No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

NE40E V800R010C10SPC500 Feature Description - Virtual Access 01

This is NE40E V800R010C10SPC500 Feature Description - Virtual Access
Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Reliability of Virtual Access

Reliability of Virtual Access

Two Control Planes

To improve system and service reliability, virtual access supports a primary and secondary control plane and provides multiple backup mechanisms.

AP Dual-homed to a Primary and Secondary Master

The control planes of nodes in a virtual access system are centralized on a master. To implement a primary and secondary control plane, an AP must be dual-homed to a primary and secondary master. You can specify a primary and secondary master for each AP. For details, see Election of Primary and Secondary Masters. After the primary and secondary masters are determined, an external communication interface on an AP has a virtual access interface on the primary and secondary masters, as shown in Figure 3-25.

Figure 3-25 AP dual-homed to a primary and secondary master

When an AP is dual-homed to a primary and secondary master, the masters separately establish a NETCONF channel to the AP and notify each other of the NETCONF channel's status (Up or Down) through the Diameter channel. In normal cases, only one master can deliver configurations to the AP at the same time. The following table lists the principles for determining the master.

Diameter Channel Status

NETCONF Channel Status on the Primary Master

NETCONF Channel Status on the Secondary Master

Master Delivering Configurations to the AP

Up

Up

Up or Down

Primary master

Down

Up

Secondary master

Down

Down

None

Down

Up

Up

Primary and secondary masters

Up

Down

Primary master

Down

Up

Secondary master

Down

Down

None

NOTE:
  • If the Diameter channel between the primary and secondary masters goes Down, the master delivering configurations to the AP performs a primary/secondary switchover based on the principles in the preceding table after a delay.

  • In the preceding table, when no master or both the primary and secondary masters deliver configurations to the AP, the virtual access system cannot work properly. You must rectify the fault.

Primary and Secondary vaPWs

When an AP is dual-homed to a primary and secondary master, the masters separately establish a vaPW to the AP and notify each other of the vaPW's status (Up or Down) through the Diameter channel. Each master determines the primary/secondary status of the local vaPW based on the principles listed in the following table.

Master's Primary/Secondary Status

Local vaPW's Status

Remote vaPW's Status

Local vaPW's Primary/Secondary Status

Primary

Up

Up or Down

Primary

Down

Up

Secondary

Down

Down

Primary

Secondary

Up

Up

Secondary

Up

Down or unknown (that is, the Diameter channel goes Down)

Primary

Down

Up or Down

Secondary

NOTE:

In the preceding table, the following situations cause service interruptions:

  • The vaPW states on both the primary and secondary masters go Down.

  • The status of the Diameter channel between the primary and secondary masters is Down, and two primary vaPWs may appear.

On the network shown in Figure 3-26, the primary/secondary status of the vaPWs is determined. When the primary and secondary vaPWs are both in the Up state, the master on which the secondary vaPW resides sets the virtual access interface to blocked to prevent loops. The master on which the primary vaPW resides delivers entries to the AP so that service traffic is imported to the primary vaPW.

Figure 3-26 Primary and secondary vaPWs
Primary and Backup vaLSPs

To carry a vaPW, each master establishes a bidirectional co-routed LSP (that is, a vaLSP) to an AP. Each vaLSP supports hot backup. The primary and backup vaLSPs form a vaTunnel, as shown in Figure 3-27.

Figure 3-27 Primary and backup vaLSPs (non-traversal LSP scenario)

When an AP is dual-homed to a primary and secondary master, both the primary and secondary masters establish the primary and backup vaLSPs to the AP. Each group of primary and backup vaLSPs form a vaTunnel. When a master calculates a path for a vaLSP, the path cannot pass through both the primary and secondary masters. On the network shown in Figure 3-28, if no backup vaLSP that passes through only the AP exists between a master and the AP, the master establishes an LSP that passes through the neighbor master and reaches the AP. This LSP is called a traversal LSP, which consists of the automatic tunnel between the master and neighbor master and the vaLSP between the neighbor master and AP. The traversal LSP provides hot backup for the primary vaLSP. The primary and secondary masters use RSVP-TE to establish an automatic tunnel.

Figure 3-28 Primary and backup vaLSPs (traversal LSP scenario)
NOTE:

If no direct link exists between the primary and secondary masters, they cannot establish an automatic tunnel. As a result, a traverse LSP cannot be established.

For details about vaPW and vaLSP establishment, see Internal Forwarding Tunnel Establishment.

Network Fault Detection

Two control planes provide multiple backup mechanisms for a virtual access system. Fast fault detection must be implemented to promptly trigger protection.

Internal Fault Detection

A virtual access system supports the following internal fault detection mechanisms:

BFD for vaLSP

A virtual access system automatically creates a BFD for vaLSP session between a master and AP to rapidly detect vaLSP faults, as shown in Figure 3-29. If the primary vaLSP fails, the master uses the BFD for vaLSP session to rapidly detect that the primary vaLSP goes Down, triggering service traffic to be switched to the backup vaLSP.

Figure 3-29 BFD for vaLSP session

BFD for vaTunnel

A vaTunnel consists of a group of primary and backup vaLSPs. A virtual access system automatically creates a BFD for vaTunnel session between a master and AP to rapidly detect vaTunnel faults, as shown in Figure 3-30. If both the primary and backup vaLSPs fail, the master uses the BFD for vaTunnel session to rapidly detect that the vaTunnel goes Down. The vaPW carried by the vaTunnel then goes Down, triggering a primary/secondary vaPW switchover. Service traffic then enters the normal vaPW.

Figure 3-30 BFD for vaTunnel session

BFD for Diameter

A virtual access system automatically creates a BFD for Diameter session between the primary and secondary masters to rapidly detect Diameter channel faults, as shown in Figure 3-31. If the Diameter channel fails, the primary and secondary masters use the BFD for Diameter session to rapidly detect that the Diameter channel goes Down, triggering a switchover to the master that delivers configurations to the AP and a vaPW switchover (for details, see, Two Control Planes).

Figure 3-31 BFD for Diameter session
NOTE:

If no direct link exists between the primary and secondary masters, the Diameter channel passes through the AP.

Detect Diameter Channel Faults Based on a DCN Route's Status

When multiple APs are dual-homed to the primary and secondary masters, the masters negotiate the primary/secondary status. When the link used by the Diameter channel is switched between the primary master's different boards, the BFD for Diameter session takes a long time for the link switchover. The BFD for Diameter session incorrectly goes Down within this time, and the secondary master considers the Diameter channel Down. As a result, the secondary master becomes the primary master. When two primary masters coexist, service traffic fails to be forwarded. To resolve this issue, configure a master to detect Diameter channel faults based on a DCN route's status. The secondary master determines the Diameter channel's status based on the DCN route's status. The DCN route goes Down only if all links between the primary and secondary masters fail. When the secondary master detects that the DCN route goes Down, the secondary master determines that the Diameter channel goes Down and becomes the primary master.

Dynamic multicast BFD monitoring an internal communication interface

In a virtual access system, manually configure a dynamic multicast BFD session on a master to rapidly detect connectivity faults on all internal communication interfaces, as shown in Figure 3-32. After the configuration, the BFD module automatically generates a dynamic multicast BFD session for each internal communication interface managed by the master, and associates the session with the interface's port state table (PST) and process interface status (PIS). If an internal communication interface or its link fails, the BFD module instructs the master to modify the interface's status in the PST, triggering fast DCN route reconvergence in the virtual access system. This affects the establishment of an internal forwarding tunnel and control channel. In addition, the BFD module immediately sends a Down message to the interface so that the interface enters the BFD Down state, which is equal to the Down state of the link protocol. In the BFD Down state, only BFD packets can be properly processed.

Figure 3-32 Deploying a multicast BFD session to monitor an internal communication interface
NOTE:

As shown in Figure 3-32, if multiple physical links exist between a master and AP, the system generates a virtual link between them and adds all physical links to the virtual link. The system generates a virtual interface on both ends of each virtual link. The virtual interface is used as an internal communication interface. The dynamic multicast BFD session goes Down only if all the physical links between virtual interfaces fail.

External Fault Detection

External fault detection enables a virtual access system to promptly detect faults on the connection to an external network and to associate an internal tunnel's status changes. A virtual access system supports the following external fault detection mechanisms:

BFD monitoring the link between a CE and AP

You can configure multicast BFD on the AP and user-side CE to rapidly detect faults on the link between the AP and CE, as shown in Figure 3-33. A multicast BFD session is bound to a physical interface on the CE, which is the same as that in a common scenario. To bind a multicast BFD session to an external communication interface on the AP, bind the multicast BFD session to the virtual access interface for the external communication interface on a master. If the link between the AP and CE fails, the multicast BFD session triggers the CE's interface, the AP's external communication interface, and the primary and secondary masters' virtual access interfaces to go Down. Both the primary and secondary vaPWs then fail. In this scenario, you must deploy an additional protection link on the user-side network to ensure that service traffic is not interrupted.

Figure 3-33 Deploying a multicast BFD session to monitor the link between the CE and AP

Internal communication interface associated with an interface monitoring group

In a virtual access scenario, if a fault occurs on the network side, the reliability mechanism deployed on the network side handles the fault. The virtual access system does not need to detect the fault. However, on the network shown in Figure 3-34, if the links between the primary master and network-side PEs all fail, downstream traffic from the network side to the user side has no available path to the primary master. The reliability mechanism on the network side then triggers downstream traffic to be switched to the path to the secondary master. If the virtual access system does not detect the fault, upstream traffic from the user side to the network side is still forwarded through the primary master. As a result, upstream traffic is lost and in turn services are interrupted. To resolve this issue, bind all the network-side interfaces on the primary master to an interface monitoring group, and associate an internal communication interface with the interface monitoring group. If the links between the primary master and network-side PEs all fail, the interface monitoring group triggers the primary master's interface communication interface to go Down. Upstream traffic is then switched to the secondary master, ensuring the normal forwarding of services.

Figure 3-34 Scenario where the links between the primary master and network-side PEs all fail

Protection Switching Mechanism

If a fault occurs on a network, the virtual access system uses a fast detection mechanism to detect the fault and trigger protection switching, ensuring service reliability. Typical protection switching scenarios include:

Scenario 1: The link between a master and AP fails.

If the link between Master1 and the AP fails, traffic is switched at the vaLSP layer and forwarded along the path CE1-AP-Master2-Master1-PE1-CE2, as shown in Figure 3-35. Table 3-10 describes control-plane switching in the virtual access system.

Figure 3-35 Scenario 1: The link between a master and AP fails.
Table 3-10 Control-plane switchover

Master's Primary/Secondary Status

Master Delivering Configurations to the AP

vaPW's Primary/Secondary Status

Master1 and Master2 are primary and secondary masters, respectively.

  • When the NETCONF channel between the AP and Master1 is Up, Master1 delivers configurations to the AP.

  • When the NETCONF channel between the AP and Master1 is Down, Master2 delivers configurations to the AP.

The vaPW between the AP and Master1 is in the primary state, and the vaPW between the AP and Master2 is in the secondary state.

NOTE:
  • If the link between the AP and Maste1 fails, the IS-IS topology of Master1 in the virtual access system is deleted. Master1 becomes the secondary Master, and Master2 becomes the primary master. The master delivering configurations becomes Master2, and the vaPW between the AP and Master2 switches to the primary state. Traffic is then switched to Master2.

  • When multiple physical links exist between the AP and Master1, the preceding protection switching is triggered only if all the links fail.

  • If the link between the AP and Master2 fails, only the establishment of the internal forwarding tunnel and control channel between the AP and Master2 is affected. No switchover occurs on the forwarding and control planes of the virtual access system.

Scenario 2: A master fails.

If Master1 fails, traffic is switched at the vaPW layer and forwarded along the path CE1-AP-Master2-PE1-CE2, as shown in Figure 3-36. Table 3-11 describes control-plane switching in the virtual access system.

Figure 3-36 Scenario 2: A master fails.
Table 3-11 Control-plane switchover

Master's Primary/Secondary Status

Master Delivering Configurations to the AP

vaPW's Primary/Secondary Status

Master1 and Master2 are secondary and primary masters, respectively.

Master2

The vaPW between the AP and Master1 is in the secondary state, and the vaPW between the AP and Master2 is in the primary state.

NOTE:
  • If Master1 fails, a primary/secondary switchover on the management plane is not immediately triggered. A primary/secondary switchover on the management plane is triggered only when the IS-IS topology of Master1 in the virtual access system is deleted.

  • If Master2 fails, only the establishment of the internal forwarding tunnel and control channel between the AP and Master2 is affected. No switchover occurs on the forwarding and control planes of the virtual access system.

Scenario 3: The link between masters fails.

On the network shown in Figure 3-37, if the link between Master1 and Master2 fails, no switchover occurs on the forwarding and control planes of the virtual access system. The fault has the following impacts on the system:

  • The Diameter channel between Master1 Master2 passes through the AP.

  • An automatic tunnel cannot be established between Master1 and Master2. A backup vaLSP that traverses Master 2 cannot be established for the primary vaLSP between the AP and Master1 (for details about a traversal LSP's establishment, see Primary and Backup vaLSPs).

Figure 3-37 Scenario 3: The link between masters fails.
Scenario 4: The link between a master and PE fails.

On the network shown in Figure 3-38, if the links between Master1 and PE1 and between Master 1 and PE2 all fail, an interface monitoring group can be associated to trigger Master1's internal communication interface to go Down (this scenario is similar to that where Master 1 fails). Traffic is then switched at the vaPW layer. Table 3-12 describes control-plane switching in the virtual access system.

Figure 3-38 Scenario 4: The link between a master and PE fails.
Table 3-12 Control-plane switchover

Master's Primary/Secondary Status

Master Delivering Configurations to the AP

vaPW's Primary/Secondary Status

Master1 and Master2 are secondary and primary masters, respectively.

Master2

The vaPW between the AP and Master1 is in the secondary state, and the vaPW between the AP and Master2 is in the primary state.

NOTE:

If other faults occur on the network side, the reliability mechanism deployed on the network side provides protection switching, which is beyond the scope of this section.

Translation
Download
Updated: 2019-01-03

Document ID: EDOC1100055053

Views: 1601

Downloads: 28

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next