still using AS 64510. Second, since ISP B needs
to do this without coordinating the simultaneous change of its ASN
with all of its eBGP peers, ISP B will configure two separate, but
related ASN migration mechanisms discussed in this document on all
eBGP sessions toward all CE devices. These mechanisms enable the
router to establish BGP neighbors using the legacy ASN, modify the
AS_PATH attribute received from a CE device when advertising it
further, and modify AS_PATH when transmitted toward CE devices to
achieve the desired effect of not increasing the length of the
AS_PATH.
At the conclusion of the ASN migration, the CE devices at the edge of
the network are not aware of the fact that their upstream router is
now in a new ASN and do not observe any change in the length of the
AS_PATH attribute. However, after the changes discussed in this
document are put in place by ISP A', there is a change to the
contents of the AS_PATH attribute to ensure the AS_PATH is not
artificially lengthened while these AS migration parameters are used.
In this use case, neither ISP is using BGP Confederations [RFC5065]
internally.
3. External BGP Autonomous System Migration Mechanisms
The following section addresses optional capabilities that are
specific to modifying the AS_PATH attribute at the Autonomous System
Border Routers (ASBRs) of an organization (typically a single Service
Provider). This ensures that external BGP customers/peers are not
forced to make any configuration changes on their CE routers before
or during the exact time the Service Provider wishes to migrate to a
new, permanently retained ASN. Furthermore, these mechanisms
eliminate the artificial lengthening of the AS_PATH both transmitted
from and received by the Service Provider that is undergoing AS
Migration, which would have negative implications on path selection
by external networks.
3.1. Modify Inbound BGP AS_PATH Attribute
The first instrument used in the process described above is called
"Local AS". This allows the router to supersede the globally
configured ASN in the "My Autonomous System" field of the BGP OPEN
[RFC4271] with a locally defined AS value for a specific BGP neighbor
or group of neighbors. This mechanism allows the PE router that was
formerly in ISP B to establish an eBGP session toward the existing CE
devices using the legacy AS -- AS 64510. Ultimately, the CE devices
(i.e., customer C) are completely unaware that ISP B has reconfigured
its router to participate as a member of a new AS. Within the
context of the former ISP B PE router, the second effect this
specific mechanism has on AS_PATH is that, by default, it prepends
all received BGP UPDATEs with the legacy AS of ISP B -- AS 64510 --
while advertising it (Adj-RIB-Out) to other BGP speakers (A').
Within the Loc-RIB on ISP B prior to the migration, the AS_PATH of
route announcements received from customer C would appear as 64496,
whereas the same RIB on ISP A' (ISP B routers post-migration) would
contain AS_PATH 64510 64496.
The second instrument, referred to as "No Prepend Inbound", is
enabled on PE routers migrating from ISP B. The "No Prepend Inbound"
capability causes ISP B's routers to not prepend the legacy AS (AS
64510), when advertising UPDATES received from customer C. This
restores the AS_PATH within ISP A' for route announcements received
from customer C so that it is just one ASN in length: 64496.
In the direction of CE -> PE (inbound):
1. "Local AS": Allows the local BGP router to generate a BGP OPEN to
an eBGP neighbor with the old, legacy ASN value in the "My
Autonomous System" field. When this capability is activated, it
also causes the local router to prepend the <old_ASN> value to
the AS_PATH when installing or advertising routes received from a
CE to Internal BGP (iBGP) neighbors inside the Autonomous System.
2. "No Prepend Inbound (of Local AS)": The local BGP router does not
prepend the <old_ASN> value to the AS_PATH when installing or
advertising routes received from the CE to iBGP neighbors inside
the Autonomous System
PE-B is a PE that was originally in ISP B, and has a customer eBGP
session to CE-B. PE-B has had its global configuration ASN changed
from AS 64510 to AS 64500 to make it part of the permanently retained
ASN. This now makes PE-B a member of ISP A'. PE-A is a PE that was
originally in ISP A, and has a customer peer CE-A. Although its
global configuration ASN remains AS 64500, throughout this exercise
we also consider PE-A a member of ISP A'.
ISP A' ISP A'
CE-A <--- PE-A <------------------- PE-B <--- CE-B
64499 New_ASN: 64500 Old_ASN: 64510 64496
New_ASN: 64500
Note: Direction of BGP UPDATE as per the arrows.
Figure 3: Local AS and No Prepend BGP UPDATE Diagram
As a result, using both the "Local AS" and "No Prepend Inbound"
capabilities on PE-B, CE-A will see an AS_PATH of 64500 64496. CE-A
will not receive a BGP UPDATE containing AS 64510 in the AS_PATH.
(If only the "Local AS" mechanism was configured without "No Prepend
Inbound" on PE-B, then CE-A would have seen an AS_PATH of 64500 64510
64496, which results in an unacceptable lengthening of the AS_PATH).
Note: If there are still routers in the old ASN (64510), it is
possible for them to accept these manipulated routes (i.e., those
with 64510 removed from the AS_PATH by this command) as if they have
not already passed through their ASN, potentially causing a loop,
since BGP's normal loop-prevention behavior of rejecting routes that
include its ASN in the path will not catch these. Careful filtering
between routers remaining in the old ASN and routers migrated to the
new ASN is necessary to minimize the risk of routing loops.
3.2. Modify Outbound BGP AS_PATH Attribute
The two aforementioned mechanisms, "Local AS" and "No Prepend
Inbound", only modify the AS_PATH attribute received by the ISP's PEs
in the course of processing BGP UPDATEs from CE devices when CE
devices still have an eBGP session established with the ISPs legacy
AS (AS64510).
In some existing implementations, "Local AS" and "No Prepend Inbound"
do not concurrently modify the AS_PATH attribute for BGP UPDATEs that
are transmitted by the ISP's PEs to CE devices. In these
implementations, with "Local AS" and "No Prepend Inbound" used on
PE-B, there is an automatic lengthening of the AS_PATH in outbound
BGP UPDATEs from ISP A' toward directly attached eBGP speakers
(customer C in AS 64496). The externally observed result is that
customer C in AS 64496 will receive the following AS_PATH: 64510
64500 64499. Therefore, if ISP A' takes no further action, there
will be an unacceptable increase in the AS_PATH length within the
customer's networks directly attached to ISP A'.
A tertiary mechanism, referred to as "Replace Old AS", is used to
resolve this problem. This capability allows ISP A' to prevent
routers from appending the globally configured ASN in outbound BGP
UPDATEs toward directly attached eBGP neighbors that are using the
"Local AS" mechanism. Instead, only the old (or previously used) AS
will be prepended in the outbound BGP UPDATE toward the customer's
network, restoring the AS_PATH length to what it what was before AS
Migration occurred.
To reuse the above diagram, but in the opposite direction, we have:
ISP A' ISP A'
CE-A ---> PE-A -------------------> PE-B ---> CE-B
64499 New_ASN: 64500 Old_ASN: 64510 64496
New_ASN: 64500
Note: Direction of BGP UPDATE as per the arrows.
Figure 4: Replace AS BGP UPDATE Diagram
By default, without the use of "Replace Old AS", CE-B would see an
AS_PATH of 64510 64500 64499. After ISP A' changes PE-B to use
"Replace Old AS", CE-B would receive an AS_PATH of 64510 64499, which
is the same AS_PATH length pre-AS migration.
3.3. Implementation
The mechanisms introduced in this section MUST be configurable on a
per-neighbor or per-neighbor-group basis to allow for maximum
flexibility. (Here, "neighbor group" refers to a group of similar
BGP neighbor statements that reuse some common configuration to
simplify provisioning.) When the "Local AS" capability is used, a
local ASN will be provided in the configuration that is different
from the globally configured ASN of the BGP router. To implement
this mechanism, a BGP speaker SHOULD send BGP OPEN [RFC4271] (see
Section 4.2) messages to the configured eBGP peer(s) using the local
ASN configured for this session as the value sent in "My Autonomous
System". The BGP router SHOULD NOT use the ASN configured globally
within the BGP process as the value sent in "My Autonomous System" in
the OPEN message. This prevents causing the eBGP neighbor to
unnecessarily generate a BGP OPEN Error message "Bad Peer AS". This
method is typically used to re-establish eBGP sessions with peers
expecting the legacy ASN after a router has been moved to a new ASN.
Implementations MAY support a more flexible model where the eBGP
speaker attempts to open the BGP session using either the ASN
configured as "Local AS" or the globally configured AS as discussed
in BGP Alias (Section 4.2). If the session is successfully
established to the globally configured ASN, then the modifications to
AS_PATH described in this document SHOULD NOT be performed, as they
are unnecessary. The benefit to this more flexible model is that it
allows the remote neighbor to reconfigure to the new ASN without
direct coordination between the ISP and the customer.
Note that this procedure will vary slightly if the locally or
globally configured ASN is a 4-octet ASN. See Section 3 of
[RFC6793].
When the BGP router receives UPDATEs from its eBGP neighbor
configured with the "Local AS" mechanism, it processes the UPDATE as
described in RFC4271 Section 5.1.2 [RFC4271]. However, the presence
of a second ASN due to "Local AS" adds the following behavior to
processing UPDATEs received from an eBGP neighbor configured with
this mechanism:
1. Internal: The router SHOULD append the configured "Local AS" ASN
in the AS_PATH attribute before installing the route or
advertising the UPDATE to an iBGP neighbor. The decision of when
to append the ASN is an implementation detail outside the scope
of this document. Some considerations factoring into this
decision include consistency in the AS_PATH throughout the AS,
and implementation of the loop detection mechanism.
2. External: The BGP router SHOULD first append the globally
configured ASN to the AS_PATH immediately followed by the "Local
AS" value before advertising the UPDATE to an eBGP neighbor.
Two options exist to manipulate the behavior of the basic "Local AS"
mechanism. They modify the behavior as described below:
1. "No Prepend Inbound": When the BGP router receives inbound BGP
UPDATEs from its eBGP neighbor configured with this option, it
MUST NOT append the "Local AS" ASN value in the AS_PATH attribute
when installing the route or advertising that UPDATE to iBGP
neighbors, but it MUST still append the globally configured ASN
as normal when advertising the UPDATE to other local eBGP
neighbors (i.e., those natively peering with the globally
configured ASN).
2. "Replace Old AS" (outbound): When the BGP router generates
outbound BGP UPDATEs toward an eBGP neighbor configured with this
option, the BGP speaker MUST NOT append the globally configured
ASN from the AS_PATH attribute. The BGP router MUST append only
the configured "Local AS" ASN value to the AS_PATH attribute
before sending the BGP UPDATEs outbound to the eBGP neighbor.
4. Internal BGP Autonomous System Migration Mechanisms
The following section describes mechanisms that assist with a gradual
and least service-impacting migration of Internal BGP sessions from a
legacy ASN to the permanently retained ASN. The following mechanism
is very valuable to networks undergoing AS migration, but its use
does not cause changes to the AS_PATH attribute.
4.1. Internal BGP AS Migration
In this case, all of the routers to be consolidated into a single,
permanently retained ASN are under the administrative control of a
single entity. Unfortunately, the traditional method of migrating
all Internal BGP speakers, particularly within larger networks, is
both time-consuming and widely service impacting.
The traditional method to migrate Internal BGP sessions was strictly
limited to reconfiguration of the global configuration ASN and,
concurrently, changing all iBGP neighbors' remote ASN from the legacy
ASN to the new, permanently retained ASN on each router within the
legacy AS. These changes can be challenging to swiftly execute in
networks with more than a few dozen internal BGP routers. There are
also the concomitant service interruptions as these changes are made
to routers within the network, resulting in a reset of iBGP sessions
and subsequent route reconvergence to reestablish optimal routing
paths. Operators often cannot make such sweeping changes given the
associated risks of a highly visible service interruption; rather,
they require a more gradual method to migrate Internal BGP sessions,
from one ASN to a second, permanently retained ASN, that is not
visibly service impacting to its customers.
With the "Internal BGP AS Migration" mechanism described herein, it
allows an Internal BGP speaker to form a single iBGP session using
either the old, legacy ASN or the new, permanently retained ASN. The
benefits of using this mechanism are several fold. First, it allows
for a more gradual and less service-impacting migration away from the
legacy ASN to the permanently retained ASN. Second, it (temporarily)
permits the coexistence of the legacy and permanently retained ASN
within a single network, allowing for uniform BGP path selection
among all routers within the consolidated network.
The iBGP router with the "Internal BGP AS Migration" capability
enabled allows the receipt of a BGP OPEN message with either the
legacy ASN value or the new, globally configured ASN value in the "My
Autonomous System" field of the BGP OPEN message from iBGP neighbors.
It is important to recognize that enablement of the "Internal BGP AS
Migration" mechanism preserves the semantics of a regular iBGP
session (i.e., using identical ASNs). Thus, the BGP attributes
transmitted by and the acceptable methods of operation on BGP
attributes received from iBGP sessions configured with "Internal BGP
AS Migration" capability are no different than those exchanged across
an iBGP session without "Internal BGP AS Migration" configured, as
defined by [RFC4271] and [RFC4456].
Typically, in medium to large networks, BGP Route Reflectors (RRs)
[RFC4456] are used to aid in reduction of configuration of iBGP
sessions and scalability with respect to overall TCP (and BGP)
session maintenance between adjacent iBGP routers. Furthermore, BGP
Route Reflectors are typically deployed in pairs within a single
Route Reflection cluster to ensure high reliability of the BGP
Control Plane. As such, the following example will use Route
Reflectors to aid in understanding the use of the "Internal BGP AS
Migration" mechanism. Note that Route Reflectors are not a
prerequisite to enable "Internal BGP AS Migration" and this mechanism
can be enabled independent of the use of Route Reflectors.
The general order of operations is as follows:
1. Within the legacy network, (the routers comprising the set of
devices that still have a globally configured legacy ASN), one
member of a redundant pair of RRs has its global configuration
ASN changed to the permanently retained ASN. Concurrently, the
"Internal BGP AS Migration" capability is enabled on all iBGP
sessions on that device. This will comprise Non-Client iBGP
sessions to other RRs as well as Client iBGP sessions, typically
to PE devices, both still utilizing the legacy ASN. Note that
during this step there will be a reset and reconvergence event on
all iBGP sessions on the RRs whose configuration was modified;
however, this should not be service impacting due to the use of
redundant RRs in each RR Cluster.
2. The above step is repeated for the other side of the redundant
pair of RRs. The one alteration to the above procedure is that
the "Internal BGP AS Migration" mechanism is now removed from the
Non-Client iBGP sessions toward the other (previously
reconfigured) RRs, since it is no longer needed. The "Internal
BGP AS Migration" mechanism is still required on all RRs for all
RR Client iBGP sessions. Also during this step, there will be a
reset and reconvergence event on all iBGP sessions whose
configuration was modified, but this should not be service
impacting. At the conclusion of this step, all RRs will have
their globally configured ASN set to the permanently retained ASN
and "Internal BGP AS Migration" enabled and in use toward RR
Clients.
3. At this point, the network administrators would then be able to
establish iBGP sessions between all Route Reflectors in both the
legacy and permanently retained networks. This would allow the
network to appear to function, both internally and externally, as
a single, consolidated network using the permanently retained
network.
4. To complete the AS migration, each RR Client (PE) in the legacy
network still utilizing the legacy ASN is now modified.
Specifically, each legacy PE would have its globally configured
ASN changed to use the permanently retained ASN. The ASN
configured within the PE for the iBGP sessions toward each RR
would be changed to use the permanently retained ASN. It is
unnecessary to enable the "Internal BGP AS Migration" mechanism
on these migrated iBGP sessions. During the same maintenance
window, External BGP sessions would be modified to include the
above "Local AS", "No Prepend", and "Replace Old AS" mechanisms
described in Section 3 above, since all of the changes are
service interrupting to the eBGP sessions of the PE. At this
point, all PEs will have been migrated to the permanently
retained ASN.
5. The final step is to excise the "Internal BGP AS Migration"
configuration from the Router Reflectors in an orderly fashion.
After this is complete, all routers in the network will be using
the new, permanently retained ASN for all iBGP sessions with no
vestiges of the legacy ASN on any iBGP sessions.
The benefit of using the aforementioned "Internal BGP AS Migration"
capability is that it is a more gradual and less externally service
impacting change to accomplish an AS migration. Previously, without
"Internal BGP AS Migration", such an AS migration change would carry
a high risk and need to be successfully accomplished in a very short
time frame (e.g., at most several hours). In addition, it would
likely cause substantial routing churn and rapid fluctuations in
traffic carried -- potentially causing periods of congestion and
resultant packet loss -- during the period when the configuration
changes are underway to complete the AS Migration. On the other
hand, with "Internal BGP AS Migration", the migration from the legacy
ASN to the permanently retained ASN can occur over a period of days
or weeks with reduced customer disruption. (The only observable
service disruption should be when each PE undergoes the changes
discussed in step 4 above.)
4.2. Implementation
The mechanism introduced in this section MUST be configurable on a
per-neighbor or per-neighbor-group basis to allow for maximum
flexibility. When configured with this mechanism, a BGP speaker MUST
accept BGP OPEN and establish an iBGP session from configured iBGP
peers if the ASN value in "My Autonomous System" is either the
globally configured ASN or a locally configured ASN provided when
this capability is utilized. Additionally, a BGP router configured
with this mechanism MUST send its own BGP OPEN [RFC4271] (see
Section 4.2) using either the globally configured or the locally
configured ASN in "My Autonomous System" as follows. To avoid
potential deadlocks when two BGP speakers are attempting to establish
a BGP peering session and are both configured with this mechanism,
the speaker SHOULD send BGP OPEN using the globally configured ASN
first, and only send a BGP OPEN using the locally configured ASN as a
fallback if the remote neighbor responds with the BGP error "Bad Peer
AS". In each case, the BGP speaker MUST treat UPDATEs sent and
received to this peer as if this was a natively configured iBGP
session, as defined by [RFC4271] and [RFC4456].
Note that this procedure will vary slightly if the locally or
globally configured ASN is a 4-octet ASN. See Section 3 of
[RFC6793].
5. Additional Operational Considerations
This document describes several mechanisms to support ISPs and other
organizations that need to perform ASN migrations. Other variations
of these mechanisms may exist, for example, in legacy router software
that has not been upgraded or reached End of Life, but continues to
operate in the network. Such variations are beyond the scope of this
document.
Companies routinely go through periods of mergers, acquisitions, and
divestitures, which in the case of the former cause them to
accumulate several legacy ASNs over time. ISPs often do not have
control over the configuration of customers' devices (i.e., the ISPs
are often not providing a managed CE router service, particularly to
medium and large customers that require eBGP). Furthermore, ISPs are
using methods to perform ASN migration that do not require
coordination with customers. Ultimately, this means there is not a
finite period of time after which legacy ASNs will be completely
expunged from the ISP's network. In fact, it is common that legacy
ASNs and the associated External BGP AS Migration mechanisms
discussed in this document can and do persist for several years, if
not longer. Thus, it is prudent to plan that legacy ASNs and
associated External BGP AS Migration mechanisms will persist in an
operational network indefinitely.
With respect to the Internal BGP AS Migration mechanism, all of the
routers to be consolidated into a single, permanently retained ASN
are under the administrative control of a single entity. Thus,
completing the migration from iBGP sessions using the legacy ASN to
the permanently retained ASN is more straightforward and could be
accomplished in a matter of days to months. Finally, good
operational hygiene would dictate that it is good practice to avoid
using "Internal BGP AS Migration" capability over a long period of
time for reasons of not only operational simplicity of the network,
but also reduced reliance on that mechanism during the ongoing life
cycle management of software, features, and configurations that are
maintained on the network.
6. Security Considerations
This document discusses a process by which one ASN is migrated into
and subsumed by another. This involves manipulating the AS_PATH
attribute with the intent of not increasing the AS_PATH length, which
would typically cause the BGP route to no longer be selected by BGP's
Path Selection Algorithm in others' networks. This could result in
sudden and unexpected shifts in traffic patterns in the network,
potentially resulting in congestion.
Given that these mechanisms can only be enabled through configuration
of routers within a single network, standard security measures should
be taken to restrict access to the management interface(s) of routers
that implement these mechanisms. Additionally, BGP sessions SHOULD
be protected using TCP Authentication Option [RFC5925] and the
Generalized TTL Security Mechanism [RFC5082]
7. References
7.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<http://www.rfc-editor.org/info/rfc2119>.
[RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A
Border Gateway Protocol 4 (BGP-4)", RFC 4271,
DOI 10.17487/RFC4271, January 2006,
<http://www.rfc-editor.org/info/rfc4271>.
[RFC4456] Bates, T., Chen, E., and R. Chandra, "BGP Route
Reflection: An Alternative to Full Mesh Internal BGP
(IBGP)", RFC 4456, DOI 10.17487/RFC4456, April 2006,
<http://www.rfc-editor.org/info/rfc4456>.
7.2. Informative References
[ALU] Alcatel-Lucent, "BGP Local AS attribute", 2006-2012,
<https://infoproducts.alcatel-lucent.com/html/0_add-h-f/
93-0074-10-01/7750_SR_OS_Routing_Protocols_Guide/
BGP-CLI.html#709567>.
[CISCO] Cisco Systems, Inc., "BGP Support for Dual AS
Configuration for Network AS Migrations", 2013,
<http://www.cisco.com/c/en/us/td/docs/ios-xml/
ios/iproute_bgp/configuration/xe-3s/asr1000/
irg-xe-3s-asr1000-book/irg-dual-as.html>.
[JUNIPER] Juniper Networks, Inc., "Understanding the BGP Local AS
Attribute", December 2013,
<http://www.juniper.net/techpubs/en_US/junos13.3/topics/
concept/bgp-local-as-introduction.html>.
[RFC5065] Traina, P., McPherson, D., and J. Scudder, "Autonomous
System Confederations for BGP", RFC 5065,
DOI 10.17487/RFC5065, August 2007,
<http://www.rfc-editor.org/info/rfc5065>.
[RFC5082] Gill, V., Heasley, J., Meyer, D., Savola, P., Ed., and C.
Pignataro, "The Generalized TTL Security Mechanism
(GTSM)", RFC 5082, DOI 10.17487/RFC5082, October 2007,
<http://www.rfc-editor.org/info/rfc5082>.
[RFC5398] Huston, G., "Autonomous System (AS) Number Reservation for
Documentation Use", RFC 5398, DOI 10.17487/RFC5398,
December 2008, <http://www.rfc-editor.org/info/rfc5398>.
[RFC5925] Touch, J., Mankin, A., and R. Bonica, "The TCP
Authentication Option", RFC 5925, DOI 10.17487/RFC5925,
June 2010, <http://www.rfc-editor.org/info/rfc5925>.
[RFC6793] Vohra, Q. and E. Chen, "BGP Support for Four-Octet
Autonomous System (AS) Number Space", RFC 6793,
DOI 10.17487/RFC6793, December 2012,
<http://www.rfc-editor.org/info/rfc6793>.
[RFC6996] Mitchell, J., "Autonomous System (AS) Reservation for
Private Use", BCP 6, RFC 6996, DOI 10.17487/RFC6996, July
2013, <http://www.rfc-editor.org/info/rfc6996>.
Appendix A. Implementation Report
As noted elsewhere in this document, this set of migration mechanisms
has multiple existing implementations in wide use.
o Cisco [CISCO]
o Juniper [JUNIPER]
o Alcatel-Lucent [ALU]
This is not intended to be an exhaustive list, as equivalent features
do exist in other implementations; however, the authors were unable
to find publicly available documentation of the vendor-specific
implementation to reference.
Acknowledgements
Thanks to Kotikalapudi Sriram, Stephane Litkowski, Terry Manderson,
David Farmer, Jaroslaw Adam Gralak, Gunter Van de Velde, Juan
Alcaide, Jon Mitchell, Thomas Morin, Alia Atlas, Alvaro Retana, and
John Scudder for their comments.
Authors' Addresses
Wesley George
Time Warner Cable
13820 Sunrise Valley Drive
Herndon, VA 20171
United States
Phone: +1 703-561-2540
Email: wesley.george@twcable.com
Shane Amante
Apple, Inc.
1 Infinite Loop
Cupertino, CA 95014
United States
Email: amante@apple.com
AI ELI5
On this page
PROBLEMWhen an ISP merges networks and retires an AS number, its eBGP peers expect the old number, and a longer AS_PATH would shift traffic in other networks.
MECHANISMLocal per-neighbor settings let a router open sessions with the old ASN and edit the AS_PATH it stores and sends (Local AS, No Prepend Inbound, Replace Old AS). A separate iBGP setting accepts either ASN inside the network.
WHY IT MATTERSCustomers can migrate when convenient and the AS_PATH length stays the same. The RFC records these existing vendor features so that future BGP work does not break them.
§1Why AS Numbers Get Migrated
RFC 7705 documents commonly used BGP mechanisms for Autonomous System Number (ASN) migration that are not formally part of BGP-4 (RFC 4271). They are local to one BGP speaker, need no negotiation with neighbors, and need not interwork with one another, so vendor implementations vary slightly. The RFC writes them down so new implementations succeed and future protocol work that reads, copies, manipulates or compares the AS_PATH attribute does not break them.
The need comes from mergers, acquisitions and divestitures. The goal is consistent configuration across the combined network, without lengthening the AS_PATH, because BGP path selection prefers the shortest AS_PATH and a longer one would likely cause sudden traffic shifts. By default a BGP session needs the ASN to match on both ends, so without these tools an ISP would have to coordinate a change with, in some cases, tens of thousands of customers.
When two companies join, one of them has to change its name. These tricks let the change happen without calling every customer on the same day.
§2The Migration Scenario
The RFC uses two ISPs, A with AS 64500 and B with AS 64510. AS 64500 is kept and AS 64510 is retired. Customer C (AS 64496) attaches to ISP B, and customer D (AS 64499) attaches to ISP A. Afterwards there is one network, ISP A', using AS 64500 everywhere, and external speakers must not see a longer AS_PATH. The examples use ASNs reserved for documentation, standing in for globally unique ASNs.
In a typical maintenance window, ISP B changes the global ASN of a Provider Edge (PE) router from 64510 to 64500. That breaks its eBGP sessions to customer devices still using 64510. So ISP B configures migration mechanisms on every customer-facing session: they let the router set up neighbors using the legacy ASN, and edit the AS_PATH on routes coming in and going out. The scenario assumes neither ISP uses BGP confederations.
Two shops merge and keep one name. The staff at the old shop's counter change their badges, but still answer to the old name when regular customers call it.
§3External Session Mechanisms
Three settings work on eBGP sessions toward customers.
- Local AS puts a locally chosen ASN in the "My Autonomous System" field of the BGP OPEN message, instead of the global ASN, for a neighbor or group of neighbors. This is how the migrated PE keeps its session with customer C using 64510. By default it also prepends the old ASN to routes received from the customer.
- No Prepend Inbound stops that prepend when the route is installed or sent to iBGP neighbors. Customer C's route stays at one ASN in length inside ISP A'. With both settings, CE-A sees AS_PATH
64500 64496. With Local AS alone it would see64500 64510 64496. - Replace Old AS handles the outbound direction. In some implementations Local AS lengthens routes sent to the customer, so customer C would see
64510 64500 64499. With Replace Old AS the router appends only the Local AS value, and the customer sees64510 64499, the same length as before migration.
These mechanisms MUST be configurable per neighbor or per neighbor group. A speaker SHOULD send OPEN with the local ASN, not the global one, so the neighbor does not raise a "Bad Peer AS" error.
One trick lets the router answer to the old name. The second stops it writing the old name on mail it passes inward. The third stops it writing both names on mail it sends outward.
| Direction and settings | AS_PATH seen |
|---|---|
| CE-A, route from customer C, Local AS only | 64500 64510 64496 |
| CE-A, route from customer C, Local AS plus No Prepend Inbound THIS RFC | 64500 64496 |
| CE-B, route from customer D, Local AS only | 64510 64500 64499 |
| CE-B, route from customer D, with Replace Old AS THIS RFC | 64510 64499 |
§4Rules for Processing Updates
A speaker follows RFC 4271 Section 5.1.2 for updates from a Local AS neighbor, plus extra steps. Internally, it SHOULD append the Local AS ASN to the AS_PATH before installing the route or advertising to an iBGP neighbor. When advertising to an eBGP neighbor, it SHOULD first append the global ASN and then the Local AS value.
With No Prepend Inbound, the router MUST NOT append the Local AS ASN for installation or iBGP advertisement, but MUST still append the global ASN as normal when advertising to other local eBGP neighbors. With Replace Old AS, outbound updates MUST NOT include the global ASN and MUST append only the Local AS ASN.
An implementation MAY try to open the session with either the Local AS or the global ASN. If the session comes up on the global ASN, the AS_PATH changes SHOULD NOT be done. That lets the customer move to the new ASN without direct coordination. Four-octet ASNs vary slightly; the RFC points to RFC 6793.
If routers remain in the old ASN, they may accept routes with 64510 removed as if they had not passed through it, which can cause a loop that normal loop prevention does not catch. Careful filtering between old-ASN and new-ASN routers is needed.
There are exact rules for which names get written on the envelope and when. If the customer has already switched to the new name, the mailroom leaves the envelope alone.
§5Internal BGP AS Migration
Inside the network, the traditional method is to change the global ASN and every iBGP neighbor's remote ASN on each router at once. That is slow, resets sessions and risks visible outages, especially beyond a few dozen routers. The Internal BGP AS Migration mechanism lets an iBGP speaker form a single iBGP session using either the old or the new ASN, so both can coexist temporarily. It does not change the AS_PATH, and the session behaves like a normal iBGP session under RFC 4271 and RFC 4456.
The RFC's example uses redundant route reflectors (RRs), though they are not required.
- On one RR of a pair, change the global ASN to the retained one and enable the mechanism on all its iBGP sessions.
- Repeat on the other RR, removing the mechanism from sessions between RRs. It stays on RR client sessions.
- Establish iBGP sessions between all RRs of both networks, so the network looks like one.
- Migrate each legacy PE: change its global ASN and RR session ASN, and add the external mechanisms on its eBGP sessions in the same window.
- Remove the mechanism from the RRs in an orderly way.
This can take days or weeks, with the only observable disruption when each PE is changed.
Inside the school, teachers answer to either the old or the new name for a while, so everyone can switch one at a time without the whole school stopping.
§6Implementation Requirements
The internal mechanism MUST be configurable per neighbor or neighbor group. When enabled, a speaker MUST accept an OPEN and establish an iBGP session if the ASN in "My Autonomous System" is either the global ASN or a locally configured ASN. It MUST send its own OPEN with one of the two. To avoid deadlock when both ends use the mechanism, it SHOULD send the global ASN first, and use the local ASN only if the peer answers with the "Bad Peer AS" error. It MUST treat updates as on a native iBGP session.
Appendix A says the mechanisms have multiple implementations in wide use and names Cisco, Juniper and Alcatel-Lucent, adding that the list is not exhaustive.
If both sides use the trick, one tries the new name first and only switches to the old name if the other side says, "I do not know that name."
§7Operations and Security
Legacy ASNs and the external mechanisms tend to stay a long time. ISPs often do not control customers' routers, and companies accumulate legacy ASNs through mergers, so the RFC advises planning for them to persist indefinitely. External legacy ASNs can last several years or longer. The internal mechanism is under one organization's control, so finishing it takes days to months, and it is good practice not to leave it enabled for a long time.
On security, the mechanisms are configured only on routers within one network, so access to router management interfaces should be restricted. The RFC also says BGP sessions SHOULD be protected with the TCP Authentication Option (RFC 5925) and the Generalized TTL Security Mechanism (RFC 5082). It warns that AS_PATH manipulation, if it changes path length, could shift traffic and cause congestion.
Old names tend to linger for years, so the mailroom plans to keep handling them. Only trusted staff should be allowed to change these rules.
§8References
[RFC4271]RFC 4271Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A Border Gateway Protocol 4 (BGP-4)", RFC 4271, January 2006.January 2006[RFC4456]RFC 4456Bates, T., Chen, E., and R. Chandra, "BGP Route Reflection: An Alternative to Full Mesh Internal BGP (IBGP)", RFC 4456, April 2006.April 2006[RFC6793]RFC 6793Vohra, Q. and E. Chen, "BGP Support for Four-Octet Autonomous System (AS) Number Space", RFC 6793, December 2012.December 2012[RFC5065]RFC 5065Traina, P., McPherson, D., and J. Scudder, "Autonomous System Confederations for BGP", RFC 5065, August 2007.August 2007[RFC5925]RFC 5925Touch, J., Mankin, A., and R. Bonica, "The TCP Authentication Option", RFC 5925, June 2010.June 2010[RFC5082]RFC 5082Gill, V., et al., "The Generalized TTL Security Mechanism (GTSM)", RFC 5082, October 2007.October 2007[RFC5398]RFC 5398Huston, G., "Autonomous System (AS) Number Reservation for Documentation Use", RFC 5398, December 2008.December 2008