VPLS PW Redundancy
Background
A redundant provider edge (PE) is often deployed to enhance service reliability. If a virtual private wire service (VPWS) or virtual private LAN service (VPLS) network uses a redundant PE, two pseudo wires (PWs) have to be deployed for service protection. This mechanism is called PW redundancy.
PW redundancy is a technique so widely used that it has developed into a reliability standard. PW redundancy improves service switchover efficiency and minimizes impact of device faults on services.
PW redundancy is best suited for point-to-point services, such as VPWS. VPLS, a point-to-multipoint service, can be viewed as a collection of point-to-point services. Therefore, VPLS can also use PW redundancy.
In VPLS, PW redundancy can speed up VPLS network convergence to reduce service downtime.
Related Concepts
Some key concepts for VPLS PW redundancy are described by using service traffic protection between CE1 and CE2 on the VPLS network shown in Figure 1 as an example.
Currently, VPLS PW redundancy can operate in either of the following modes (the operating mode is specified on PE1):
- Master/slave mode: PE1 determines whether a local PW is in the active or standby state based on the PW's preset forwarding priority.
- Independent mode: PE1 determines whether a local PW is in the active or standby state based on the master/backup status of PE2 and PE5.
The endpoint PEs of a PW protection group must negotiate PW status to ensure that they select the same PW to transmit packets. The following concepts are introduced for PW status negotiation inside a PW protection group:
Primary/secondary: used to describe the forwarding priority of a PW and can be configured. A smaller value indicates a higher priority. A PW with the higher forwarding priority is the primary PW.
The forwarding priorities take effect only if PE1 uses the master/slave mode as the PW redundancy mode. In master/slave mode, PE1 instructs PE2 and PE5 to change the forwarding status of PWs to be the same as that on PE1. In independent mode, the master/backup status of PE2 and PE5 determines the forwarding status of PWs on PE1.
Active/standby: used to describe the forwarding status of a PW and cannot be configured. Only the active PW can be used to forward traffic. The standby PW may be used to receive traffic.
In some documents, Huawei uses active/inactive or primary/backup to describe PW status. These terms have the same meaning as term active/standby defined in draft. They all indicate the PW forwarding status.
Implementation
To keep the original forwarding behavior, the endpoint PEs of a PW protection group must use the same PW to transmit traffic and ensure that only the PW used to transmit traffic is in the active state. To achieve this goal, the PW protection group must use a signaling mechanism.
Relevant standards specify the PW Status TLV to transmit the PW forwarding status. The PW Status TLV, a 32-bit status code field, is carried in a Label Mapping or LDP Notification message. PW redundancy uses a new PW status code, 0x00000020, to indicate that a PW is in the standby state.
Only PWE3 VPLS supports PW redundancy.
The forwarding priorities must be specified for PWs in a PW protection group. The PW with the higher forwarding priority is preferentially selected as the active PW to forward traffic. The remaining PW stays in the standby state to protect the primary PW.
The forwarding status of a PW determines whether the PW is used to forward traffic. The PW forwarding status depends on:
- Local and remote PW signaling status: A PE monitors its local signaling status and uses PW redundancy signaling to obtain the remote signaling status from a remote PE.
- PW redundancy mode: The PW redundancy mode, which can be master/slave or independent, is specified on PE1.
- PW forwarding priority: The PW forwarding priority, which determines whether a PW is the primary or secondary PW, is specified on PE1.
On the network shown in Figure 1, VPLS PW redundancy is configured on PE1. In normal cases, the local and remote PW signaling status on PE1 are both Up. The following describes how the endpoint PEs of a PW protection group choose the same PW to transmit traffic based on PW redundancy modes:
In master/slave mode, PE1 determines the local PW forwarding status based on preset forwarding priorities and informs PE2 and PE5 of the PW forwarding status; PE2 and PE5 determine their PW forwarding status based on the received PW forwarding status.
In independent mode, PE1 determines the local PW forwarding status based on the PW forwarding status learned from PE2 and PE5; PE2 and PE5 determine their PW forwarding status based on signaling, which can be enhanced trunk (E-Trunk) or Virtual Router Redundancy Protocol (VRRP) signaling, and notify PE1 of the forwarding status.
In master/slave or independent mode, if the primary PW becomes faulty, it becomes inactive and the corresponding secondary PW becomes active. PW-side faults do not affect the AC status. In independent mode, if an AC-side fault occurs, for example, a PE or AC is faulty, the PW forwarding status will change, because the status is determined by the master/backup status of the dual-homing devices. In master/slave mode, if an AC-side fault occurs, the PW forwarding status will not change, because the status is determined by preset forwarding priorities.
VPLS PW redundancy is similar to VPWS PW redundancy. The difference lies in that a virtual switching instance (VSI) has multiple PWs destined for different PEs. These PWs may form multiple PW protection groups. The forwarding status changes of PWs in one group do not affect the forwarding status of PWs in other groups.
Currently, the VPLS PW redundancy mode can only be master/slave.
Derivative Function
In addition to providing real-time service protection against network faults, VPLS PW redundancy also allows you to manually switch traffic between PWs in a PW protection group during network operation and maintenance. For example, if you want to maintain a device that serves as the endpoint of a primary PW, you can switch traffic to the secondary PW before maintenance and switch traffic back to the primary PW after maintenance.
Usage Scenario
VPLS PW redundancy can be used on hierarchical virtual private LAN service (HVPLS) networks and virtual leased line (VLL) accessing VPLS networks. These two types of networks can bear many types of services. It is recommended that you determine which types of services to deploy based on the networking characteristics of the two types of networks if they are newly planned or deployed:
- HVPLS networks are best suited for multicast services, such as Internet Protocol television (IPTV) services, because HVPLS networks can save VPLS core network bandwidth. For more information, see VPLS PW Redundancy for Protecting Multicast Services.
- VLL accessing VPLS networks are best suited for unicast services, such as high-speed internet (HSI) and Voice over Internet Protocol (VoIP) services, because VLL PEs do not need to learn user MAC addresses. For more information, see VPLS PW Redundancy for Protecting Unicast Services.
VPLS PW redundancy can also be used to improve the reliability of existing networks. On the VPLS network shown in Figure 1, CE1 communicates with CE2, CE3, and CE4 over PWs between PE1 and PE2, PE3, and PE4. As services develop, services between CE1 and CE2 and between CE1 and CE3 raise higher reliability requirements. To meet the reliability requirements, PE5 and PE6 are deployed on the VPLS network to provide VPLS PW redundancy protection for PE2 and PE3, respectively. In addition, multiple PW protection groups are configured on PE1. As a result, VPLS PW redundancy protects services against network-side, AC-side, and PE failures without affecting existing services, improving network reliability.
VPLS PW redundancy can be configured for desired services without affecting services on other PWs, reducing costs and maximizing profits.
