Internet DRAFT - draft-ietf-ccamp-gmpls-resource-sharing-proc
draft-ietf-ccamp-gmpls-resource-sharing-proc
CCAMP Working Group Xian Zhang
Internet-Draft Haomian Zheng, Ed.
Intended Status: Informational Huawei
Expires: May 24, 2015 Rakesh Gandhi, Ed.
Zafar Ali
Gabriele Maria Galimberti
Cisco Systems, Inc.
Pawel Brzozowski
ADVA Optical
November 20, 2014
RSVP-TE Signaling Procedure for GMPLS Restoration and Resource Sharing-
based LSP Setup and Teardown
draft-ietf-ccamp-gmpls-resource-sharing-proc-00
Abstract
In transport networks, there are requirements where Generalized
Multi-Protocol Label Switching (GMPLS) end-to-end recovery scheme
needs to employ restoration Label Switched Path (LSP) while keeping
resources for the working and/or restoration LSPs reserved in the
network after the failure occurs. This document reviews how the LSP
association is to be provided using Resource Reservation Protocol -
Traffic Engineering (RSVP-TE) signaling in the context of GMPLS end-
to-end recovery when using restoration LSP where failed LSP is not
torn down.
This document compliments existing standards by explaining the
missing pieces of information during the RSVP-TE signaling procedure
in support of resource sharing-based LSP setup/teardown in
GMPLS-controlled circuit networks. No new procedures or mechanisms
are defined by this document, and it is strictly informative in
nature.
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
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
Zhang, et al Expires May 24, 2015 [Page 1]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
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/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
Copyright Notice
Copyright (c) 2014 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. Problem Statement . . . . . . . . . . . . . . . . . . . . . . . 4
2.1. GMPLS Restoration . . . . . . . . . . . . . . . . . . . . . 4
2.1.1. 1+R Restoration . . . . . . . . . . . . . . . . . . . . 4
2.1.2. 1+1+R Restoration . . . . . . . . . . . . . . . . . . . 5
2.2. Resource Sharing-based LSP Setup/Teardown . . . . . . . . . 6
3. RSVP-TE Signaling For Restoration LSP Association . . . . . . . 7
4. RSVP-TE Signaling For Resource Sharing During LSP
Setup/Teardown . . . . . . . . . . . . . . . . . . . . . . . . 8
4.1. LSPs with Identical Tunnel ID . . . . . . . . . . . . . . . 8
4.1.1. Restoration LSP Setup . . . . . . . . . . . . . . . . . 8
4.1.2. LSP Reversion . . . . . . . . . . . . . . . . . . . . . 10
4.1.2.1. Make-while-break Reversion . . . . . . . . . . . . 11
4.1.2.2. Make-before-break Reversion . . . . . . . . . . . . 13
4.1.3. Re-optimization LSP Setup and Reversion . . . . . . . . 15
4.2. LSPs with Different Tunnel IDs . . . . . . . . . . . . . . 15
5. Security Considerations . . . . . . . . . . . . . . . . . . . . 16
6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 16
7. Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . 16
8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 17
8.1. Normative References . . . . . . . . . . . . . . . . . . . 17
8.2. Informative References . . . . . . . . . . . . . . . . . . 17
9. Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 19
Zhang, et al Expires May 24, 2015 [Page 2]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
1. Introduction
Generalized Multi-Protocol Label Switching (GMPLS) [RFC3945] defines
a set of protocols, including Open Shortest Path First - Traffic
Engineering (OSPF-TE) [RFC4203] and Resource ReserVation Protocol -
Traffic Engineering (RSVP-TE) [RFC3473]. These protocols can be used
to create Label Switched Paths (LSPs) in a number of deployment
scenarios with various transport technologies. The GMPLS protocol
set extends MPLS, which supports only Packet Switch Capable (PSC) and
Layer 2 Switch Capable interfaces (L2SC), to also cater for
interfaces capable of Time Division Multiplexing (TDM), Lambda
Switching (LSC) and Fiber Switching (FSC). These switching
technologies provide several protection schemes [RFC4426][RFC4427]
(e.g., 1+1, 1:N and M:N). Resource Reservation Protocol - Traffic
Engineering (RSVP-TE) signaling has been extended to support various
GMPLS recovery schemes [RFC4872][RFC4873], to establish Label
Switched Paths (LSPs), typically for working LSP and protecting LSP.
[RFC4427] Section 7 specifies various schemes for GMPLS recovery.
In GMPLS recovery schemes generally considered, restoration LSP is
signaled after the failure has been detected and notified on the
working LSP. In non-revertive recovery mode, working LSP is assumed
to be removed from the network before restoration LSP is signaled.
For revertive recovery mode, a restoration LSP is signaled while
working LSP and/or protecting LSP are not torn down in control plane
due to a failure. In transport networks, as working LSPs are
typically signaled over a nominal path, service providers would like
to keep resources associated with the working LSPs reserved. This is
to make sure that the service (working LSP) can use the nominal path
when the failure is repaired to provide deterministic behavior and
guaranteed Service Level Agreement (SLA). Consequently, revertive
recovery mode is usually preferred by recovery schemes used in
transport networks.
The Make-Before-Break (MBB) mechanisms exploiting the Shared-Explicit
(SE) reservation style can be employed in MPLS networks to avoid
double booking of resource during the process of LSP re-optimization
as specified in [RFC3209]. This method is also used in GMPLS-
controlled networks [RFC4872] [RFC4873] for end-to-end and segment
recovery of LSPs. This was further generalized to support resource
sharing oriented applications in MPLS networks as well as non-LSP
contexts, as specified in [RFC6780].
Due to the fact that the features of GMPLS-controlled networks
(specifically for TDM, LSC and FSC), are not identical to that of the
MPLS networks, additional considerations for resource sharing based
LSP association are needed. As defined in [RFC4872] and being
considered in this document, "fully dynamic rerouting switches normal
Zhang, et al Expires May 24, 2015 [Page 3]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
traffic to an alternate LSP that is not even partially established
only after the working LSP failure occurs. The new alternate route
is selected at the LSP head-end node, it may reuse resources of the
failed LSP at intermediate nodes and may include additional
intermediate nodes and/or links". During the signaling procedure for
resource sharing based LSP setup/teardown, the behaviors of the nodes
along the path may be different from that in the MPLS networks as
well as the effect it may have on the traffic delivery.
As described in [RFC6689], ASSOCIATION Object is used to identify the
LSPs for restoration using association type "Recovery" [RFC4872] and
for resource sharing using association type "Resource Sharing"
[RFC4873].
Following section describes the problem statements for the GMPLS
restoration and resource sharing based LSP setup and teardown.
2. Problem Statement
Problem statements for the GMPLS restoration schemes and resource
sharing-based LSP setup and teardown are described in this section.
2.1. GMPLS Restoration
2.1.1. 1+R Restoration
One example of the recovery scheme considered in this document is 1+R
recovery. The 1+R recovery is exemplified in Figure 1. In this
example, working LSP on path A-B-C-Z is pre-established. Typically
after a failure detection and notification on the working LSP, a
second LSP on path A-H-I-J-Z is established as a restoration LSP.
Unlike protection LSP, restoration LSP is signaled per need basis.
+-----+ +-----+ +-----+ +-----+
| A +----+ B +-----+ C +-----+ Z |
+--+--+ +-----+ +-----+ +--+--+
\ /
\ /
+--+--+ +-----+ +--+--+
| H +-------+ I +--------+ J |
+-----+ +-----+ +-----+
Figure 1: An Example of 1+R Recovery Scheme
During failure switchover with 1+R recovery scheme, in general,
working LSP resources are not released and working and restoration
LSPs coexist in the network. Nonetheless, working and restoration
Zhang, et al Expires May 24, 2015 [Page 4]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
LSPs can share network resources. Typically when failure is
recovered on the working LSP, restoration LSP is no longer required
and torn down (e.g., revertive mode).
2.1.2. 1+1+R Restoration
Another example of the recovery scheme considered in this document is
1+1+R. In 1+1+R, a restoration LSP is signaled for the working LSP
and/or the protecting LSP after the failure has been detected and
notified on the working LSP or the protecting LSP. The 1+1+R
recovery is exemplified in Figure 2.
+-----+ +-----+ +-----+
| D +-------+ E +--------+ F |
+--+--+ +-----+ +--+--+
/ \
/ \
+--+--+ +-----+ +-----+ +--+--+
| A +----+ B +-----+ C +-----+ Z |
+--+--+ +-----+ +-----+ +--+--+
\ /
\ /
+--+--+ +-----+ +--+--+
| H +-------+ I +--------+ J |
+-----+ +-----+ +-----+
Figure 2: An Example of 1+1+R Recovery Scheme
In this example, working LSP on path A-B-C-Z and protecting LSP on
path A-D-E-F-Z are pre-established. After a failure detection and
notification on a working LSP or protecting LSP, a third LSP on path
A-H-I-J-Z is established as a restoration LSP. The restoration LSP
in this case provides protection against a second order failure.
Restoration LSP is torn down when the failure on the working or
protecting LSP is repaired.
[RFC4872] Section 14 defines PROTECTION Object for GMPLS recovery
signaling. As defined, the PROTECTION Object is used to identify
primary and secondary LSPs using S bit and protecting and working
LSPs using P bit. Furthermore, [RFC4872] defines the usage of
ASSOCIATION Object for associating GMPLS working and protecting LSPs.
[RFC6689] Section 2.2 reviews the procedure for providing LSP
associations for GMPLS end-to-end recovery and covers the schemes
where the failed working LSP and/or protecting LSP are torn down.
This document reviews how the LSP association is to be provided for
GMPLS end-to-end recovery when using restoration LSP where working
Zhang, et al Expires May 24, 2015 [Page 5]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
and protecting LSP resources are kept reserved in the network after
the failure.
2.2. Resource Sharing-based LSP Setup/Teardown
+-----+ +-----+
| F +------+ G +--------+
+--+--+ +-----+ |
| |
| |
+-----+ +-----+ +--+--+ +-----+ +--+--+
| A +----+ B +-----+ C +--X---+ D +-----+ E |
+-----+ +-----+ +-----+ +-----+ +-----+
Figure 3: A Simple OTN Network
Using the Optical Transport Network (OTN) topology shown in Figure 3
as an example, GMPLS-controlled circuit LSP1 (A-B-C-D-E) is the
working LSP and it allows for resource sharing when the LSP is
dynamically rerouted due to link failure. Upon detecting the failure
of a link along the LSP1, e.g. Link C-D, node A needs to decide on
which alternate path it will establish an LSP to reroute the traffic.
In this case, A-B-C-F-G-E is chosen as the alternative path for the
LSP and the resources on the path segment A-B-C are re-used by this
LSP. Since this is an OTN network, which is different from the
packet-switching network, the label has a mapping into the data plane
resource used (e.g. wavelength) and also the nodes along the path
need to send triggering commands to data plane nodes for setting up
cross-connection accordingly during the RSVP-TE signaling process.
In this case, the following issues are left un-described in the
existing standards for resource sharing based LSP setup/teardown in
GMPLS-controlled circuit networks:
- Reservation style Shared-Explicit (SE) as defined in [RFC3209] may
not be applicable due to the nature of the GMPLS-controlled circuits.
It is not clear how reservation style is to be used by the GMPLS
LSPs for resource sharing.
- As described in [RFC3209], the purpose of Make-Before-Break (MBB)
is to "not disrupt traffic or adversely impact network operations
while TE tunnel rerouting is in progress". Due to the nature of the
GMPLS-controlled circuit networks, this may not be fulfilled under
certain scenarios. Thus, the name "Make-Before-Break" may no longer
hold true.
- The existing MBB method may not be sufficient to support LSP setup
and teardown with resource sharing.
Zhang, et al Expires May 24, 2015 [Page 6]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
- In [RFC3209], the MBB method assumes the old and new LSPs share the
same tunnel ID (i.e., sharing the same source and destination nodes).
[RFC4873] does not impose this constraint but limit the resource
sharing usage in LSP recoveries only. [RFC6780] generalizes the
resource sharing application, based on the ASSOCIATION Object, to be
useful in MPLS networks as well as in non-LSP association such as
Voice Call-Waiting. Recently, there are also requirements to
generalize resource sharing of LSPs with different tunnel IDs, such
as the one mentioned in [PCEP-RSO] and LSPs with LSP-stitching across
multi-domains. In this case, how the signaling process can make
intermediate nodes aware of the resource sharing constraint and
behave accordingly is an issue that needs to be described.
- The node behavior during traffic reversion in the GMPLS-controlled
circuit network is missing and should be clarified.
This document reviews the signaling procedure for resource
sharing-based LSP setup and teardown for GMPLS-based circuits in OTN
networks. This includes the node behavior description, besides
clarifying some un-discussed points for this process. Two typical
examples mentioned in this document are LSP restoration and LSP re-
optimization, where it is desirable to share resources. This
document does not define any RSVP-TE signaling extensions. If
necessary, discussion is provided to identify potential extensions to
the existing RSVP-TE protocol. It is expected that the extensions,
if there are any, will be addressed in separate documents.
3. RSVP-TE Signaling For Restoration LSP Association
Where GMPLS end-to-end recovery scheme needs to employ restoration
LSP while keeping resources for the working and/or protecting LSPs
reserved in the network after the failure, restoration LSP is
signaled with ASSOCIATION Object that has association type set to
"Recovery" [RFC4872] with the association ID set to the LSP ID of the
LSP it is restoring. For example, when a restoration LSP is signaled
for a working LSP, the ASSOCIATION Object in the restoration LSP
contains the association ID set to the LSP ID of the working LSP.
Similarly, when a restoration LSP is signaled for a protecting LSP,
the ASSOCIATION Object in the restoration LSP contains the
association ID set to the LSP ID of the protecting LSP.
The procedure for signaling the PROTECTION Object is specified in
[RFC4872]. Specifically, restoration LSP being used as a working LSP
is signaled with P bit cleared and being used as a protecting LSP is
signaled with P bit set.
Zhang, et al Expires May 24, 2015 [Page 7]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
As discussed in Section 2 of this document, [RFC6689] Section 2.2
reviews the procedure for providing LSP associations for the GMPLS
end-to-end recovery scheme using restoration LSP where the failed
working LSP and/or protecting LSP are torn down.
4. RSVP-TE Signaling For Resource Sharing During LSP Setup/Teardown
For LSP restoration upon failure, as explained in Section 11 of
[RFC4872], the purpose of using MBB is to re-use existing resources.
Thus, the behavior of the intermediate nodes during rerouting process
will not further impact traffic since it has been interrupted due to
the already broken working LSP. However, for the following two
cases, the behavior of intermediate nodes may impact the traffic
delivery: (1) LSP reversion; (2) LSP re-optimization.
Another dimension that needs separate attention is how to correlate
the two LSPs sharing resource. For the LSPs with the same Tunnel ID,
[RFC4872] and reviewed in this section. For the LSPs with different
Tunnel IDs, signaling procedure is clarified in Section 4.2 of this
document.
4.1. LSPs with Identical Tunnel ID
For resource sharing among LSPs with identical Tunnel IDs, SE flag
and ASSOCIATION Object are used together. The SE flag is to enable
resource sharing and the ASSOCIATION Object with association type
"Resource Sharing" [RFC4873] is to identify the associated LSPs.
As a first step, in order to allow resource sharing, the original LSP
setup should explicitly carry the SE flag in the SESSION_ATTRIBUTE
Object during the initial LSP setup, irrespective of the purpose of
resource sharing.
The basic signaling procedure for alternative LSP setup has been
described by the existing standards. In [RFC3209], it describes the
basic MBB signaling flow for MPLS-TE networks. [RFC4872] adds
additional information when using MBB for LSP rerouting.
As mentioned before, for LSP setup/teardown in GMPLS-controlled
circuit networks, the network elements along the path need to send
cross-connection setup/teardown commands to data plane node(s) either
during the PATH message forwarding phase or the RESV message
forwarding phase.
4.1.1. Restoration LSP Setup
For LSP restoration, the complete signaling flow processes for both
Zhang, et al Expires May 24, 2015 [Page 8]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
LSP restorations upon failure and LSP reversion upon link failure
recovery are described in this section.
Table 1: Node Behavior during Restoration LSP Setup
---------+---------------------------------------------------------
Category | Node Behavior during Restoration LSP setup
---------+---------------------------------------------------------
C1 + Reusing existing resource on both input and output
+ interfaces.
+ This type of nodes only needs to book the existing
+ resource when receiving the PATH message and no cross-
+ connection setup command is needed when receiving
+ the RESV message.
---------+----------------------------------------------------------
C2 + Reusing existing resource only on one of the interfaces,
+ either input or output interfaces and need to use new
+ resource on the other interface.
+ This type of nodes needs to book the resources on the
+ interface where new resource are needed and re-use the
+ existing resource on the other interface when it receives
+ the PATH message. Upon receiving the RESV message, it
+ needs to send the re-configuration the cross-connection
+ command to its corresponding data plane node.
---------+---------------------------------------------------------
C3 + Using new resource on both interfaces.
+ This type of nodes needs to book the new resource when
+ receiving PATH and send the cross-connection setup
+ command upon receiving RESV.
---------+---------------------------------------------------------
For LSP rerouting upon working LSP failure, using the network shown
in Figure 3 as an example.
Working LSP: A-B-C-D-E
Restoration LSP: A-B-C-F-G-E
The restoration LSP may be calculated by the head-end node or a Path
Computation Element (PCE) [RFC4655]. Assuming that the
cross-connection configuration command is sent by the control plane
nodes during the RESV forwarding phrase, the node behavior for
setting up the alternative LSP can be classified into the following
three categories as shown in Table 1.
Zhang, et al Expires May 24, 2015 [Page 9]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
+---+ +---+ +---+ +---+ +---+ +---+
| A | | B | | C | | F | | G | | E |
+-+-+ +-+-+ +-+-+ +-+-+ +-+-+ +-+-+
| | | | | |
| PATH | | | | |
C1 +----------X+ C1 | | | |
| | PATH | | | |
| +----------X+ C2 | | |
| | | PATH | | |
| | +----------X+ C3 | |
| | | | PATH | |
| | | +----------X+ C3 |
| | | | | PATH |
| | | | +-----------X+ C2
| | | | | |
| | | | | |
| | | | | RESV |
| | | | C3 +X-----------+ C2
| | | | RESV | |
| | | C3 +X----------+ |
| | | RESV | | |
| | C2 +X----------+ | |
| | RESV | | | |
| C1 +X----------+ | | |
| RESV | | | | |
C1 +X----------+ | | | |
Figure 4: Restoration LSP Setup Signaling Procedure
As shown in Figure 4, depending on whether the resource is re-used or
not, the node behaviors differ. This deviates from normal LSP setup
since some nodes do not need to re-configure the cross-connection,
and thus should not be viewed as an error. Also, the judgment
whether the control plane node needs to send a cross-connection
setup/modification command to its corresponding data plane node(s)
relies on the check whether the following two cases holds true: (1)
the PATH message received include a SE reservation style; (2) the
PATH message identifies a LSP that sharing the same tunnel ID as the
LSP to share resource with. For the second point, the processing
rules and configuration of ASSOCIATION Object defined in [RFC4872]
are followed.
4.1.2. LSP Reversion
If the LSP rerouting is revertive, traffic can be reverted to the
working or protecting LSP after its failure is recovered. From
resource sharing perspective reversion can be divided into two types:
Zhang, et al Expires May 24, 2015 [Page 10]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
o Make-while-break reversion, where resources associated with
working or protecting LSP are reconfigured while removing
reservations for restoration LSP.
o Make-before-break reversion, where resources associated with
working or protecting LSP are reconfigured before removing
restoration LSP.
It is worth mentioning that in GMPLS-controlled circuit OTN networks
both reversion types will result in a short traffic disruption.
4.1.2.1. Make-while-break Reversion
In this technique, restoration LSP is simply requested to be deleted.
Removing reservations for restoration LSP triggers reconfiguration of
resources associated with working or protecting LSP on every node
where resources are shared. Hence, whenever reservation for
restoration LSP is removed from a node, data plane configuration
changes to reflect reservations of working or protection LSP as
signaling progresses. Eventually, after the whole restoration LSP is
deleted, data plane configuration will fully match working or
protecting LSP reservations on the whole path. Thus reversion is
complete.
+---+ +---+ +---+ +---+ +---+ +---+
| A | | B | | C | | F | | G | | E |
+-+-+ +-+-+ +-+-+ +-+-+ +-+-+ +-+-+
| | | | | |
| PATHTEAR | | | | |
D1 +----------X+ D1 | | | |
| | PATHTEAR | | | |
| +----------X+ D2 | | |
| | | PATHTEAR | | |
| | +----------X+ D3 | |
| | | | PATHTEAR | |
| | | +----------X+ D3 |
| | | | | PATHTEAR |
| | | | +----------X+ D2
| | | | | |
Figure 5: Signaling Procedure for LSP Make-while-break Reversion
Figure 5 shows signaling process of make-while-break reversion of LSP
PathTear message. For alarm-free LSP deletion, the mechanisms
described in Section 6 of [RFC4208] should be followed. Resource
sharing between working and restoration LSP takes place on nodes A,
B, C and E. These are the nodes where reconfiguration of resources
associated with working LSP can take place.
Zhang, et al Expires May 24, 2015 [Page 11]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
Node behavior upon removing reservation for restoration LSP depends
on how resources are shared with working or protecting LSP:
Table 2: Node behavior during LSP make-while-break reversion
---------+---------------------------------------------------------
Category | Node behavior during LSP make-while-break reversion
---------+---------------------------------------------------------
D1 + Working and restoration LSP share resources on both
+ incoming and outgoing interface.
+
+ CP change: Reservation for restoration LSP is removed.
+ DP change: None, as data plane configuration already
+ reflects working LSP reservation.
---------+----------------------------------------------------------
D2 + Working and restoration LSP share resources on one of the
+ interfaces.
+
+ CP change: Reservation for restoration LSP is removed.
+ DP change: Resource on the interface that is not shared
+ between working and restoration LSP is freed.
+ Cross-connection is updated to reflect working LSP
+ reservation.
---------+----------------------------------------------------------
D3 + Working and restoration LSP do not share resources.
+
+ CP change: Reservation for restoration LSP is removed.
+ DP change: Resources associated with restoration LSP are
+ freed.
---------+----------------------------------------------------------
Make-while-break, while being relatively simple in its logic, has a
few limitations which may be not acceptable in some implementations:
o No rollback
Deletion of a LSP is not a revertive process. If for some
reason reconfiguration of data plane on one of the nodes to
match working or protection LSP reservations fails, falling back
to restoration LSP is no longer an option, as its state might
have already been removed from other nodes.
o No completion guarantee
Deletion of a LSP provides no guarantees of completion. In
particular, if RSVP packets are lost due to nodal or DCN
failures it is probable for a LSP to be only partially deleted.
To mitigate this, RSVP could maintain soft state reservations
Zhang, et al Expires May 24, 2015 [Page 12]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
and hence eventually remove remaining reservations due to
refresh timeouts. This approach is not feasible in circuit
networks however, since control and data channels are often
separated and hence soft state reservations are not used.
Finally, one could argue that graceful LSP deletion [RFC3473]
would provide guarantee of completion. While this is true for
most cases, many implementations will timeout graceful deletion
if LSP is not removed within certain amount of time, e.g. due to
a transit node fault. After that, deletion procedures that
provide no completion guarantees will be attempted. Hence in
corner cases completion guarantee cannot be provided.
o No explicit notification of completion to ingress node
In some cases it may be useful for ingress node to know when the
data plane has been reconfigured to match working or protection
LSP reservations. This knowledge could be used for initiating
operations like enabling alarm monitoring, power equalization
and others. Unfortunately, for the reasons mentioned above,
make-while-break reversion lacks such explicit notification.
4.1.2.2. Make-before-break Reversion
MBB reversion can be used to overcome limitations of make-while-break
reversion. It is similar in spirit to MBB concept used for
restoration. Instead of relying on deletion of restoration LSP, it
chooses to establish a new LSP to reconfigure resources on the
working or protection LSP path. Only if setup of this LSP is
successful will other LSPs be deleted. MBB reversion consists of two
parts:
A) Make part:
Creating a new reversion LSP following working or protection
LSP's path - see Figure 6. Reversion LSP is sharing resources
both with working and restoration LSPs. As reversion LSP is
created, resources are reconfigured to match its reservations -
nodes follow procedures described in Table 1. Hence after
reversion LSP is created, data plane configuration essentially
reflects working or protecting LSP reservations.
B) Break part:
After 'make' part is finished, working and restoration LSPs are
torn down. Removing reservations for working and restoration
LSPs does not cause any resource reconfiguration on reversion
LSP's path - nodes follow same procedures as for 'break' part of
any MBB operation. Hence after working and restoration LSPs are
removed, data plane configuration is exactly the same as before
Zhang, et al Expires May 24, 2015 [Page 13]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
starting restoration. Thus reversion is complete.
+---+ +---+ +---+ +---+ +---+
| A | | B | | C | | D | | E |
+-+-+ +-+-+ +-+-+ +-+-+ +-+-+
| | | | |
| PATH | | | |
C1 +----------X+ C1 | | |
| | PATH | | |
| +----------X+ C2 | |
| | | PATH | |
| | +----------X+ C1 |
| | | | PATH |
| | | +----------X+ C2
| | | | |
| | | | |
| | | | RESV |
| | | C1 +X----------+ C2
| | | RESV | |
| | C2 +X----------+ |
| | RESV | | |
| C1 +X----------+ | |
| RESV | | | |
C1 +X----------+ | | |
Figure 6: 'Make': Reversion LSP Setup follows Working LSP's Path
Figure 6 shows signaling process of reversion LSP setup for working
LSP from Section 4.1.1. In this example, resource sharing between
reversion and restoration LSP takes place on nodes A, B, C and E.
Resource sharing between working and reversion LSP takes place on
whole working LPS's path, i.e. A, B, C, D and E. Before reversion
LSP is signaled, data plane configuration on nodes A, B, C and E
match restoration LSP reservations. On node D data plane
configuration matches working LSP reservations.
As already mentioned, MBB reversion uses make-before-break
characteristics to overcome challenges related to make-while-break
reversion:
o Rollback
If 'make' part fails, restoration LSP will still be used to
carry existing traffic. Same logic applies here as for any MBB
operation failure.
o Completion guarantee
Zhang, et al Expires May 24, 2015 [Page 14]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
LSP setup is resilient against RSVP message loss, as PATH and
RESV messages are refreshed periodically. Hence, given that
network recovers its DCN eventually, setup is guaranteed to
finish with either success or failure.
o Explicit notification of completion to ingress node
Ingress knows that data plane has been reconfigured to match
working or protection LSP reservations when it receives RESV for
the reversion LSP.
4.1.3. Re-optimization LSP Setup and Reversion
For LSP re-optimization where the new LSP and old LSPs share
resource, the signaling flow for new LSP setup and old LSP teardown
is similar to those shown in Figures 4 and 5.
The issue that should be noted is the traffic will be disrupted if
the new path setup process changes the cross-connection configuration
of the nodes along the old LSP. If no traffic interruption is
desirable, it should either ensure that the old and new LSP do not
share the resource other than the source and destination nodes or use
other mechanisms. This is out the scope of this document.
Similarly, if LSP re-optimization fails and there is a need for LSP
reversion, the traffic may be disrupted when resources are shared and
cross-connections need to be reconfigured and reverted.
4.2. LSPs with Different Tunnel IDs
For two LSPs with different Tunnel IDs, the ASSOCIATION Object is
used to specify that they are sharing resource (by setting
ASSOCIATION type as "Resource Sharing" (value 2) as well as to
identify these correlated LSPs. There are two types:
(1) Sharing the common nodes, such as segment recovery, the source
and destination nodes of the segment recovery LSP is the
intermediate nodes along the working LSPs;
(2) Resource sharing is used in a generalized context (such as
multi-layer or multi-domain networks); it may result in either
sharing source nodes in common, or destination nodes in common, or
non end-points in common, if viewed from one domain's perspective.
The path computation can either be performed by the source node or
edge nodes for the path/path segment or carried out by the PCE, such
as the one explained in [PCEP-RSO]. This document does not impose
any constraint with regard to path computation.
Zhang, et al Expires May 24, 2015 [Page 15]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
[RFC4873] considers resource sharing for LSP segment recovery. The
ASSOCIATION Object usage is limited. [RFC6780] extends the usage of
ASSOCIATION Object to cover generalized resource sharing
applications. The extended ASSOCIATION Object is primarily defined
for MPLS-TP, but it can be applied in a wider scope [RFC6780]. It
can be used in the second types mentioned above. The configuration
and processing rules of extended ASSOCIATION Object defined in
[RFC6780] should be followed. The only issue that need pay attention
to is that uniqueness of LSP association for the second type should
be guaranteed when crossing the layer or domain boundary. The
mechanisms for how to ensure this are outside the scope of this
document.
Other than this, the signaling flow for this type of resource sharing
is similar to the description provided in Section 4.1.1. Similar to
what is discussed in previous sections, the traffic delivery may be
interrupted. Depending on whether the short traffic interruption is
acceptable or not, additional mechanisms may be needed and are
outside the scope of this document.
5. Security Considerations
This document reviews procedures defined in [RFC4872] and [RFC6689]
and does not define any new procedure. This document does not incur
any new security issues other than those already covered in [RFC3209]
[RFC4872] [RFC4873] and [RFC6780].
6. IANA Considerations
This informational document does not make any requests for IANA
action.
7. Acknowledgement
The authors would like to thank George Swallow for the discussions on
the GMPLS restoration.
Zhang, et al Expires May 24, 2015 [Page 16]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
8. References
8.1. Normative References
[RFC3209] D. Awduche et al, "RSVP-TE: Extensions to RSVP for LSP
Tunnels", RFC 3209, December 2001.
[RFC3473] L. Berger, Ed., "Generalized Multi-Protocol Label
Switching (GMPLS) Signaling Resource ReserVation
Protocol-Traffic Engineering (RSVP-TE) Extensions", RFC
3473, January 2003.
[RFC3945] Mannie, E., "Generalized Multi-Protocol Label Switching
(GMPLS) Architecture", RFC 3945, October 2004.
[RFC4203] Kompella, K., and Rekhter, Y., "OSPF Extensions in
Support of Generalized Multi-Protocol Label Switching
(GMPLS)", RFC 4203, October 2005.
[RFC4872] J.P. Lang et al, "RSVP-TE Extensions in Support of End-
to-End Generalized Multi-Protocol Label Switching (GMPLS)
Recovery", RFC 4872, May 2007.
[RFC4873] L. Berger et al, "GMPLS Segment Recovery", RFC 4873, May
2007.
[RFC6689] L. Berger, "Usage of the RSVP ASSOCIATION Object", RFC
6689, July 2012.
[RFC6780] L. Berger et al, "RSVP ASSOCIATION Object Extensions",
RFC 6780, October 2012.
8.2. Informative References
[PCEP-RSO] X. Zhang, et al, "Extensions to Path Computation Element
Protocol (PCEP) to Support Resource Sharing-based Path
Computation", work in progress, February 2014.
[RFC4426] Lang, J., Rajagopalan, B., and Papadimitriou, D.,
"Generalized Multiprotocol Label Switching (GMPLS)
Recovery Functional Specification", RFC 4426, March 2006.
[RFC4427] Mannie, E., and Papadimitriou, D., "Recovery (Protection
and Restoration) Terminology for Generalized Multi-
Protocol Label Switching", RFC 4427, March 2006.
Zhang, et al Expires May 24, 2015 [Page 17]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
[RFC4655] A. Farrel et al, "A Path Computation Element (PCE)-Based
Architecture", RFC 4655, August 2006.
[RFC4208] Swallow, G., Drake, J., Ishimatsu, H., Rekhter, Y.,
"Generalized Multiprotocol Label Switching (GMPLS)
User-Network Interface (UNI): Resource ReserVation
Protocol-Traffic Engineering (RSVP-TE) Support for the
Overlay Model", RFC 4208, October 2005.
Zhang, et al Expires May 24, 2015 [Page 18]
Internet-Draft GMPLS Restoration and Resource SharingNovember 20, 2014
9. Authors' Addresses
Xian Zhang
Huawei Technologies
F3-1-B R&D Center, Huawei Base
Bantian, Longgang District
Shenzhen 518129 P.R.China
Email: zhang.xian@huawei.com
Haomian Zheng (editor)
Huawei Technologies
F3-1-B R&D Center, Huawei Base
Bantian, Longgang District
Shenzhen 518129 P.R.China
Email: zhenghaomian@huawei.com
Rakesh Gandhi (editor)
Cisco Systems, Inc.
Email: rgandhi@cisco.com
Zafar Ali
Cisco Systems, Inc.
Email: zali@cisco.com
Gabriele Maria Galimberti
Cisco Systems, Inc.
Email: ggalimbe@cisco.com
Pawel Brzozowski
ADVA Optical
Email: PBrzozowski@advaoptical.com
Zhang, et al Expires May 24, 2015 [Page 19]
