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

NE20E-S V800R010C10SPC500 Configuration Guide - MPLS 01

This is NE20E-S V800R010C10SPC500 Configuration Guide - MPLS
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).
Licensing Requirements and Limitations for MPLS TE

Licensing Requirements and Limitations for MPLS TE

Restrictions and Guidelines

Restrictions

Guidelines

Impact

A loopback interface address must be used in the mpls lsr-id command.

None

If an LSR ID is not set to a loopback address, an IGP floods the LSR ID and a path is successfully computed. After a TE tunnel is established, application nor protocol packets that depend on the loopback address cannot be handled. For example, in VPN iteration, the peer BGP address is used as a next-hop IP address (which should have been a loopback interface address), and the local end is iterated to the tunnel destined for the peer BGP address. Since the loopback interface address is not used, the iteration fails.

The destination IP address of the tunnel must be set to the MPLS LSR ID configured on the destination router.

None

If an LSR ID is not set to a loopback address, a path is successfully computed. After a TE tunnel is established, application nor protocol packets that depend on the loopback address cannot be handled. For example, in VPN iteration, the peer BGP address is used as a next-hop IP address (which should have been a loopback interface address), and the local end is iterated to the tunnel destined for the peer BGP address. Since the loopback interface address is not used, the iteration fails.

If a TE tunnel group and dynamic load balancing (load-balance dynamic-adjust enable) are configured, and dynamic load-balancing does not take effect.

None

If a TE tunnel group is configured, the dynamic load balancing function does not take effect.

The TE group feature requires that the inbound interface reside on a NSP-50/NSP-50-E/NSP-120/NSP-120-E/NSP-A/NSP-B/NSP-C/NSP-D board.

None

Non-NSP-50/NSP-50-E/NSP-120/NSP-120-E/NSP-A/NSP-B/NSP-C/NSP-D boards do not support TE tunnel groups.

After re-marking is performed in MF traffic classification, packets with the same priority are added to the same TE group.

After re-marking is performed in MF traffic classification, packets with the same priority are added to the same TE group.

In a TE tunnel group scenario, after CAR and re-marking are performed in MF traffic classification, packets may be out of order.

In a TE tunnel group scenario, if TE tunnels work with physical interfaces or LDP LSPs to balance traffic, although the bandwidth or weight is configured for tunnels of each priority, unequal load balancing based on the bandwidth or weight is not supported. Equal cost load balancing is performed among tunnels of each priority.

None

In a TE tunnel group scenario, if TE tunnels work with physical interfaces or LDP LSPs to balance traffic, although the bandwidth or weight is configured for tunnels of each priority, unequal load balancing based on the bandwidth or weight is not supported. Equal cost load balancing is performed among tunnels of each priority.

P2MP TE does not support the following outbound interfaces:serial interface, MP-group interface, Eth-Trunk dot1q sub-interface for VLAN tag termination, Eth-Trunk QinQ sub-interface for VLAN tag termination, Trunk-Serial interfaces, Global-MP-Group, and POS-Trunk interface.

Run the mpls te p2mp-te disable command on the interface to prevent MPLS traffic from traveling through it.

Traffic over a tunnel fails to be forwarded through the outbound interface.

An inter-IGP area P2MP TE tunnel is no supported.

Plan all P2MP TE tunnels within the same IGP area.

An S2L cannot be created on a leaf node that is in a different IGP area than the ingress of a P2MP TE tunnel.

When a Huawei forwarder runs PCEP to communicate with a non-Huawei controller, the non-Huawei controller may deliver an initiated LSP name carrying special characters, such as a question mark (?), a colon (;), or %, to the forwarder. The forwarder, however, does not support such characters. When a Huawei forwarder runs PCEP to communicate with a Huawei controller, this problem does not occur.

Do not specify an initiated LSP name carrying the special characters.

The forwarder cannot execute a parameter carrying the special characters.

PCEP is used only to connect to a Huawei controller. PCEP and NETCONF cannot be used together to modify a stitching label.

If both PCEP and NETCONF are used to modify a stitching label due to mal-operation, on the forwarder, run the undo mpls te pce delegate command to cancel tunnel delegation and the mpls te pce delegate command to delegate the tunnel to the controller again.

The Huawei controller incorrectly remains the stitching label.

Entropy label feature:

RSVP P2P LSPs and LDP P2P LSPs support entropy label negotiation. Other types of LSPs do not support this function.

None

None

A neighbor node supports GR. If a local node functions as the GR Helper, the mpls rsvp-te hello command must be run on an RSVP-TE tunnel after the mpls rsvp-te hello support-peer-gr command is run in the MPLS view.

None

If the RSVP-TE Hello capability is not configured on the RSVP-TE tunnel, the mpls rsvp-te hello support-peer-gr command functioNonelity fails. When the RSVP node performs a master/slave main control board swithocver, the RSVP adjacency between the local ndoe and its neighbor node is torn down because of the sigNoneling protocol timeout. As a result, a CR-LSP is torn down, and services transmitted over teh CR-LSP are interrupted.

When multiple MPLS TE-enabled IGP processes are deployed on a device, CSPF cannot compute an optimal path among these IGP processes. An optimal path computed within a single process may be different than the expected one. As a result, services are congested or interrupted on a TE tunnel.

  • Solution: An explicit path is configured in the system view and used in the TE tunnel interface view.
  • AlterNonetive solution: The mpls te cspf preferred-igp command is run in the MPLS view to enable CSPF to compute a path within a selected IGP process.

TE tunnels and multiple IGP processes are configured on the ingress and egress. The paths of the TE tunnels are automatically computed based on IGP paths. No explicit paths or the other constraints are configured. As a result, the TE tunnels pass through paths different than the expected ones, causing service congestion or interruptions.

In an LDP over TE scenario, after the mpls te igp advertise or mpls te igp advertise shortcut command is run in the view of a TE tunnel interface, a remote LDP session must be established based on the tunnel destination address after the route advertisement function is enabled.

In the preceding scenario, a remote LDP session must be established based on the tunnel destiNonetion.

If no remote LDP session is established, no LDP LSP is established, which interrupts services.

If a TE tunnel passes through multiple IGP areas, an explicit path with inter-node IP addresses included needs to be configured for the tunnel, and CSPF must be configured on the ingress and inter-area nodes.

Plan all TE tunnels within the same IGP area.

A tunnel fails to be established or forward data packets.

When you configure a split tunnel, not that:

A split tunnel cannot be used as an outbound interface of a route. The mpls te igp mpls te igp shortcut and mpls te reserved-for-binding commands cannot be run simultaneously for a split tunnel.

None

None

If the mpls te timer auto-bandwidth command must be run in the MPLS view when automatic bandwidth adjustment is configured on a tunnel.

None

Automatic bandwidth adjustment does not take effect.

The include and exclude attributes cannot be simultaneously configured for an affinity on a tunnel.

None

The configuration fails.

When the bidirectional attribute is enabled for a tunnel, the tunnel cannot be bound to a static bidirectional co-routed CRLSP on the egress.

None

The configuration fails.

The bidirectional co-routed attribute is mutually exclusive with the dynamic reverse LSP attribute.

None

The configuration fails.

A primary tunnel cannot be configured as a protection tunnel. A primary tunnel cannot configure itself as a protection tunnel.

None

The configuration fails.

The bandwidth configured in the bypass-attribute command must be lower than or equal to the bandwidth of the primary tunnel. The priority configured in this command cannot be greater than the priority of the primary tunnel.

None

The configuration fails.

The FF style is mutually exclusive with FRR.

The FF style is mutually exclusive with backup.

The FF style is mutually exclusive with automatic bandwidth adjustment.

The FF style is mutually exclusive with soft preemption.

None

The configuration fails.

The FF style is mutually exclusive with PCE delegation.

None

The configuration fails.

Bit error-triggered switching is mutually exclusive with the protection group function.

None

The configuration fails.

Hot standby is mutually exclusive with ordinary backup.

None

The latest configuration overrides the previous one.

Hot standby is mutually exclusive with the offload function.

Hot standby is mutually exclusive with the bypass tunnel function.

Hot standby is mutually exclusive with a protected interface.

Hot standby is mutually exclusive with the FF style.

Hot standby is mutually exclusive with the reroute disable command.

None

The configuration fails.

Ordinary backup is mutually exclusive with the off-load function.

Ordinary backup is mutually exclusive with the best-effort path function.

Ordinary backup is mutually exclusive with the FF style.

Ordinary backup is mutually exclusive with the reroute disable command.

Ordinary backup is mutually exclusive with reverse LSPs.

None

The configuration fails.

Fast reroute is mutually exclusive with the offload function.

Fast reroute is mutually exclusive with a bypass tunnel.

Fast reroute is mutually exclusive with a protected interface.

Fast reroute is mutually exclusive with FF style.

Fast reroute is mutually exclusive with reverse LSPs.

Fast reroute is mutually exclusive with the reroute disable command.

None

The configuration fails.

A bypass tunnel is mutually exclusive with FRR.

A bypass tunnel is mutually exclusive with the detour LSP function.

A bypass tunnel is mutually exclusive with backup.

A bypass tunnel is mutually exclusive with the reroute disable command.

None

The configuration fails.

A protected interface is mutually exclusive with FRR.

A protected interface is mutually exclusive with the detour LSP function.

A protected interface is mutually exclusive with backup.

A protected interface is mutually exclusive with ordinary backup and best-effort functions.

None

The configuration fails.

The best-effort path function is mutually exclusive with a bypass tunnel.

The best-effort path function is mutually exclusive with a protected interface.

The best-effort path function is mutually exclusive with the FF style.

The best-effort path function is mutually exclusive with the offload function.

The best-effort path function is mutually exclusive with ordinary backup.

The best-effort path function is mutually exclusive with the reroute disable command.

None

The configuration fails.

Soft preemption is mutually exclusive with the FF style.

Soft preemption is mutually exclusive with the reroute disable command.

None

The configuration fails.

The bypass attribute is mutually exclusive with the detour LSP function.

None

The configuration fails.

The include-all and include-any attributes cannot be simultaneously configured for an affinity. The latest configuration overrides the previous one.

None

A later configuration takes effect.

When the reverse LSP function is used, the global label allocation mode must be set to the non-null mode.

Modify the global label allocation mode to the non-null mode

Reverse LSPs do not take effect.

The reverse LSP function is mutually exclusive with backup.

The reverse LSP function is mutually exclusive with FRR.

None

The configuration fails.

The reverse LSP function is mutually exclusive with the detour LSP function.

None

The configuration fails.

A label and an IP address cannot be configured in a mixed way for an explicit path.

None

The configuration fails.

PCE delegation is mutually exclusive with the detour LSP function.

PCE delegation is mutually exclusive with the reroute disable command.

None

The configuration fails.

The detour LSP function is mutually exclusive with the offload function.

The detour LSP function is mutually exclusive with a bypass tunnel.

The detour LSP function is mutually exclusive with a protected interface.

The detour LSP function is mutually exclusive with the FF style.

The detour LSP function is mutually exclusive with the bypass attribute.

The detour LSP function is mutually exclusive with reverse LSPs.The detour LSP function is mutually exclusive with the bypass attribute.The detour LSP function is mutually exclusive with the reroute disable command.The detour LSP function is mutually exclusive with PCE delegation.

None

The configuration fails.

The reroute disable command is mutually exclusive with PCE delegation.

The reroute disable command is mutually exclusive with tunnel re-optimization.

The reroute disable command is mutually exclusive with soft preemption.

The reroute disable command is mutually exclusive with the detour LSP function.

The reroute disable command is mutually exclusive with ordinary backup.

The reroute disable command is mutually exclusive with the best-effort path.

The reroute disable command is mutually exclusive with FRR.

The reroute disable command is mutually exclusive with backup.

The reroute disable command is mutually exclusive with a bypass tunnel.

The reroute disable command is mutually exclusive with automatic bandwidth adjustment.

None

The configuration fails.

When a GMPLS tunnel is bound to a GMPLS-UNI interface and the switch-type is EVPL, the process-pst parameter cannot be configured.

None

The configuration fails.

When BFD monitors a GMPLS-UNI, the process-pst parameter is configured, and the GMPLS tunnel is bound to the GMPLS-UNI, the switch-type parameter cannot be set to EVPL.

None

The configuration fails.

A static tunnel or a static CR tunnel cannot be bound to a reverse RSVP-TE LSP.

None

The configuration fails.

LSP loopback is supported only on a transit node of a static bidirectional co-routed LSP.

None

The configuration fails.

The name of a reverse LSP in a static bidirectional co-routed tunnel must be the same as that of a forward LSP.

None

The configuration fails.

If the TP OAM lock function is enabled on an interface, the interface cannot be used as the outbound or inbound interface of a static tunnel.

None

The configuration fails.

BFD and MPLS OAM cannot be configured simultaneously for a tunnel.

None

The configuration fails.

The bandwidth of the primary tunnel cannot be greater than that of a backup tunnel.

None

The configuration fails.

After the reverse LSP in a working tunnel must be different from that in a protection tunnel.

None

The configuration fails.

A dynamic tunnel cannot be bound to a reverse LSP of a static protocol.

None

The configuration fails.

The hard pipe is mutually exclusive with bandwidth attributes.

The hard pipe attribute is mutually exclusive with the diffserv-mode command.

An inbound interface must be configured for a hard pipe LSP.

An outbound interface must be configured for a hard pipe LSP.

None

The configuration fails.

An inter-area auto bypass tunnel is not supported because LSP cannot be bound to an FRR path in an inter-area scenario.

Do not use Auto FRR in an inter-area scenario.

A cross-IGP area node cannot create an Auto bypass tunnel to protect nodes or links on a primary tunnel.

An inter-area detour LSP is not supported.

Do not use Detour FRR in an inter-area scenario. The primary LSP cannot be bound to a Detour FRR LSP in the inter-area scenarios.

A cross-IGP area node cannot create a detour LSP to protect nodes or links on a primary tunnel.

Inter-IGP-area re-optimization is not supported.

None

If an LSP is established across IGP areas, the LSP cannot be re-optimized.

An Segment Routing MPLS tunnel must be used together with detection function. Segment Routing MPLS tunnels do not use signaling protocols. BFD for TE LSP must be configured to monitor the connectivity of the primary and backup LSPs before HSB protection is performed.

An Segment Routing MPLS tunnel must be used together with detection function.

Without a detection protocol enabled, Segment Routing MPLS LSPs cannot detect link faults or switch traffic, resulting in traffic loss.

An Segment Routing MPLS tunnel must be used together with detection function. Segment Routing MPLS tunnels do not use signaling protocols. BFD for tunnel must be configured to associate the BFD status with the tunnel status.

An Segment Routing MPLS tunnel must be used together with detection function.

Without a detection protocol enabled, Segment Routing MPLS LSPs cannot detect link faults or notify an App of the tunnel fault. As a result, services iterated to an Segment Routing MPLS LSP cannot be switched to a protection LSP, resulting in traffic loss.

After a labeled explicit path is configured for an Segment Routing MPLS tunnel, there is no response to label stack packets delivered by the schema or PCEP path information messages.

None

An Segment Routing MPLS tunnel established over a labeled explicit path does not respond to messages of the other label stacks or to the make-before-break mechanism. Path re-optimization is not performed, and the origiNonel LSP and traffic are not affected.

If static and dynamic BFD for TE LSP is simultaneously configured, the static BFD session takes effect.

None

If static and dynamic BFD for TE LSP is simultaneously configured, the static BFD session takes effect.

Re-optimization can be performed only after CSPF is enabled globally.

Re-optimization and CSPF are used together.

If CSPF is not enabled, pre-computation cannot be performed, and re-optimization does not take effect.

Automatic bandwidth adjustment can be performed only after CSPF is enabled globally.

Automatic bandwidth adjustment and CSPF are used together.

If CSPF is not enabled, pre-computation cannot be performed, and automatic bandwidth adjustment does not take effect.

Diabling the automatic configuration of global physical interface bandwidth is supported only in the PTN mode.

None

The automatic configuration of physical interface bandwidth is not supported and does not take effect in the CX mode.

Translation
Download
Updated: 2019-01-03

Document ID: EDOC1100055103

Views: 20168

Downloads: 35

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