CN108243102B

CN108243102B - Method for realizing fast rerouting and PE equipment

Info

Publication number: CN108243102B
Application number: CN201611227013.8A
Authority: CN
Inventors: 卢祖友
Original assignee: Maipu Communication Technology Co Ltd
Current assignee: Maipu Communication Technology Co Ltd
Priority date: 2016-12-27
Filing date: 2016-12-27
Publication date: 2022-03-11
Anticipated expiration: 2036-12-27
Also published as: CN108243102A

Abstract

The embodiment of the invention discloses a method for realizing fast rerouting and PE (provider edge) equipment, relates to the technical field of communication, and aims to solve the problems that in the prior art, a scheme for realizing VPN FRR (virtual private network) is incompatible and control is complex when a user side link fails. The method comprises the following steps: after the second PE device detects the link failure with the CE device, the next hop of the route pointing to the CE device is switched to the standby next hop; deleting a first private network route, and deleting ILM table entries corresponding to the first private network route in the ILM table in a delayed manner so as to search a next hop from the undeleted ILM table entries when a message pointing to the CE equipment needs to be forwarded, wherein the first private network route is a Border Gateway Protocol (BGP) private network route learned from the CE equipment; and when the first private network route is deleted, sending a standard BGP notification message to the first PE device of the opposite end, wherein the notification message is used for indicating that the second PE device deletes the first private network route.

Description

Method for realizing fast rerouting and PE equipment

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method for implementing fast reroute and a PE device.

Background

Today, when a network is developed at a high speed, operators pay great attention to the service convergence speed during network failure, when any node fails, the service switching of adjacent nodes is less than 50ms, and the end-to-end service convergence is less than 1s, which gradually becomes the threshold level index of a bearer network.

In order to meet the requirements that the service Switching of adjacent nodes is less than 50ms and the end-to-end service convergence is less than 1s, the MPLS (Multi-Protocol Label Switching, multiprotocol Label Switching) TE (Traffic Engineering) FRR (Fast Re-Route, Fast reroute) technology and the IGP (internet Gateway Protocol) routing Fast convergence technology (IP FRR) are developed by themselves, but they cannot solve the problem of end-to-end Fast convergence of the service when a PE device node fails in a network of CE (Customer Edge, user network Edge) dual-homed PE (Provider Edge, service Provider network Edge).

The Virtual Private Network (VPN) FRR in the prior art is a means for solving the problem of end-to-end service convergence of a common Network model, such as a CE dual-homed PE, and controls the convergence time of end-to-end service under a PE node failure condition within 1s by issuing backup path information in advance and switching to a backup path in case of failure.

In the typical CE dual-homed PE network shown in fig. 1, a main/standby path relationship is formed on PE1 and PE2, that is, a VPN FRR networking is formed. When PE2 fails, as shown in fig. 1, switching from the main path CE1 → PE1 → PE2 → CE2 to the standby path CE1 → PE1 → PE3 → CE2 is done on PE 1; when the user side link between PE2 and CE2 fails, referring to fig. 2, switching from the primary path CE1 → PE1 → PE2 → CE2 to the alternate path CE1 → PE1 → PE2 → PE3 → CE2 is done on PE 2.

However, when the link between PE2 and CE2 fails, the above-mentioned VPN FRR switching process has at least the following problems:

when a user side link between PE2 and CE2 fails (for example, a device port shutdown), BGP (Border Gateway Protocol) between PE2 and CE2 senses and cancels BGP private network routing (a VPN private network label originally allocated by PE2 does not exist), and the cancellation of far-end PE1 routing is relatively slow, which causes PE1 to still use PE2 to forward through a private network label issued by BGP, and PE2 cannot perform label termination, resulting in serious packet loss.

In the prior art, there is a scheme for solving the above packet loss problem, and the key points are as follows:

1. after the link between PE2 and CE2 fails, the route learned from CE2 continues to be retained, while the next hop of the route is set to PE 3;

2. PE2 informs PE1 to delete the route from CE2 (using the BGP extension attribute);

3. after PE2 receives the success of route deletion of PE1, it deletes the route learned locally from CE 2.

The disadvantages of this solution are:

1. PE2 informs PE1 of deleting the routing message to use the extended attribute of BGP, and simultaneously requires PE1 to notify PE2 after the deletion is successful, and the actions are not standard protocol behaviors; compatibility issues with the solution when PE1 is a friend device;

2. route deletion between PE2 and all other PEs requires validation and is complex to implement.

Disclosure of Invention

Embodiments of the present invention provide a method for implementing fast reroute and a PE device, so as to solve the problem that a scheme for implementing VPN FRR in the prior art is incompatible and the problem of complex control when a user side link fails.

In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:

in a first aspect, an embodiment of the present invention provides a method for implementing fast reroute, which is applied to a VPN network of a CE dual homing provider network edge PE in a user network edge, in which a CE device is connected to a second PE device and a third PE device, the first PE device is an opposite-end PE device to the second PE device and the third PE device, the second PE device enables a VPN FRR, and an address of the third PE device is configured as a next hop of a route pointing to the CE device, and the method includes:

after detecting the link failure between the second PE equipment and the CE equipment, switching the next hop of the route pointing to the CE equipment to the standby next hop;

deleting a first private network route, and delaying and deleting an ILM table item corresponding to the first private network route in an incoming label mapping ILM table so as to search a next hop from the un-deleted ILM table item when a message pointing to the CE equipment needs to be forwarded, wherein the first private network route is a Border Gateway Protocol (BGP) private network route learned from the CE equipment;

and when the first private network route is deleted, sending a standard BGP notification message to the first PE equipment of the opposite end, wherein the notification message is used for indicating that the second PE equipment deletes the first private network route.

Optionally, the performing delayed deletion of the ILM entry corresponding to the first private network route in the incoming label mapping ILM table includes: searching an ILM table item to be deleted corresponding to the prefix of the first private network route in the ILM table; setting delay time for a first batch of ILM entries to be deleted after the link failure, and starting a timer by using the delay time of the first batch of ILM entries to be deleted as timing time, wherein the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted, the waiting time of each ILM entry to be deleted is the same, and the waiting time is the time from the time of setting delay deletion to the time of executing deletion of the ILM entries to be deleted; setting delay time for the non-first ILM entries to be deleted after the link failure, wherein the delay time of the non-first ILM entries to be deleted is the time from the current starting time of a timer to the time for deleting the ILM entries to be deleted; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time in the refreshed delay time as the timing time, wherein the refreshed delay time is the difference between the delay time of the non-overtime ILM table items to be deleted and the timing time overtime of the timer at this time.

Optionally, the performing delayed deletion of the ILM entry corresponding to the first private network route in the incoming label mapping ILM table further includes: and acquiring the waiting time of each ILM table item to be deleted according to the scale of the VPN network.

Optionally, the performing delayed deletion of the ILM entry corresponding to the first private network route in the incoming label mapping ILM table includes: searching an ILM table item to be deleted corresponding to the prefix of the first private network route in the ILM table; acquiring the waiting time of each ILM table item to be deleted, wherein the waiting time is the time from the time of setting delayed deletion to the time of executing deletion of the ILM table item to be deleted; setting delay time for the first ILM list items to be deleted after the link failure, and starting a timer by taking the shortest delay time of the first ILM list items to be deleted as timing time, wherein the delay time of the first ILM list items to be deleted is the waiting time of the first ILM list items to be deleted; setting delay time for the non-first ILM list items to be deleted after the link failure, wherein the delay time of the non-first ILM list items to be deleted is the time from the current starting time of a timer to the time for deleting the ILM list items to be deleted; if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, wherein the first ILM table entry to be deleted is the ILM table entry of which the delay time is set after the timer is started currently, and the first delay time is the time from the current refreshing time to the deletion of the ILM table entry to be deleted; restarting the timer by taking the shortest first delay time as the timing time; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the ILM table items to be deleted which are not overtime into second delay time, and restarting the timer by taking the shortest delay time in the second delay time as the timing time, wherein the second delay time is the difference between the delay time of the ILM table items to be deleted which are not overtime and the timing time of the current time of the timer.

Optionally, the setting a delay time for the to-be-deleted ILM entry includes: and adding a delay time item in the ILM table so as to record the delay time of the ILM table entry to be deleted in the delay time item.

Optionally, the method further includes: acquiring an ILM table entry to be added, and checking whether a first ILM table entry which is set to be deleted in a delayed manner and is not deleted exists in the ILM table and has the same prefix with the ILM table entry to be added; if the first ILM table entry does not exist, adding the ILM table entry to be added into the ILM table; if the first ILM table entry exists, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same, wherein the attributes comprise an operation code and a next hop; if the ILM table entries to be added are the same, the ILM table entries to be added are not added, and the setting of the delayed deletion of the first ILM table entry is cancelled; and if not, adding the ILM table entry to be added, and deleting the first ILM table entry.

In a second aspect, an embodiment of the present invention provides a PE device, which is one of two PE devices connected to a CE device in a network of CE dual-homed PEs, where the PE device enables a VPN FRR, and configures an address of another PE device connected to the CE device as a standby next hop of a route pointing to the CE device, and the PE device includes:

the detection module is used for detecting a link between the CE devices connected with the detection module;

a switching processing module, configured to instruct the route updating module to switch a next hop of a route pointing to the CE device to the standby next hop after the detection module detects a link failure; deleting a first private network route, and delaying and deleting an ILM table item corresponding to the first private network route in an incoming label mapping ILM table so as to search a next hop from the un-deleted ILM table item when a message pointing to the CE equipment needs to be forwarded, wherein the first private network route is a Border Gateway Protocol (BGP) private network route learned from the CE equipment; when the first private network route is deleted, sending a standard BGP notification message to a first PE device of an opposite end of the first private network route, wherein the notification message is used for indicating that the second PE device deletes the first private network route;

and the route updating module is used for updating the route according to the indication of the switching processing module.

Optionally, the switching processing module is specifically configured to search, in the ILM table, an ILM entry to be deleted corresponding to a prefix of the first private network route; setting delay time for a first batch of ILM entries to be deleted after the link failure, and starting a timer by using the delay time of the first batch of ILM entries to be deleted as timing time, wherein the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted, the waiting time of each ILM entry to be deleted is the same, and the waiting time is the time from the time of setting delay deletion to the time of executing deletion of the ILM entries to be deleted; setting delay time for the non-first ILM entries to be deleted after the link failure, wherein the delay time of the non-first ILM entries to be deleted is the time from the current starting time of a timer to the time for deleting the ILM entries to be deleted; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time in the refreshed delay time as the timing time, wherein the refreshed delay time is the difference between the delay time of the non-overtime ILM table items to be deleted and the timing time overtime of the timer at this time.

Optionally, the switching processing module is further specifically configured to obtain, according to the scale of the VPN network, a waiting time for each item of the ILM to be deleted.

Optionally, the switching processing module is specifically configured to search, in the ILM table, an ILM entry to be deleted corresponding to a prefix of the first private network route; acquiring the waiting time of each ILM table item to be deleted, wherein the waiting time is the time from the time of setting delayed deletion to the time of executing deletion of the ILM table item to be deleted; setting delay time for the first ILM list items to be deleted after the link failure, and starting a timer by taking the shortest delay time of the first ILM list items to be deleted as timing time, wherein the delay time of the first ILM list items to be deleted is the waiting time of the first ILM list items to be deleted; setting delay time for the non-first ILM list items to be deleted after the link failure, wherein the delay time of the non-first ILM list items to be deleted is the time from the current starting time of a timer to the time for deleting the ILM list items to be deleted; if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, wherein the first ILM table entry to be deleted is the ILM table entry of which the delay time is set after the timer is started currently, and the first delay time is the time from the current refreshing time to the deletion of the ILM table entry to be deleted; restarting the timer by taking the shortest first delay time as the timing time; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the ILM table items to be deleted which are not overtime into second delay time, and restarting the timer by taking the shortest delay time in the second delay time as the timing time, wherein the second delay time is the difference between the delay time of the ILM table items to be deleted which are not overtime and the timing time of the current time of the timer.

Optionally, the switching processing module is further specifically configured to add a delay time entry in the ILM table, so as to record the delay time of the ILM entry to be deleted in the delay time entry.

Optionally, the apparatus further comprises: an adding module, configured to obtain an ILM entry to be added, and check whether a first ILM entry in the ILM table has a same prefix as the ILM entry to be added, where the first ILM entry is an ILM entry in the ILM table that has been set to be deleted in a delayed manner and is not deleted; if the first ILM table entry does not exist, adding the ILM table entry to be added into the ILM table; if the first ILM table entry exists, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same, wherein the attributes comprise an operation code and a next hop; if the ILM table entries to be added are the same, the ILM table entries to be added are not added, and the setting of the delayed deletion of the first ILM table entry is cancelled; and if not, adding the ILM table entry to be added, and deleting the first ILM table entry.

When a second PE device (PE2) detects that a main path is not forwarded successfully, the second PE device switches to a standby next hop, immediately deletes a private network route learned from the main path, but deletes an ILM (Internet Link management) table entry corresponding to the private network route in a delayed manner, and immediately notifies a first PE device (PE1) at an opposite end after deleting the private network route, wherein the PE2 deletes the private network route learned from the main path. Therefore, the problems that in the prior art, the PE1 still uses the PE2 to forward through the private network label released by BGP, so that the PE1 cannot terminate the label and further causes packet loss are solved, and meanwhile, the problems of incompatibility and complex operation of the prior art are solved. Therefore, the packet loss of the operator network is reduced to a greater extent, the performance index of the equipment is improved, the satisfaction degree of customers is improved, better experience is provided for users, and the service quality of the operator network is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic diagram of VPN FRR switching for node failure in a CE dual-homed PE network in the prior art;

fig. 2 is a schematic diagram of VPN FRR switching for a user side link failure in a CE dual-homed PE network in the prior art;

fig. 3 is a flowchart of an implementation method of a VPN FRR according to an embodiment of the present invention;

fig. 4 is a flowchart of an implementation method of a VPN FRR according to an embodiment of the present invention;

fig. 5 is a flowchart of an implementation method of a VPN FRR according to an embodiment of the present invention;

fig. 6 is a block diagram of a PE device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The term "and/or" in the embodiment of the present invention is only one kind of association relationship describing an associated object, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" in the embodiment of the present invention generally indicates that the preceding and following related objects are in an "or" relationship.

In addition, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless otherwise specified.

Example one

The embodiment of the invention provides a method for realizing fast rerouting, when a user side link fails, a route of the failed link is deleted, but an ILM (Incoming Label Map) table entry corresponding to the route is deleted in a delayed manner, and when a message exists, a corresponding next hop is found by searching the ILM table entry deleted in the delayed manner. Therefore, packet loss is reduced in the process of realizing VPN FRR, and the problems of incompatibility of schemes for realizing VPN FRR and complex control in the prior art are solved.

The following describes an embodiment of the present invention in detail by taking the VPN network of the CE dual-homed service provider network edge PE shown in fig. 2 as an example. In the VPN network, a CE device (CE2) is connected to a second PE device (PE2) and a third PE device (PE3), respectively, a first PE device (PE1) is a PE device that is opposite to PE2 and PE3, respectively, PE1 is connected to CE1, a VPN FRR is enabled on PE2, and an address of PE3 is configured as a standby next hop of a route pointing to CE 2.

It should be noted that, in the embodiment of the present invention, for any PE device that enables VPN FRR, only one PE device is taken as an example for description, for example, a PE3 device is taken as a PE2 device pointing to a standby next hop of CE2, but is not limited to one, and multiple PE devices may be taken as standby next hops; meanwhile, there may be one or more P devices between the opposite-end PE devices, for example, between PE1 and PE2 or PE3, which is not described herein again.

Specifically, VPN FRR related configuration is performed on PE2, including configuration of VPN FRR routing policy, configuration of backup next hop, and enabling VPN FRR. The spare next hop configured for the route pointing to the CE2 on the PE2 points to the PE3, that is, the spare next hop is a loopback port address of the PE3 (the loopback address of the PE3 can be used to identify the device, PE 3). VPN FRR related configuration statements reference is made to the prior art.

According to the standard MPLS L3VPN technology, PE2 and PE3 both publish BGP private network routes to PE1 that point to CE2 and assign private network labels. PE1 is able to learn BGP private network routes learned on the CE2 → PE2 → PE1 path, and BGP private network routes learned on the CE2 → PE3 → PE1 path. The PE1 selects eligible BGP private network routes according to the matching strategy, wherein the BGP private network routes comprise suboptimal BGP private network routes published by PE3 in addition to the optimized BGP private network routes published by PE 2; the PE1 fills BGP private network routing information (including forwarding prefix, inner Label, and selected outer LSP (Label Switched Path) tunnel) issued by the PE2 and PE3 in the forwarding table entry to guide forwarding.

Under normal conditions, PE1 only performs forwarding processing according to BGP private network routing information issued by preferred PE2, so the path for CE1 to access CE2 is: CE1 → PE1 → PE2 → CE 2. When PE2 detects a link failure, a VPN FRR switch will be triggered. As shown in fig. 3, the method specifically includes:

s101 (optional), PE2 detects a link failure between PE2 and CE 2.

The BFD detection function is enabled on PE2 and CE2, and CE2 and PE2 are configured as BFD peer devices for detecting connectivity of a bidirectional forwarding path between CE2 and PE 2. PE2 discovers a link failure with CE2 through BFD detection.

S102, PE2 switch the next hop of the route directed to CE2 to standby next hop PE 3.

Since the alternate next hop for the route directed to CE2 configured on PE2 is the IP address of PE3, PE2 updates the next hop for the route directed to CE1 from the IP address of PE2 to the IP address of PE3 here.

S103, PE2 deletes the first private network route. And the ILM table entry corresponding to the first private network route in the ILM table is deleted in a delayed mode.

Wherein the first private network route is a border gateway protocol BGP private network route learned from CE 2.

Conventionally, at this time, the BGP private network route learned from the CE2 (i.e., the BGP private network route learned from the CE2 → PE2 path) and the ILM entry corresponding to the BGP private network route need to be deleted, but in the embodiment of the present invention, only the BGP private network route is deleted, and the ILM entry corresponding to the BGP private network route is deleted in a delayed manner.

Therefore, when a message pointing to the CE2 needs to be forwarded on the PE2, the next hop is searched from the undeleted ILM entry.

PE1 perceives that the link between PE2 and CE2 is down and the process of performing a route update lasts for a period of time during which the un-updated route is then used to direct forwarding, i.e. take the backup path: CE1 → PE1 → PE2 → PE3 → CE 2. Specifically, if the PE1 receives the message intended for the CE2 when the operation of revoking the BGP route has not been completed, the private network tag allocated by the PE2 is used and forwarded to the PE2, and since the next hop of the route pointing to the CE2 has been switched to the IP address of the PE3 on the PE2, but the ILM entry of the BGP private network route has not been deleted yet due to the delayed deletion, the PE1 can terminate and forward the private network tag of the message, and the message can be forwarded from the path CE1 → PE1 → PE2 → PE3 → CE2 to the CE2, and there is no packet loss in the entire process.

Referring to fig. 4, the delay deletion of the ILM entry corresponding to the first private network route in the ILM table specifically includes the following steps:

and S1031, searching ILM table items to be deleted corresponding to the prefix of the first private network route in the ILM table.

An ILM table entry corresponding to the first private network route is recorded in the ILM table. When detecting a link failure between PE2 and CE2, PE2 deletes the first private network route, but before PE1 reconverges and updates the route, in order to reduce packet loss, in the embodiment of the present invention, the ILM entry corresponding to the first private network route is deleted in a delayed manner. Therefore, after the first private network route needing to be deleted is determined, according to the prefix of the first private network route, the ILM item corresponding to the first private network route is searched in the ILM table, delayed deletion is set, and the searched ILM item is the item to be deleted.

S1032 (optional) adds a delay time entry in the ILM table, so as to record the delay time of the ILM entry to be deleted in the delay time entry.

In the embodiment of the present invention, the ILM table includes a prefix, an opcode, and a next hop entry, which are shown in table 1, and may also include other entries, which are not described herein again, and refer to the prior art specifically.

TABLE 1

ILM meter	Prefix	Operation code	Next hop
				Table item 1	xxxxx	xxxxx	xxxxx
Table entry 2	xxxxx	xxxxx	xxxxx
				......	......	......	......

It should be noted that: the actual ILM table may not have entries of table entry 1 and table entry 2, which is only present in the embodiment of the present invention for the sake of clarity of the table.

There are many methods for setting a delay to an ILM entry corresponding to a first private network route. Preferably, a delay time entry may be added to the original ILM table to store the delay time of each ILM entry, as shown in table 2.

TABLE 2

ILM meter	Prefix	Operation code	Next hop	Time delay (New increasing item)
					Table item 1	xxxxx	xxxxx	xxxxx
Table entry 2	xxxxx	xxxxx	xxxxx
					......	......	......	......

Optionally, a new delay deletion table may be constructed without changing the original ILM table, and is used to store the delay time of each ILM entry. For example, the new delayed deletion table includes a code (for representing each ILM entry) and a delay time, which may be, for example, the number of the ILM entry, such as entry 1, entry 2, etc., the prefix in the ILM entry, etc., as shown in table 3.

TABLE 3

ILM meter	Time delay
		Table item 1
Table entry 2
		......

S1033 (optional), obtaining the waiting time of each to-be-deleted ILM entry according to the scale of the VPN network, where the waiting time of each to-be-deleted ILM entry is the same.

At this time, PE2 has deleted the first private network route, and has found the corresponding ILM according to the label of the first private network route as setting the delayed deletion to the entry to be deleted. When the delay deletion is set, the waiting time of each ILM entry needs to be known, wherein the waiting time is the time from the time of setting the delay deletion to the time of executing the deletion of the ILM entry to be deleted, and the waiting time of each ILM entry to be deleted in the step is the same.

Preferably, the waiting time may be set according to the size of the VPN network. In fact, the length of the waiting time is related to the time for deleting the routing operation on the PE device and the routing convergence time, and the larger the network size is, the longer the time is, so that the waiting time can be set according to the size of the VPN network. The size of the VPN network may be determined according to the number of routes, the number of PE devices, and the network size level of the VPN network. For example, the larger the number of routes in the VPN network, the larger the network size, the longer the waiting time; moreover, higher levels indicate larger network sizes and longer latencies.

Optionally, the waiting time of each ILM entry may also be set in advance, or determined by those skilled in the art according to actual situations.

S1034, setting delay time for the first batch of ILM table items to be deleted after the link failure, starting a timer by taking the delay time of the first batch of ILM table items to be deleted as timing time, and setting delay time for the non-first batch of ILM table items to be deleted after the link failure.

Because there are multiple routes between PE2 and CE2, route deletion takes time, and a route may be deleted in a batch, where a batch may be one or multiple, and then the ILM entry to be deleted is also a set delay time in a batch.

Preferably, only one timer is set, so that all the ILM entries to be deleted share one timer, the timer may be overtime in the whole process, the timer is restarted at a new timing time after the expiration, and each time the timer is restarted, the zero time of the timer start is all the time, so the set delay time is changed according to the start time of each timer.

Specifically, after the delay time is set for the first group of ILM entries to be deleted, the delay time of the first group of ILM entries to be deleted is used as the timing time and the timer is started at the same time. Therefore, for the first batch of ILM entries to be deleted, the zero time when the timer is started is the delay time set for the first batch of ILM entries to be deleted, that is, the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted. And the delay time of the non-first batch of ILM table items to be deleted is the time from the current starting time of the timer to the time for deleting the ILM table items to be deleted. The current starting time of the timer refers to the time when the timer is started at the new timing time last time.

Optionally, because the waiting time of all the ILM entries to be deleted is the same, a timer may be set for each batch of ILM entries to be deleted, so that the delay time of each batch of ILM entries to be deleted is the waiting time, and the timing time of each timer is the waiting time. And when one timer is overtime, deleting all the ILM entries to be deleted corresponding to the timer. Setting the timer in this way is relatively simple to operate, but is relatively wasteful.

And S1035, when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time after refreshing as the timing time.

Specifically, when the timer times out for a certain time, deleting the ILM entry to be deleted that times out in the ILM table, that is, deleting the ILM entry to be deleted whose delay time is the same as the current timing time of the timer, where the ILM entry to be deleted whose delay time is different from the current timing time of the timer is the ILM entry to be deleted that has not timed out. And refreshing the delay time of the ILM list item to be deleted which is not overtime, wherein the refreshed delay time is the difference between the delay time of the ILM list item to be deleted which is not overtime and the overtime timing time of the timer. And restarting the timer by taking the shortest delay time after refreshing as the timing time.

For example, assuming that the waiting time is 5 minutes, a batch has only one ILM entry to be deleted, and a delay time is set for one ILM entry to be deleted every other minute. Referring to table 4, setting a delay time of 5 for a first batch of ILM entries to be deleted (i.e., entry 1), and starting a timer at time 0; when the timer is started for 1 minute, setting the delay time of 6 for the 2 nd batch of ILM table items to be deleted (namely table items 2), namely adding the time passed by the timer to the waiting time (1+ 5); by analogy, when the timer is started for 4 minutes, the delay time is set to 9(4+5) for the 5 th ILM entry to be deleted (i.e., entry 5). TABLE 4

ILM meter	Prefix	Operation code	Next hop	Time delay (minutes)
					Table item 1	xxxxx	xxxxx	xxxxx	5
Table entry 2	xxxxx	xxxxx	xxxxx	6
					Table item 3	xxxxx	xxxxx	xxxxx	7
Table entry 4	xxxxx	xxxxx	xxxxx	8
					Table item 5	xxxxx	xxxxx	xxxxx	9
Table entry 6	xxxxx	xxxxx	xxxxx	10

When the timer is started for 5 minutes, a delay time is set for the 6 th ILM table entry to be deleted (namely, table entry 6), the delay time of the table entry 6 is 10(5+5), and when the timer is overtime, as shown in the table 5, the table entry 1 is deleted when the time is overtime, and the delay times from the table entry 2 to the table entry 6 are refreshed, respectively

Changes to 1, 2, 3, 4, 5, and then the timer is restarted at a timing of 1 minute.

TABLE 5

ILM meter	Prefix	Operation code	Next hop	Time delay (minutes)
					Table entry 2	xxxxx	xxxxx	xxxxx	1
Table item 3	xxxxx	xxxxx	xxxxx	2
					Table entry 4	xxxxx	xxxxx	xxxxx	3
Table item 5	xxxxx	xxxxx	xxxxx	4
					Table entry 6	xxxxx	xxxxx	xxxxx	5
Table entry 7	xxxxx	xxxxx	xxxxx	6

When the timer is started for 1 minute currently, delay time is set for the 7 th ILM table entry to be deleted (namely, the table entry 7), the delay time of the table entry 7 is 6(1+5), and when the timer is overtime, the table entry 2 is overtime deleted, the delay time from the table entry 3 to the table entry 7 in refreshing is respectively changed into 1, 2, 3, 4 and 5, and at this time, the timer is restarted by taking 1 minute as the timing time.

And repeating the steps until all ILM entries to be deleted are deleted after time is out, and closing the timer.

It should be noted that the above example is only for explaining how to set the delay time for the ILM entry to be deleted, and how to refresh the delay time every time the timer times out, but the above data does not represent an actual situation.

Or the following steps may be included after the above step S1032 (optional):

and S1036, obtaining the waiting time of each ILM table item to be deleted, wherein the waiting time of each ILM table item to be deleted is definitely the same or different.

And the waiting time is the time from the time of setting the delayed deletion to the time of executing the deletion of the ILM table entry to be deleted.

Specific method of acquiring the waiting time refers to the above-mentioned step S1033,

s1037, setting delay time for the first batch of ILM entries to be deleted after the link failure, starting a timer by taking the shortest delay time of the first batch of ILM entries to be deleted as timing time, and setting delay time for the non-first batch of ILM entries to be deleted after the link failure.

The delay time of the first batch of ILM table items to be deleted is the waiting time of the first batch of ILM table items to be deleted, and the delay time of the non-first batch of ILM table items to be deleted is the time from the current starting time of the timer to the time for deleting the ILM table items to be deleted.

Since the waiting time may be different, when the timer is started for the first time, the timer is started with the shortest delay time among the delay times of the first batch of entries to be deleted. And if only one ILM table entry to be deleted exists in the first batch, starting a timer by taking the delay time as the timing time. Otherwise, refer to step S1034, which is not described herein again.

S1038, if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, and taking the shortest of the first delay times as the timing time to restart the timer.

Because the waiting time may be different, after a certain time of timer starting, the delay time of the ILM entry to be deleted for which the delay time is newly set may be smaller than the timing time of the current timer, and at this time, the delay time needs to be refreshed and the timer needs to be restarted.

Specifically, if the delay time of the first to-be-deleted ILM entry is shorter than the current timing time of the timer, the delay time of each to-be-deleted ILM entry is refreshed into a first delay time, the first to-be-deleted ILM entry is an ILM entry for which the delay time is set after the timer is currently started, and the first delay time is the time from the current refreshing time to the time for deleting the ILM entry to be deleted; and the shortest one of the first delay times is taken as the timing time to restart the timer.

For example, assuming that there are two ILM entries to be deleted in a batch, a delay time is set every one minute for a batch of ILM entries to be deleted. Firstly, the waiting time of each ILM table entry to be deleted is respectively as follows: table 1 to table 4 are 6 minutes, 7 minutes, 3 minutes, 6 minutes, respectively.

Referring to table 6, the first list items of the ILM to be deleted are list item 1 and list item 2, the delay times are 6 and 7, respectively, and the timer is started with the short delay time 6 as the timing time; when the timer is started for 1 minute, the second ILM entries to be deleted are entry 3 and entry 4, and the delay times are 4(1+3) and 7(1+6), respectively. It can be known that the delay time 4 of the table entry 3 is less than the current timing time 6 of the timer, and the delay time of each to-be-deleted ILM entry needs to be refreshed, as shown in table 7, the delay times are 5, 6, 3, and 6 respectively after the refresh, and the timer is restarted with the minimum 3 as the timing time.

TABLE 6

ILM meter	Prefix	Operation code	Next hop	Time delay (minutes)
					Table item 1	xxxxx	xxxxx	xxxxx	6
Table entry 2	xxxxx	xxxxx	xxxxx	7
					Table item 3	xxxxx	xxxxx	xxxxx	4
Table entry 4	xxxxx	xxxxx	xxxxx	7

TABLE 7

ILM meter	Prefix	Operation code	Next hop	Time delay (minutes)
					Table item 1	xxxxx	xxxxx	xxxxx	5
Table entry 2	xxxxx	xxxxx	xxxxx	6
					Table item 3	xxxxx	xxxxx	xxxxx	3
Table entry 4	xxxxx	xxxxx	xxxxx	6

It should be noted that: the above example is only for illustration, how to set the delay time for each to-be-deleted ILM entry, and how to refresh the delay time of each to-be-deleted ILM entry and restart the timer when the delay time of an entry is smaller than the delay time of the current timer, but the above data does not represent an actual situation.

S1039, when the timer times out, deleting the ILM table entry to be deleted which times out in the ILM table, refreshing the delay time of the ILM table entry to be deleted which does not time out into second delay time, and taking the shortest of the second delay time as the timing time to restart the timer.

Specifically, the second delay time is a difference between the delay time of the ILM entry to be deleted that is not timed out and the timing time of the timer when the time is timed out this time. The second delay time is the delay time after the refresh in step S1035, and the specific process refers to the description of step S1034, which is not described herein again.

It should be noted that: in the embodiment of the invention, delay time is set for each batch of ILM table items to be deleted in sequence, when the delay time is set for the first batch of ILM table items to be deleted, a timer is started for the first time, then each time the timer is overtime, the ILM table items to be deleted which are overtime are deleted, the delay time of the existing ILM table items which are not overtime and are to be deleted is refreshed, and the timer is restarted with the shortest delay time among the refreshed delay times. And if the delay time of the new ILM table entry to be deleted with the set delay time is less than the timing time of the current timer after the timer is started, the delay time of the ILM table entry to be deleted is also refreshed virtually, and the timer is restarted with the shortest delay time after refreshing. And the order of the steps is not fixed and can be exchanged.

And S104, when the first private network route is deleted from the PE2, the PE2 sends a standard BGP notification message to the PE1 device at the opposite end of the PE2, wherein the notification message is used for indicating that the PE2 deletes the first private network route.

When the PE2 deletes the first private network route, an advertisement message is immediately sent to the PE1 device. After receiving the standard BGP advertisement message sent by the PE2 device, the PE1 learns that the PE2 has deleted the BGP private network route learned from the CE 2. Thus PE1 knows that the link between PE2 and CE2 failed, deletes the BGP private network route learned from PE2, and updates the route. Specifically, PE1 may implement the deletion of BGP routes through a BGP route update process. The BGP route update procedure may include: the routing information learned from PE2 (including the forwarding prefix, the inner label, the selected outer LSP tunnel) is removed from the forwarding table entry. Thereafter, the path for accessing CE2 by CE1 is switched from CE1 → PE1 → PE2 → CE2 to CE1 → PE1 → PE3 → CE 2.

In the course of performing the above steps S103-S104, it is possible that the link failure between PE2 and CE2 has been repaired, at which point the link from PE2 to CE2 becomes reachable again, at which point it may be necessary to add an ILM entry in the ILM table. Specifically, referring to fig. 5, the method includes the following steps:

s201, obtaining an ILM table entry to be added.

When the fault is repaired, the route converges, and a new BGP private network route learned from CE2 is added, the corresponding ILM entry needs to be added in the ILM table.

S202, checking whether a first ILM table entry in the ILM table has the same prefix with the ILM table entry to be added.

The first ILM table entry is an ILM table entry which is set to be deleted in a delayed way and is not deleted.

If the first ILM entry does not exist, step S203 is executed, and if the first ILM entry exists, step S204 is executed.

S203, adding an ILM table entry to be added into the ILM table.

And if the first ILM table entry with the same prefix as the ILM table entry to be added does not exist, the ILM table entry to be added is directly added.

S204, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same.

Wherein the attributes include an opcode and a next hop.

The method comprises the steps of having a first ILM table entry with the same prefix as an ILM table entry to be added, indicating that the ILM table entry with the same prefix as the ILM table entry to be added may exist, checking whether the operation code and the next hop of the ILM table entry to be added and the first ILM table entry are the same, if so, indicating that the ILM table entry to be added and the first ILM table entry are completely the same, and if not, indicating that the ILM table entry to be added and the first ILM table entry are not completely the same and the ILM table entry to be added is a more optimal route.

If the two are the same, step S205 is executed, and if the two are not the same, step S206 is executed.

S205, not adding ILM table items to be added, and canceling the delayed deletion of the first ILM table item.

And if the ILM table entry to be added is completely the same as the first ILM table entry, only the delayed deletion of the first ILM table entry needs to be cancelled without adding the ILM table entry. This prevents the flow of current from being interrupted when the route is switched back.

S206, adding ILM table items to be added and deleting the first ILM table item.

And deleting the first ILM table entry and adding the more optimal ILM table entry to be added.

The embodiment of the invention provides a method for realizing fast reroute, when detecting that a main path is not forwarded successfully, switching to a standby next hop, immediately deleting a private network route learned from the main path, but deleting ILM table items corresponding to the private network route in a delayed manner, and immediately notifying opposite-end PE equipment (PE1) after deleting the private network route, wherein the PE2 deletes the private network route learned from the main path. Therefore, the problems that in the prior art, the PE1 still uses the PE2 to forward through the private network label released by BGP, so that the PE1 cannot terminate the label and further causes packet loss are solved, and meanwhile, the problems of incompatibility and complex operation of the prior art are solved. Therefore, the packet loss of the operator network is reduced to a greater extent, the performance index of the equipment is improved, the satisfaction degree of customers is improved, better experience is provided for users, and the service quality of the operator network is improved.

Example two

The embodiment of the present invention provides a PE device, which is one of two PE devices connected to a CE device in a network of CE dual-homed PE, where the PE device enables a VPN FRR, and configures an address of another PE device connected to the CE device as a standby next hop of a route pointing to the CE device, and implementation of each functional module in the device may refer to the first embodiment, and is not described herein again. The apparatus shown with reference to fig. 4 comprises:

a detection module 11, configured to detect a link between CE devices connected to the device;

a switching processing module 12, configured to instruct the route updating module to switch a next hop of a route directed to the CE device to the standby next hop after the detection module detects a link failure; deleting a first private network route, and delaying and deleting an ILM table item corresponding to the first private network route in an incoming label mapping ILM table so as to search a next hop from the un-deleted ILM table item when a message pointing to the CE equipment needs to be forwarded, wherein the first private network route is a Border Gateway Protocol (BGP) private network route learned from the CE equipment; when the first private network route is deleted, sending a standard BGP notification message to a first PE device of an opposite end of the first private network route, wherein the notification message is used for indicating that the second PE device deletes the first private network route;

and a route updating module 13, configured to update a route according to the instruction of the switching processing module.

Optionally, the switching processing module 12 is specifically configured to search, in the ILM table, an ILM entry to be deleted corresponding to a prefix of the first private network route; setting delay time for a first batch of ILM entries to be deleted after the link failure, and starting a timer by using the delay time of the first batch of ILM entries to be deleted as timing time, wherein the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted, the waiting time of each ILM entry to be deleted is the same, and the waiting time is the time from the time of setting delay deletion to the time of executing deletion of the ILM entries to be deleted; setting delay time for the non-first ILM entries to be deleted after the link failure, wherein the delay time of the non-first ILM entries to be deleted is the time from the current starting time of a timer to the time for deleting the ILM entries to be deleted; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time in the refreshed delay time as the timing time, wherein the refreshed delay time is the difference between the delay time of the non-overtime ILM table items to be deleted and the timing time overtime of the timer at this time.

Optionally, the switching processing module 12 is further specifically configured to obtain, according to the scale of the VPN network, a waiting time for each item of the ILM to be deleted.

Optionally, the switching processing module 12 is specifically configured to search, in the ILM table, an ILM entry to be deleted corresponding to a prefix of the first private network route; acquiring the waiting time of each ILM table item to be deleted, wherein the waiting time is the time from the time of setting delayed deletion to the time of executing deletion of the ILM table item to be deleted; setting delay time for the first ILM list items to be deleted after the link failure, and starting a timer by taking the shortest delay time of the first ILM list items to be deleted as timing time, wherein the delay time of the first ILM list items to be deleted is the waiting time of the first ILM list items to be deleted; setting delay time for the non-first ILM list items to be deleted after the link failure, wherein the delay time of the non-first ILM list items to be deleted is the time from the current starting time of a timer to the time for deleting the ILM list items to be deleted; if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, wherein the first ILM table entry to be deleted is the ILM table entry of which the delay time is set after the timer is started currently, and the first delay time is the time from the current refreshing time to the deletion of the ILM table entry to be deleted; restarting the timer by taking the shortest first delay time as the timing time; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the ILM table items to be deleted which are not overtime into second delay time, and restarting the timer by taking the shortest delay time in the second delay time as the timing time, wherein the second delay time is the difference between the delay time of the ILM table items to be deleted which are not overtime and the timing time of the current time of the timer.

Optionally, the switching processing module 12 is further specifically configured to add a delay time entry in the ILM table, so as to record the delay time of the ILM entry to be deleted in the delay time entry.

Optionally, the switching processing module 12 is further configured to obtain an ILM entry to be added, and check whether a first ILM entry in the ILM table has a same prefix as the ILM entry to be added, where the first ILM entry is an ILM entry in the ILM table that has been set to be deleted in a delayed manner and is not deleted; if the first ILM table entry does not exist, adding the ILM table entry to be added into the ILM table; if the first ILM table entry exists, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same, wherein the attributes comprise an operation code and a next hop; if the ILM table entries to be added are the same, the ILM table entries to be added are not added, and the setting of the delayed deletion of the first ILM table entry is cancelled; and if not, adding the ILM table entry to be added, and deleting the first ILM table entry.

It should be noted that, in this embodiment, each functional module may be a separately established processor, or may be implemented by being integrated in a certain processor of the PE device, or may be stored in a memory of the PE device in the form of program codes, and the certain processor of the PE device calls and executes the functions of the above units. The processor may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present invention.

The embodiment of the invention provides PE (provider edge) equipment, which is switched to a standby next hop when detecting that a main path is not forwarded, immediately deletes a private network route learned from the main path, but deletes an ILM (enterprise lead management) table entry corresponding to the private network route in a delayed manner, and immediately informs PE equipment (PE1) at an opposite end after deleting the private network route, wherein the private network route learned from the main path is deleted by the equipment. Therefore, the problem of packet loss is solved, and the problems of incompatibility and complex operation of the prior art are solved.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be physically included alone, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A method for realizing fast rerouting is applied to a VPN network of CE dual-homed PE, in the VPN network, CE devices are respectively connected to a second PE device and a third PE device, a first PE device and the second PE device and the third PE device are opposite-end PE devices respectively, a VPN FRR is enabled on the second PE device, and an address of the third PE device is configured as a standby next hop of a route pointing to the CE devices, and the method is characterized by comprising the following steps:

2. The method of claim 1, wherein the delaying deletion of the ILM table entry in the incoming label mapping ILM table corresponding to the first private network route comprises:

searching an ILM table item to be deleted corresponding to the prefix of the first private network route in the ILM table;

setting delay time for a first batch of ILM entries to be deleted after the link failure, and starting a timer by using the delay time of the first batch of ILM entries to be deleted as timing time, wherein the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted, the waiting time of each ILM entry to be deleted is the same, and the waiting time is the time from the time of setting delay deletion to the time of executing deletion of the I LM entries to be deleted;

setting delay time for the non-first ILM entries to be deleted after the link failure, wherein the delay time of the non-first ILM entries to be deleted is the time from the current starting time of a timer to the time for deleting the ILM entries to be deleted;

and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time in the refreshed delay time as the timing time, wherein the refreshed delay time is the difference between the delay time of the non-overtime ILM table items to be deleted and the timing time overtime of the timer at this time.

3. The method of claim 2, wherein said delaying deletion of an ILM entry in the incoming label mapped ILM table corresponding to the first private network route, further comprises:

and acquiring the waiting time of each ILM table item to be deleted according to the scale of the VPN network.

4. The method of claim 1, wherein the delaying deletion of the ILM table entry in the incoming label mapping ILM table corresponding to the first private network route comprises:

acquiring the waiting time of each ILM table item to be deleted, wherein the waiting time is the time from the time of setting delayed deletion to the time of executing deletion of the ILM table item to be deleted;

setting delay time for the first ILM list items to be deleted after the link failure, and starting a timer by taking the shortest delay time of the first ILM list items to be deleted as timing time, wherein the delay time of the first ILM list items to be deleted is the waiting time of the first ILM list items to be deleted;

setting delay time for the non-first ILM list items to be deleted after the link failure, wherein the delay time of the non-first ILM list items to be deleted is the time from the current starting time of a timer to the time for deleting the ILM list items to be deleted;

if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, wherein the first ILM table entry to be deleted is the ILM table entry of which the delay time is set after the timer is started currently, and the first delay time is the time from the current refreshing time to the deletion of the ILM table entry to be deleted; restarting the timer by taking the shortest first delay time as the timing time;

and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the ILM table items to be deleted which are not overtime into second delay time, and restarting the timer by taking the shortest delay time in the second delay time as the timing time, wherein the second delay time is the difference between the delay time of the ILM table items to be deleted which are not overtime and the timing time of the current time of the timer.

5. The method according to any of claims 2-4, wherein the setting a delay time for the ILM entry to be deleted comprises:

and adding a delay time item in the ILM table so as to record the delay time of the ILM table entry to be deleted in the delay time item.

6. The method of claim 1, further comprising:

acquiring an ILM table entry to be added, and checking whether a first ILM table entry which is set to be deleted in a delayed manner and is not deleted exists in the ILM table and has the same prefix with the ILM table entry to be added;

if the first ILM table entry does not exist, adding the ILM table entry to be added into the ILM table;

if the first ILM table entry exists, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same, wherein the attributes comprise an operation code and a next hop;

if the ILM table entries to be added are the same, the ILM table entries to be added are not added, and the setting of the delayed deletion of the first ILM table entry is cancelled;

and if not, adding the ILM table entry to be added, and deleting the first ILM table entry.

7. A PE device that is one of two PE devices connected to a CE device in a network of CE dual homed PEs, the PE device having VPN FRR enabled thereon and configuring an address of another PE device connected to the CE device as a standby next hop of a route to the CE device, the device comprising:

a switching processing module, configured to instruct a route updating module to switch a next hop of a route pointing to the CE device to the standby next hop after the detection module detects a link failure; deleting a first private network route, and delaying and deleting an ILM table item corresponding to the first private network route in an incoming label mapping ILM table so as to search a next hop from the un-deleted ILM table item when a message pointing to the CE equipment needs to be forwarded, wherein the first private network route is a Border Gateway Protocol (BGP) private network route learned from the CE equipment; when the first private network route is deleted, sending a standard BGP notification message to a first PE device of an opposite end of the first private network route, wherein the notification message is used for indicating that a second PE device deletes the first private network route;

8. The PE device of claim 7, wherein the switching processing module is specifically configured to search, in the ILM table, an ILM entry to be deleted that corresponds to a prefix of the first private network route; setting delay time for a first batch of ILM entries to be deleted after the link failure, and starting a timer by using the delay time of the first batch of ILM entries to be deleted as timing time, wherein the delay time of the first batch of ILM entries to be deleted is the waiting time of the first batch of ILM entries to be deleted, the waiting time of each ILM entry to be deleted is the same, and the waiting time is the time from the time of setting delay deletion to the time of executing deletion of the ILM entries to be deleted; setting delay time for the non-first ILM entries to be deleted after the link failure, wherein the delay time of the non-first ILM entries to be deleted is the time from the current starting time of a timer to the time for deleting the ILM entries to be deleted; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the non-overtime ILM table items to be deleted, and restarting the timer by taking the shortest delay time in the refreshed delay time as the timing time, wherein the refreshed delay time is the difference between the delay time of the non-overtime ILM table items to be deleted and the timing time overtime of the timer at this time.

9. The PE device of claim 8, wherein the switching processing module is further specifically configured to obtain a waiting time for each to-be-deleted ILM entry according to a scale of a VPN network.

10. The PE device of claim 7, wherein the switching processing module is specifically configured to search, in the ILM table, an ILM entry to be deleted that corresponds to a prefix of the first private network route; acquiring the waiting time of each ILM table item to be deleted, wherein the waiting time is the time from the time of setting delayed deletion to the time of executing deletion of the ILM table item to be deleted; setting delay time for the first ILM list items to be deleted after the link failure, and starting a timer by taking the shortest delay time of the first ILM list items to be deleted as timing time, wherein the delay time of the first ILM list items to be deleted is the waiting time of the first ILM list items to be deleted; setting delay time for the non-first ILM list items to be deleted after the link failure, wherein the delay time of the non-first ILM list items to be deleted is the time from the current starting time of a timer to the time for deleting the ILM list items to be deleted; if the delay time of the first ILM table entry to be deleted is less than the current timing time of the timer, refreshing the delay time of each ILM table entry to be deleted into a first delay time, wherein the first ILM table entry to be deleted is the ILM table entry of which the delay time is set after the timer is started currently, and the first delay time is the time from the current refreshing time to the deletion of the ILM table entry to be deleted; restarting the timer by taking the shortest first delay time as the timing time; and when the timer is overtime, deleting the overtime ILM table items to be deleted in the ILM table, refreshing the delay time of the ILM table items to be deleted which are not overtime into second delay time, and restarting the timer by taking the shortest delay time in the second delay time as the timing time, wherein the second delay time is the difference between the delay time of the ILM table items to be deleted which are not overtime and the timing time of the current time of the timer.

11. The PE device according to any one of claims 8 to 10, wherein the switching processing module is further specifically configured to add a delay time entry in the ILM table, so as to record the delay time of the ILM entry to be deleted in the delay time entry.

12. The PE device of claim 7, wherein the device further comprises:

an adding module, configured to obtain an ILM entry to be added, and check whether a first ILM entry in the ILM table has a same prefix as the ILM entry to be added, where the first ILM entry is an ILM entry in the ILM table that has been set to be deleted in a delayed manner and is not deleted; if the first ILM table entry does not exist, adding the ILM table entry to be added into the ILM table; if the first ILM table entry exists, checking whether the attributes of the ILM table entry to be added and the first ILM table entry are the same, wherein the attributes comprise an operation code and a next hop; if the ILM table entries to be added are the same, the ILM table entries to be added are not added, and the setting of the delayed deletion of the first ILM table entry is cancelled; and if not, adding the ILM table entry to be added, and deleting the first ILM table entry.