Internet DRAFT - draft-ietf-trill-centralized-replication
draft-ietf-trill-centralized-replication
TRILL Working Group W. Hao
INTERNET-DRAFT Y. Li
Intended Status: Standards Track Huawei Technologies
Updates: 6325 M. Durrani
Equinix
S. Gupta
IP Infusion
A. Qu
MediaTec
Expires: July 2018 January 25, 2018
Centralized Replication for Active-Active BUM Traffic
draft-ietf-trill-centralized-replication-13.txt
Abstract
In TRILL active-active access, an RPF (Reverse Path Forwarding) check
failure issue may occur when using the pseudo-nickname mechanism
specified in RFC 7781. This draft describes a solution to resolve
this RPF check failure issue through centralized replication. All
ingress RBridges send BUM (Broadcast, Unknown unicast and Multicast)
traffic to a centralized node with unicast TRILL encapsulation.
When the centralized node receives the BUM traffic, it decapsulates
the packets and forwards them to their destination RBridges using
a distribution tree established as per TRILL base protocol RFC 6325.
To avoid RPF check failure on a RBridge sitting between the ingress
RBridge and the centralized replication node, some change in the
RPF calculation algorithm is required. RPF checks on each RBridge
MUST be calculated as if the centralized node was the ingress
RBridge, instead of being calculated using the actual ingress
RBridge. This document updates RFC 6325.
Status of this Memo
This Internet-Draft is submitted to IETF in full conformance with
the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
Hao & Li, et al Expires July 25, 2018 [Page 1]
Internet-Draft Centralized replication for BUM traffic January 2018
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six
months and may be updated, replaced, or obsoleted by other
documents at any time. It is inappropriate to use Internet-Drafts
as reference material or to cite them other than as "work in
progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/1id-abstracts.html
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
Copyright Notice
Copyright (c) 2018 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 ................................................3
2. Conventions used in this document............................4
3. Centralized replication solution overview....................5
4. Frame duplication from remote RBridge........................6
5. Local forwarding behavior on ingress RBridge.................7
6. Loop prevention among RBridges in an edge group..............8
7. Centralized replication forwarding process...................9
8. BUM traffic load balancing among multiple centralized nodes..11
9. Co-existing with the CMT solution........................... 12
Hao & Li, et al Expires July 25, 2018 [Page 2]
Internet-Draft Centralized replication for BUM traffic January 2018
10. Network upgrade analysis................................... 13
11. TRILL protocol extensions.................................. 13
11.1. "R" and "C" Flag in the Nickname Flags APPsub-TLV......13
12. Security Considerations.................................... 14
13. IANA Considerations........................................ 15
14. References ................................................ 15
14.1. Normative References.................................. 15
14.2. Informative References................................ 16
15. Acknowledgments ........................................... 16
1. Introduction
The IETF TRILL (Transparent Interconnection of Lots of Links)
[RFC6325] protocol provides loop free and per hop based multipath
data forwarding with minimum configuration. TRILL uses IS-IS
[RFC6165] [RFC7176] as its control plane routing protocol and
defines a TRILL specific header for user data.
Customer Equipment (CE) devices can be multi-homed to a set of edge
RBridges forming an edge group where active-active service can be
provided. In that case, all of the uplinks from a CE are handled via
a Local Active-Active Link Protocol (LAALP [RFC7379]) such as Multi-
Chassis Link Aggregation (MC-LAG) or Distributed Resilient Network
Interconnect (DRNI) [802.1AX]. An active-active flow-based load
sharing mechanism can achieve better load balancing and high
reliability. A CE device can be a layer 3 end system by itself or a
bridge switch through which layer 3 end systems access to TRILL
campus.
In active-active access, the pseudo-nickname solution in [RFC7781]
can be used to avoid MAC flip-flop on remote RBridges. The basic
idea is to use a virtual RBridge (RBv) with a single pseudo-nickname
to represent an edge group. Any member RBridge of that edge group
uses this pseudo-nickname rather than its own nickname as the
ingress nickname when it injects TRILL data frames to TRILL campus.
The use of the nickname solves the address flip-flop issue by
setting the MAC address learnt by a remote RBridge to the pseudo-
nickname. However, it introduces another issue of incorrect packet
dropping as follows: When a pseudo-nickname is used by an edge
RBridge as the ingress nickname to forward BUM (Broadcast, Unknown
unicast and Multicast) traffic, any RBridges (RBn) sitting between
the ingress RBridge and the distribution tree root will treat the
traffic as if it was ingressed from the virtual RBridge RBv. If the
same distribution tree is used by different edge RBridges of the
same RBv, the traffic may arrive at some RBn from different ports.
Then the RPF (Reverse Path Forwarding) check required by TRILL
Hao & Li, et al Expires July 25, 2018 [Page 3]
Internet-Draft Centralized replication for BUM traffic January 2018
[RFC6325] fails, and the BUM traffic received on unexpected ports
will be dropped by RBn.
This document specifies a centralized replication solution for
broadcast, unknown unicast and multicast (BUM) traffic forwarding to
resolve the issue of incorrect packet drop caused by the RPF check
failure in the virtual RBridge case. The basic idea is that all
ingress RBridges send BUM traffic to a centralized node, which
MUST be a distribution tree root, using unicast TRILL
encapsulation. When the centralized node receives the packets, it
decapsulates and forwards them to their destination RBridges using a
distribution tree established as per the TRILL base protocol. This
document updates [RFC6325]; under [RFC6325] multi-destination
traffic is ingressed to a multi-destination TRILL Data packet
but under this document, when using the centralized replication
feature, multi-destination traffic is initially ingressed to a
unicast TRILL Data packet.
2. Conventions used in this document
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in BCP 14
[RFC2119] [RFC8174] when, and only when, they appear in all capitals,
as shown here.
The acronyms and terminology in [RFC6325] are used herein with the
following additions:
BUM - Broadcast, Unknown unicast, and Multicast
CE - As in [RFC7783], Customer Equipment device (end station or
bridge). The device can be either physical or virtual equipment.
Data Label - VLAN or FGL [RFC7172]
DF - Designated Forwarder [RFC7781]
FGL - Fine Grained Label [RFC7172].
LAALP - Local Active-Active Link Protocol [RFC7379].
MAC flip flop - A problem where the attachment point of a MAC
address appears to a remote switch to keep changing. See Section 3.3
of [RFC7379].
MC-LAG - Multi-Chassis Link Aggregation.
RPF - Reverse Path Forwarding.
Hao & Li, et al Expires July 25, 2018 [Page 4]
Internet-Draft Centralized replication for BUM traffic January 2018
3. Centralized replication solution overview
When an edge RBridge receives BUM traffic from a CE device, it uses
unicast TRILL encapsulation instead of multicast encapsulation to
send the packets to a centralized node. The centralized node MUST be
a distribution tree root. Distribution tree roots are normally
chosen to be high capacity core RBridges with many high bandwidth
adjacencies and this constraint makes its practical, as described
below, to support centralized replication with only software changes
to transit RBridges.
The TRILL header of the unicast TRILL encapsulation contains an
"ingress RBridge nickname" field and an "egress RBridge nickname"
field [RFC6325]. If the ingress RBridge receives the BUM packet from
a port that is in an active-active edge group using [RFC7781], it
sets the ingress RBridge nickname to be the pseudo-nickname rather
than its own nickname to avoid MAC flip-flop (see Section 3.3 of
[RFC7379]) on remote RBridges. The egress RBridge nickname is set
to a special nickname of the centralized node which is used to
differentiate the centralized replication purpose unicast TRILL
encapsulation from a normal unicast TRILL encapsulation. This
special nickname is called an R-nickname.
When the centralized RBridge receives a unicast TRILL encapsulated
packet with its R-nickname as egress nickname, it decapsulates the
packet. Then the centralized RBridge replicates and forwards the
BUM packet to the packet's destination RBridges using one of the
distribution trees established as per TRILL base protocol. It MUST
use a distribution tree whose tree root is the centralized RBridge
itself. (An RBridge may by the root of more than one tree.) When the
centralized RBridge forwards the BUM traffic, it simply sends it on
the distribution tree as if it was a locally ingressed frame except
that the ingress nickname remains the same as that in the packet it
received to ensure that the MAC address learning by all egress
RBridges is bound to the pseudo-nickname.
When the replicated packet is forwarded by each RBridge along the
distribution tree starting from the centralized node, an RPF check
is performed as per [RFC6325]. For any RBridge sitting between the
ingress RBridge and the centralized replication node, the incoming
port of such BUM packet should be the centralized node facing port
as the multicast traffic always comes from the centralized node in
this solution. However the RPF port as the result of distribution
tree calculation as specified in [RFC6325] will be the real ingress
RBridge facing port as it uses the edge group's virtual RBridge as
the ingress RBridge, so the RPF check will fail.
Hao & Li, et al Expires July 25, 2018 [Page 5]
Internet-Draft Centralized replication for BUM traffic January 2018
To solve this problem, some change in the RPF test is required. In
this case, the RPF calculation on each RBridge should use the
centralized node as the ingress RBridge for each tree for which that
node is the root instead of the real ingress virtual RBridge to
perform the calculation. As a result, RPF check will accept traffic
on the centralized node facing port of the RBridge for multi-
destination traffic. This prevents incorrect frame drops by the RPF
check.
The change in the actual RPF check on a received multi-destination
TRILL data packet is easy. The [RFC6325] RPF check is a check to see
if a triple of {ingress nickname, tree, receiving RBridge port} is
allowed. (The tree is indicated by the nickname of its root which is
stored in the TRILL Header "egress nickname" field.) When
determining the RFC check, if "ingress nickname" is using centralized
replication (indicated by a C-nickname, see Section 9), then the check
is based on distribution from the tree root. If "ingress nickname"
is not using centralized replication, then the check is based on
distribution from the RBridge having the ingress nickname.
To differentiate the centralized replication unicast TRILL
encapsulation from normal unicast TRILL encapsulation, the R-
nickname is introduced for centralized replication. When the
centralized node receives unicast TRILL encapsulation traffic with
the egress nickname R-nickname, it decapsulates the packet and then
forwards the packet to the destination RBridges through a
distribution tree for which it is the root by re-encapsulation as
aforementioned. In TRILL, RBridges can hold multiple nicknames so
the centralized RBridge simply obtains another nickname to use as
the R-nickname. The centralized RBridge or RBridges should announce
their R-nickname to all TRILL campus through the TRILL LSP extension
specified in Section 11.
4. Frame duplication from remote RBridge
Frame duplication may occur when a remote host sends a multi-
destination frame to a local CE which has an active-active
connection to the TRILL campus. To avoid local CE receiving multiple
copies from a remote RBridge, the designated forwarder (DF)
mechanism is supported for egress direction multicast traffic.
The DF election mechanism [RFC7781] allows only one port of one
RBridge in an active-active group to forward multicast traffic from
the TRILL campus to the local access side for each VLAN. The basic
idea of DF is to elect one RBridge per VLAN from an edge group to be
Hao & Li, et al Expires July 25, 2018 [Page 6]
Internet-Draft Centralized replication for BUM traffic January 2018
responsible for egressing the BUM traffic. [RFC7781] describes the
DF election mechanism among member RBridges involving in an edge
group.
If the DF election mechanism is used for frame duplication
prevention, access ports on an RBridge are categorized as one of
three types: non-group, group DF port and group non-DF port. The
last two types can be called group ports. Each of the group ports
is associated with a pseudo-nickname. If consistent nickname
allocation to edge group RBridges is used, it is possible that same
pseudo-nickname is associated with more than one port on a single
RBridge. A typical scenario is that CE1 is connected to RB1 & RB2
by LAALP1 while CE2 is connected to RB1 & RB2 by LAALP2.
In order to conserve the number of pseudo-nicknames used, member
ports for both LAALP1 and LAALP2 on RB1 & RB2 are all associated
with the same pseudo-nickname.
5. Local forwarding behavior on ingress RBridge
When an ingress RBridge (RB1) receives BUM traffic from a local
active-active connected CE (CE1) device, the traffic will be
injected into the TRILL campus with TRILL encapsulation, and it will
be replicated and forwarded to all destination RBridges through
central replication, including the ingress RBridge itself, along a
TRILL distribution tree. To avoid the traffic looping back to the
original sender CE, an ingress nickname of the CE group's pseudo-
nickname is used for traffic filtering.
However, if there are two CEs, say CE1 and CE2, connecting to the
ingress RB1 and each associated with the same pseudo-nickname, RB1
needs to locally replicate and forward to CE2, because another copy
of the BUM traffic between CE1 and CE2 through TRILL campus will be
blocked by the traffic filtering.
If CE1 and CE2 are not associated with the same pseudo-nickname, the
copy of the BUM traffic between CE1 and CE2 through TRILL campus
won't be blocked by the traffic filtering. To avoid duplicated
traffic on receiver CE, there cannot be local replicated BUM traffic
between these two CEs on ingress RB1.
In summary, to ensure correct BUM traffic forwarding behavior for
each CE, the local replication behavior on the ingress RBridge is as
follows:
1. Replicate to the active-active group ports associated with the
same pseudo-nickname as that associated with the incoming port.
Hao & Li, et al Expires July 25, 2018 [Page 7]
Internet-Draft Centralized replication for BUM traffic January 2018
2. Do not replicate to active-active group ports associated with
other pseudo-nicknames.
3. Do not replicate to non-edge-group ports.
The above local forwarding behavior on the ingress RBridge of RB1
can be called centralized replication local forwarding behavior A.
If ingress RBridge RB1 itself is the centralized replication node,
BUM traffic injected by RB1 into the TRILL campus won't loop back to
RB1. In this case, the local forwarding behavior is called
centralized replication local forwarding behavior B. Behavior B on
RB1 is as follows:
1. Local replication to the ports associated with the same
pseudo-nickname as that associated with the incoming port.
2. Local replication to the group DF port associated with
different pseudo-nicknames. Do not replicate to group non-DF port
associated with different pseudo-nicknames.
3. Local replication to non-edge-group ports.
6. Loop prevention among RBridges in an edge group
If a CE sends a broadcast, unknown unicast, or multicast (BUM)
packet through a DF port to an ingress RBridge, that RBridge will
forward that packet to all or a subset of the other RBridges that
only have non-DF ports for that active-active group. Because BUM
traffic forwarding to non-DF ports isn't allowed, in this case the
frame won't loop back to the CE.
If a CE sends a BUM packet through a non-DF port to an ingress
RBridge, say RB1, then RB1 will forward that packet to other
RBridges that have a DF port for that active-active group. In this
case the frame will loop back to the CE and the traffic split-
horizon filtering mechanism is used to avoid looping back among
RBridges in the edge group.
This split-horizon mechanism relies on the ingress nickname field in
the TRILL header to check if a packet's egress port belongs to a
same active-active group as the packet's incoming port to the TRILL
campus.
Hao & Li, et al Expires July 25, 2018 [Page 8]
Internet-Draft Centralized replication for BUM traffic January 2018
When the ingress RBridge receives BUM traffic from an active-active
connected CE device, the traffic will be sent through the TRILL
campus with TRILL encapsulation to a centralized RBridge. There it
will be replicated and forwarded to its destination RBridges, which
include ingress RBridge itself, through a TRILL distribution tree.
If the same pseudo-nickname is used for two active-active access CEs
as ingress nickname, an egress RBridge can use that nickname to
filter traffic forwarding to all local CEs. In this case, the
traffic between these two CEs goes through the local RBridge and
another copy of the traffic from the TRILL campus is filtered. If
different ingress nicknames are used for two connecting CE devices,
the access ports connecting to these two CEs should be isolated from
each other. The BUM traffic between these two CEs should go through
the TRILL campus, otherwise the destination CE connected to same
RBridge with the sender CE will receive two copies of the traffic.
7. Centralized replication forwarding process
+-----------+
| (RB5) |
+-----------+
|
+-----------+
| (RB4) |
+-----------+
| | |
-------- | --------
| | |
+------+ +------+ +------+
|(RB1) | |(RB2) | | (RB3)|
+------+ +------+ +------+
* | * | * | ^
* | * | * | ^
* ----------*-------------*-- ^
***************************** | ^
* | ^
LAALP1 * LAALP2 | ^
+------+ +------+ +------+
| CE1 | | CE2 | | CE3 |
+------+ +------+ +------+
Note: The asterisk line, hyphen & vertical bar line, and circumflex
line in this figure indicate the connection of the various CEs to
the various RBs.
Figure 1 TRILL Active-active access
Assuming the centralized replication solution is used in the example
network of above figure 1, RB5 is the distribution tree root and
Hao & Li, et al Expires July 25, 2018 [Page 9]
Internet-Draft Centralized replication for BUM traffic January 2018
centralized replication node, CE1 and CE2 are active-active accessed
to RB1, RB2, and RB3 through LAALP1 and LAALP2 respectively, CE3 is
single homed to RB3. The RBridge's own nicknames of RB1 to RB5 are
nick1 to nick5 respectively. RB1, RB2, and RB3 use the same pseudo-
nickname for LAALP1 and LAALP2; that pseudo-nickname is P-nick. The
R-nickname on the centralized replication node of RB5 is S-nick.
The BUM traffic forwarding process from CE1 to CE2 and CE3 is as
follows:
1. CE1 sends BUM traffic to RB3.
2. RB3 replicates and sends the BUM traffic to CE2 locally. RB2
also sends the traffic to RB5 using unicast TRILL encapsulation. In
the TRILL Header, the ingress nickname is set as P-nick and the
egress nickname is set as S-nick.
3. RB5 decapsulates the unicast TRILL Data packet. Then it uses a
distribution tree for which it is the root to forward the packet as
a multi-destination TRILL Data packet. The egress nickname in the
multi-destination TRILL Header is the nick5 and the ingress nickname
is still P-nick. If RB3 had sent the unicast to some nickname that
was not an R-nickname, the packet will not be re-encapsulate. If it
is sent to an R-nickname that is not a tree root, it will either be
not forwarded at all or, if it is re-encapsulated and forwarded,
will be subject to incorrect pruning and will not be delivered to
all of its intended recipients.
4. RB4 receives multicast TRILL traffic from RB5. Traffic
incoming port is the up port facing the distribution tree root,
RB4's RPF check will be correct based on the changed RPF port
calculation algorithm in this document. After the RPF check is
performed, it forwards the traffic to all other egress RBridges (RB1,
RB2, and RB3).
5. RB3 receives multicast TRILL traffic from RB4. It decapsulates
the multi-destination TRILL Data packet. Because the ingress
nickname of P-nick is equivalent to the nickname of local LAALPs
connecting to CE1 and CE2, RB3 doesn't forward the traffic to CE1
and CE2 to avoid duplicated frame. RB3 only forwards the packet to
CE3.
6. RB1 and RB2 receive multicast TRILL traffic from RB4. The
forwarding process is similar to the process on RB3, i.e., because
the ingress nickname of P-nick is equivalent to the nickname of the
Hao & Li, et al Expires July 25, 2018 [Page 10]
Internet-Draft Centralized replication for BUM traffic January 2018
local LAALPs connecting CE1 and CE2, they also don't forward the
traffic to local CE1 and CE2.
8. BUM traffic load balancing among multiple centralized nodes
To support unicast TRILL encapsulation BUM traffic load balancing,
multiple centralized replication nodes can be deployed and the
traffic can be spread over these nodes based on data label (VLAN or
FGL). Furthermore, if it was desirable for a centralized node to be
sent more of this BUM traffic, it could hold two or more R-nicknames.
The share of BUM traffic it would receive would be proportional to
the number of R-nicknames it held.
Assuming there are k different R-nicknames held by centralized nodes
in a TRILL campus. The VLAN-based (or FGL-based [RFC7172]) load
balancing algorithm used by ingress active-active access RBridge is
as follows:
1. All R-nicknames are ordered and numbered from 0 to k-1 in
ascending order treating the nicknames as unsigned 16-bit integers.
2. For data label ID m, choose the R-nickname whose index is
given by (m mod k) as egress nickname for BUM traffic unicast TRILL
encapsulation.
For examples, there are 3 R-Nicknames (RN). The RNs will be ordered
RN0 to RN2. Assuming there are 5 VLANs from VLAN ID 1 to ID 5
spreading among edge RBridges, the traffic in VLAN 1 will go to RN1,
VLAN 2 will go to RN2, and so on.
When an ingress RBridge participating in active-active connection
receives BUM traffic from local CE, the RBridge decides which R-
nickname to send the traffic to based on the VLAN-based load
spreading algorithm, thus data-label-based load balancing for the
BUM traffic can be achieved using multiple centralized
nodes/multiple R-nicknames.
Hao & Li, et al Expires July 25, 2018 [Page 11]
Internet-Draft Centralized replication for BUM traffic January 2018
9. Co-existing with the CMT (RFC 7783) solution
+------+ +------+
|(RB6) | |(RB7) |
+------+ +------+
------------------|-----------|----------------------
| | | | |
+------+ +------+ +------+ +------+ +------+
|(RB1) | |(RB2) | |(RB3) | |(RB4) | |(RB5) |
+------+ +------+ +------+ +------+ +------+
| | | | |
------------ -------------------------
| |
+------+ +------+
| CE1 | | CE2 |
+------+ +------+
Figure 2 CMT and centralized replication co-existing scenario
Both the centralized replication solution and the Coordinated
Multicast Trees (CMT) [RFC7783] solution rely on using
pseudo-nicknames to avoid MAC flip-flop on remote RBridges.
These two solutions can co-exist in a single TRILL campus.
Each solution can be selected by each active-active edge
group of RBridges independently.
As illustrated in Figure 2, RB1 and RB2 use CMT for CE1's active-
active access, RB3, RB4, and RB5 use the centralized replication for
CE2's active-active access.
For the centralized replication solution, edge group RBridges MUST
announce the local pseudo-nickname using Nickname Flags APPsub-TLV
with C-flag set. A nickname with the C-flag set is called a "C-
nickname". A transit RBridge will perform the centralized
replication specific RPF check algorithm if it receives TRILL Data
packets with a C-nickname as ingress nickname.
An edge group using CMT [RFC7783] MUST NOT set the C-nickname flag
on the pseudo-nickname it is using. This is already mandatory
behavior because any RBridge originating a Nickname Flags APPsub-TLV
is required by [RFC7780] to set any flag bit it does not know about
to zero. If a edge RBridge using CMT [RFC7783] nevertheless set the
C-bit for an edge group pseudo-nickname, it is very likely that BUM
traffic encapsulated with that nickname as ingress would be
incorrectly pruned early in its distribution and would thus reach
few (possibly none) of its intended targets. To avoid confusion, a
Hao & Li, et al Expires July 25, 2018 [Page 12]
Internet-Draft Centralized replication for BUM traffic January 2018
pseudo-nickname MUST NOT be shared between a centralized replication
edge group and a CMT-based edge group.
10. Network upgrade analysis
Centralized nodes will typically need software and hardware upgrades
to support centralized replication, which stitches together the
TRILL unicast traffic decapsulation process and the process of
normal TRILL multicast traffic forwarding along distribution tree.
Active-active connection edge RBridges will typically need software
and hardware upgrade to support unicast TRILL encapsulation for BUM
traffic; the process is similar to other head-end replication
processes.
Transit nodes typically need only a software upgrade to support the
changed RPF port calculation algorithm.
11. TRILL protocol extensions
Two new flags, "R" and "C", are specified in the Nickname Flags
APPsub-TLV [RFC7780]. A nickname with the "R" flag set is called an
R-nickname and a nickname with the the "C" flag set is called a C-
nickname. The R-nickname is a specialized nickname attached to a
centralized node to differentiate unicast TRILL encapsulated BUM
traffic from normal unicast TRILL traffic. The C-nickname flag is
set on the pseudo-nickname for each edge group that uses the
centralized replication. A C-nickname is a specialized pseudo-
nickname for which transit RBridges perform a different RPF check
algorithm for TRILL data packets with the C-nickname in the ingress
nickname field.
When active-active edge RBridges use centralized replication to
forward BUM traffic, the R-nickname is used as the egress nickname
and the C-nickname is used as ingress nickname in the TRILL header
for the unicast TRILL encapsulation of BUM traffic.
11.1. "R" and "C" Flag in the Nickname Flags APPsub-TLV
If this APPsub-TLV is being advertised by an RBridge that does not
have the nickname appearing in the Nickname Flags APPsub-TLV, the R
and C flag bits in the APPsub-TLV MUST be treated as if they were
zero. If an RBridge that is not a distribution tree root advertises
an R-nickname, that nickname MUST NOT be treated as an R-nickname
but rather as an ordinary nickname, that is, the R-nickname flag is
Hao & Li, et al Expires July 25, 2018 [Page 13]
Internet-Draft Centralized replication for BUM traffic January 2018
ignored for all purposes if the nickname is held by an RBridge that
is not a tree root.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
| Nickname |
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
|IN|SE|R | C| RESV |
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
NICKFLAG RECORD
o R = If R flag is one, it indicates that the advertising
TRILL switch holding Nickname is a centralized replication node, and
Nickname is used as egress nickname for edge group RBridges to
inject BUM traffic into the TRILL campus when the edge group
RBridges use this centralized replication solution for active-active
access. If flag is zero, that nickname will not be used for that
purpose.
o C = If C flag is one, it indicates that the TRILL traffic
with this nickname as an ingress nickname that requires the special
RPF check algorithm specified in Section 3. If flag is zero, that
nickname will not be used for that purpose.
It is possible, due to errors or due to transient inconsistent LSPs
when the link state database is converging after a configuration
change or the like for there to be inconsistent Nickname Flags
APPsub-TLVs for the same nickname. In this case it is RECOMMENDED
that the nickname be treated as if the R / C flag was set if any
Nickname Flags APPsub-TLV for that nickname has the R / C flag set.
12. Security Considerations
This draft does not introduce any extra security risks. A rogue
RBridge that is a tree root can attract traffic by advertising an R-
nickname. However, this does not represent a substantial increase in
risk as RBridges could cause problems in a number of other ways by
advertising low cost adjacencies or making themselves the highest
priority tree root or the like. In general, the protection against
an untrusted device acting as an RBridge and wrecking havoc is to
use IS-IS authentication [RFC5310] and configure and administer the
TRILL campus so that only trusted RBridges have the authentication
key.
Hao & Li, et al Expires July 25, 2018 [Page 14]
Internet-Draft Centralized replication for BUM traffic January 2018
For general TRILL Security Considerations, see [RFC6325]. For
Security Considerations related to pseudo-nickname active-active,
see [RFC7781].
13. IANA Considerations
IANA is requested to assign two bits in the Nickname Flags APPsubTLV
flags for the R and C bits discussed in Section 11.1 [Bits 3 and 4
suggested] and update the "NickFlags" Bits registry on the TRILL
Parameters page as follows:
Bit Mnemonic Description Reference
--- -------- -------------------- -----------
3 R Replication Nickname [This document]
4 C Special RFC Check [This document]
14. References
14.1. 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>.
[RFC5310] Bhatia, M., Manral, V., Li, T., Atkinson, R., White,
R., and M. Fanto, "IS-IS Generic Cryptographic Authentication", RFC
5310, DOI 10.17487/RFC5310, February 2009, <https://www.rfc-
editor.org/info/rfc5310>.
[RFC6165] Banerjee, A. and D. Ward, "Extensions to IS-IS for
Layer-2 Systems", RFC 6165, DOI 10.17487/RFC6165, April 2011,
<http://www.rfc-editor.org/info/rfc6165>.
[RFC6325] Perlman, R., Eastlake 3rd, D., Dutt, D., Gai, S., and A.
Ghanwani, "Routing Bridges (RBridges): Base Protocol Specification",
RFC 6325, DOI 10.17487/RFC6325, July 2011, <http://www.rfc-
editor.org/info/rfc6325>.
[RFC7172] Eastlake 3rd, D., Zhang, M., Agarwal, P., Perlman, R.,
and D. Dutt, "Transparent Interconnection of Lots of Links (TRILL):
Fine-Grained Labeling", RFC 7172, DOI 10.17487/RFC7172, May 2014,
<http://www.rfc-editor.org/info/rfc7172>.
[RFC7176] Eastlake 3rd, D., Senevirathne, T., Ghanwani, A., Dutt,
D., and A. Banerjee, "Transparent Interconnection of Lots of Links
Hao & Li, et al Expires July 25, 2018 [Page 15]
Internet-Draft Centralized replication for BUM traffic January 2018
(TRILL) Use of IS-IS", RFC 7176, DOI 10.17487/RFC7176, May 2014,
<http://www.rfc-editor.org/info/rfc7176>.
[RFC7780] Eastlake 3rd, D., Zhang, M., Perlman, R., Banerjee, A.,
Ghanwani, A., and S. Gupta, "Transparent Interconnection of Lots of
Links (TRILL): Clarifications, Corrections, and Updates", RFC 7780,
DOI 10.17487/RFC7780, February 2016, <http://www.rfc-
editor.org/info/rfc7780>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in
RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>.
14.2. Informative References
[RFC7781] Zhai, H., Senevirathne, T., Perlman, R., Zhang, M., and
Y. Li, "Transparent Interconnection of Lots of Links (TRILL):
Pseudo-Nickname for Active-Active Access", RFC 7781, DOI
10.17487/RFC7781, February 2016, <http://www.rfc-
editor.org/info/rfc7781>.
[RFC7379] Li, Y., Hao, W., Perlman, R., Hudson, J., and H. Zhai,
"Problem Statement and Goals for Active-Active Connection at the
Transparent Interconnection of Lots of Links (TRILL) Edge", RFC 7379,
DOI 10.17487/RFC7379, October 2014, <http://www.rfc-
editor.org/info/rfc7379>.
[RFC7783] Senevirathne, T., Pathangi, J., and J. Hudson,
"Coordinated Multicast Trees (CMT) for Transparent Interconnection
of Lots of Links (TRILL)", RFC 7783, DOI 10.17487/RFC7783, February
2016, <http://www.rfc-editor.org/info/rfc7783>.
15. Acknowledgments
The authors wish to acknowledge the important contributions of
Donald Eastlake, Hongjun Zhai, Xiaomin Wu, Liang Xia, Francis Dupont.
Hao & Li, et al Expires July 25, 2018 [Page 16]
Internet-Draft Centralized replication for BUM traffic January 2018
Authors' Addresses
Weiguo Hao
Huawei Technologies
101 Software Avenue,
Nanjing 210012
PR China
Email: haoweiguo@huawei.com
Yizhou Li
Huawei Technologies
101 Software Avenue,
Nanjing 210012
PR China
Email: liyizhou@huawei.com
Muhammad Durrani
Equinix
Email: mdurrani@equinix.com
Sujay Gupta
IP Infusion
RMZ Centennial
Mahadevapura Post
Bangalore - 560048
India
Email: sujay.gupta@ipinfusion.com
Andrew Qu
MediaTec
Email: laodulaodu@gmail.com
Hao & Li, et al Expires July 25, 2018 [Page 17]
