Reliability
General Reliability Deployment Solution
Layer | Reliability Technique | Usage Scenario |
---|---|---|
Physical and link layers | Link aggregation (Eth-Trunk), Link Aggregation Control Protocol (LACP), dual-homing, ring, MSTP/RSTP/STP | Any type of Ethernet links are used. |
Network layer | IGP fast convergence (FC), directional forwarding detection (BFD) for IGP, and IGP graceful restart (GR) | OSPF or IS-IS is used. |
BGP indirect next hop, BFD for BGP, and BGP GR | BGP is used. | |
MPLS LDP fast reroute (FRR), IGP-LDP synchronization, and BFD for LDP | MPLS LDP is used. | |
MPLS TE FRR, TE hot standby, and BFD for TE | MPLS TE tunnels are used. | |
Service bearer layer | mVRRP, BFD for VRRP, and ARP multi-device hot backup | A CE is dual-homed to PEs. |
PW redundancy, VLL FRR and BFD for PW | A CE is dual-homed to PEs, and an L2VPN is used to transmit services. | |
VPN FRR and IP FRR | A CE is dual-homed to PEs, and an L3VPN is used to transmit services. | |
VLL GR and VPLS GR | An L2VPN is used to transmit services. | |
VPN GR | An L3VPN is used to transmit services. |
L3VPN Service Reliability
- PE-to-CE services: protected using VRRP or route convergence.
- PE-to-PE services: protected using LSP reliability and VPN FRR.
Configuration Item | Description |
---|---|
PE-to-CE service reliability | One or more PE-to-CE service reliability functions
can be configured as needed:
|
PE-to-PE service reliability | LSP reliability
BFD for VPN FRR
|
Reliability Functions for the E-Line/VLL Service
Select reliability techniques for the E-Line/VLL service as needed:
- 3PE scenario: Using the PW redundancy protection technique is recommended, as shown in Figure 7-5.
- 4PE scenario: Using the PW redundancy protection technique is recommended, as shown in Figure 7-6.
Configuration Item | Description |
---|---|
Eth-Trunk interface |
|
E-Trunk |
|
BFD for E-Trunk |
|
Configure service PWs. |
|
Intelligent combiner bypass (ICB) PW (also called bypass PW) | A bypass PW is established between UPE2 and UPE3. The bypass
PW configuration is similar to the primary PW configuration. The configuration
includes an LDP session, an LSP, and a VC on an AC interface on UPE2
and UPE3. The difference is that different command lines are used
to configure VCs:
|
management PW (mPW) |
|
Bind each service PW to an mPW. |
|
BFD for PW |
|
Configuration Item | Description |
---|---|
Eth-Trunk interface | Configure an Eth-Trunk interface. The configuration is the same as the configurations on UPE2 and UPE3 in Figure 7-5. For configuration details, see Table 7-3. |
BFD for AC |
The BFD for AC configuration of UPE1 is similar to BFD for AC configurations of UPE3 and UPE4. The configuration procedure is not provided. NOTE:
BFD is also configured on CE1 and CE2 to monitor
PE-to-CE links. For configuration details, see the documents of CE
devices. |
service PWs and ICB PWs (also called bypass PWs) |
|
management VRRP (mVRRP) | mVRRP is configured to determine the primary/secondary PW status when the independent PW redundancy mode is used. In the following example, mVRRP is configured on UPE1 and UPE2. The mVRRP configuration on UPE3 is similar to that on UPE1, and that on UPE4 is similar to that on UPE2. The configuration procedure is not provided.
|
Bind each service PW to the mVRRP backup group | In the following example, bind a service PW to an mVRRP backup group on UPE1. The configurations on UPE2, UPE3, and UPE4 are similar to the configuration on UPE1. The configuration procedure is not provided. # interface eth-trunk 1.1 //Enter the view of an AC interface bound to a PW. vlan-type dot1q 100 mpls l2vc track admin-vrrp interface eth-trunk1.2 vrid 1 pw-redundancy mpls l2vc track admin-vrrp secondary |
Reliability Functions for the E-VLAN (VPLS) Service
The following reliability functions can be used to prevent loops for VPLS services:
- Split horizon that VPLS itself supports
- E-Trunk, E-VRRP, or MSTP that prevents loops at the access layer
- Label switched path (LSP) protection functions that protect traffic within a VPLS network
Configuration Item | Description |
---|---|
PE-CE service reliability | One or more PE-to-CE service reliability functions
can be configured as needed:
For configuration details about the Eth-Trunk interface, an E-Trunk interface, and BFD for E-Trunk, see Table 7-3. In the following example, E-VRRP is configured on UPE2. # efm enable //Enable EFM OAM globally. # interface eth-trunk 1.1 //Enter the AC interface view. vlan-type dot1q 100 l2 binding vsi a2 vrrp vrid 1 virtual-ip 192.168.1.254 //Assign a VRRP virtual IP address. It must be on the same network segment as that of the CE. vrrp vrid 1 priority 120 //Set a VRRP priority value so that UPE2 functions as the VRRP master. The default priority value is 100. The greater the value, the higher the priority. vrrp vrid 1 track efm interface eth-trunk 1 //Associate the VRRP backup group with an EFM session. efm enable //Enable EFM OAM on the AC interface. efm error-frame notification enable //(Optional) Enable errored frame detection on the interface. efm error-code notification enable //(Optional) Enable errored code detection on the interface. efm error-frame-second notification enable //(Optional) Enable errored frame second detection on the interface. NOTE:
The configuration of UPE3 is similar to that of UPE2. The difference is that the VRRP priority value on UPE3 is smaller than that on UPE2. For MSTP configuration guide, seeMSTP Configuration. |
PW reliability | Configure LSP protection functions to improve reliability within a VPLS network. For configuration details, see Table 7-5. |
The following reliability functions can be used to prevent loops between PWs on an HVPLS network:
- A VPLS PW can be used to block one PW when PW redundancy is deployed to establish PWs between a UPE and an SPE.
- When the PW status changes on a UPE, the SPE is configured to send a MAC Withdraw message to delete a MAC address entry.
Configuration Item | Description |
---|---|
Eth-Trunk, E-Trunk and BFD for E-Trunk | Configure NPE1 and NPE2. If NE20Es are used as NPEs,the configuration on NPE1 is similar to that on UPE2, and that on NPE2 is similar to that on UPE3, in Figure 7-5. For configuration details, see Table 7-3. |
PW redundancy | # vsi vsi1 static pwsignal ldp vsi-id 1 peer 2.2.2.2 //Configure a PW to SPE1. peer 3.3.3.3 //Configure a PW to SPE2. protect-group vsi1 //Create a redundancy PW protection group. protect-mode pw-redundancy master //Enable the master/slave PW redundancy mode. reroute delay 60 //Set the switchback delay time to 60s. On a large network, a delayed switchback helps prevent packet loss during a switchback due to asynchronous network convergence. peer 2.2.2.2 preference 1 //Set a PW priority value. The smaller the value, the higher the priority. The primary PW has the highest priority. peer 3.3.3.3 preference 2 # |
BFD for PW | Configure a UPE. # bfd vplspw1 bind pw vsi vsi1 peer 2.2.2.2 remote-peer 2.2.2.2 pw-ttl auto-calculate //Configure BFD to monitor the primary PW. discriminator local 100 //Set the local discriminator. discriminator remote 101 //Set the remote discriminator. # Configure SPE1. # bfd vplspw1 bind pw vsi vsi1 peer 1.1.1.1 remote-peer 1.1.1.1 pw-ttl auto-calculate //Configure BFD to monitor the primary PW. discriminator local 101 //Set the local discriminator. discriminator remote 100 //Set the remote discriminator. # |