KR101291191B1 - Method and system for seamless mobility support using hierarchical home agents - Google Patents

Abstract

A method and system for seamless mobility support using a hierarchical home agent is disclosed. According to the method for seamless mobility support using the hierarchical home agent according to the present invention, the failure of the home agent is not known to the mobile node, thus reducing the burden for the mobile node to detect and recover the failure of its home agent. It minimizes the period of service interruption, requires no change on the mobile node side, and does not need to have multiple home agents in one subnet to handle the failure of the home agent.

Description

Method and system for seamless mobility support using hierarchical home agents}

1 is a diagram illustrating a mobile IPv6 adoption scenario.

2 is a diagram illustrating a corporate domain connected to the Internet through a border gateway.

3 is a diagram illustrating a signal exchange between the home agent information server HIS and the home agent HA when the home agent information server HIS starts to operate.

4 is a diagram illustrating a sequence of various operations that occur when detecting and processing a failure of a home agent (HA).

5 is a diagram illustrating a sequence of various operations that occur when a home agent (HA) recovers from a failure.

6 is a diagram illustrating a sequence of operations that occur when a router starts operation.

FIG. 7 is a diagram illustrating signal exchange between a home agent information server (HIS) and a router when the home agent information server (HIS) starts to operate.

Terms and Definitions Used herein

HA: Home Agent

MIPv6: Mobility support in IPv6

MN: Mobile Node

HIS: Home Agent Information Server

BU: Binding Update

MPS: Mobile Prefix Solicitation

DHAD: Dynamic Home Agent discovery

BG: Border Gateway

ND: Neighbor Discovery

IP: Internet Protocol

IKE: Internet Key Exchange

Ack: Acknowledgment

The present invention relates generally to mobile communication technology. Furthermore, the present invention relates to a hierarchical management domain in mobile communication. More specifically, the present invention relates to methods and systems for seamless mobility support using hierarchical home agents.

Mobile IPv6 is described in references D. Johnson, C. E. Perkins, and J. Arkko, “Mobility Support in IPv6”, RFC 3775, June 2004.

The reference describes a protocol that allows nodes to remain reachable while roaming on the IPv6 Internet. A mobile node is always identified by its home address, regardless of its current point of attachment to the Internet. While away from home, the mobile node registers its address at the remote access point, called the care-of address, with the router on its home link, so that the router can serve as the "home agent" of the mobile node. request. The home agent (HA) is now responsible for forwarding (delivering) packets forwarded to the mobile node's home link to the mobile node's current location.

The mobile node MN will rely on this home agent HA for its connectivity. Thus, the home agent (HA) has the potential to be one point of failure in mobile IPv6. The home agent (HA) may be responsible for a plurality of mobile nodes (MN) in the home link. In this case, a failure of one home agent HA may cause a loss of connection of a plurality of mobile nodes MN located at various locations of the Internet.

To overcome this problem, Mobile IPv6 allows employing a plurality of home agents (HAs) on the home link, so that if a home agent (HA) in service breaks down, another home agent (HA) has failed home It takes the role of the agent (HA), so that it is possible to provide a continuous service to the mobile node (MN) registered to the failed home agent (HA). The move of services from this broken home agent (HA) to a new working home agent (HA) is problematic, and the current Mobile IPv6 specification does not offer a solution to these problems.

1 shows a plurality of home agents HA_x, x = 1, 2,... In the same home link to provide continuous service to the mobile node MN in case the home agent HA providing the service fails. The basic adoption scenario in the case where., n) is present is shown. The mobile node (MN) has a mobile IPv6 mobile node function that provides mobility signaling messages protected by IP security protocol (IPsec). All home agents (HAs) also feature mobile IPv6 home agents by IPsec server software. Initially, the mobile node (MN) registers with HA_1 and has an IPv6 tunnel.

Referring to FIG. 1, consider a situation in which HA_1 fails and a service moves (switching) to a new home agent HA_x (x = 2, ..., n) on the same home link. Describes the fault detection and switching process and related problems.

<Detection of HA Failure>

The service movement from the failed HA_1 to the new HA_x occurs after the failure is detected by the mobile node (MN). The mobile node MN can only detect the failure of HA_1 under certain conditions. These conditions are as follows.

When a mobile node (MN) sends a Binding Updata (BU) message to a failed HA_1 and does not receive a corresponding Binding Acknowledge (BA) message, the mobile node binds until a timeout occurs. Resend updates (BUs). When the timeout occurs, the mobile node (MN) becomes aware of the failure of HA.

Similarly, when the mobile node (MN) sends a Mobile Prefix Solicitation (MPS) message to the failed HA_1 and does not receive the Mobile Prefix Advertisement, it retransmits the MPSs before the timeout occurs. By doing so, it is found that HA_1 has failed.

According to Mobile IPv6, the mobile node (MN) waits for the initial timeout period after sending the first BU or MPS to the failed HA_1, which is INITIAL_BINDACK_TIMEOUT (1 second) for BU and INITIAL_SOLICIT_TIMER (3 for MPS). Seconds). This timeout period is doubled for subsequent BU or MPS messages, increasing until its value is MAX_BINDACK_TIMEOUT (32 seconds). The mobile node MN may send unlimited BUs or MPSs to the failed HA_1 until the last timeout occurs.

Thus, detection of a failed HA_1 can be delayed for a significant time. Also, the BU and MPS are on request rather than periodic. The mobile node (MN) sends a BU message only when registering a new secondary address or extending the registration period to the currently serving HA. Similarly, the mobile node (MN) transmits an MPS message only when the address of the serving HA is invalidated. As a result, the mobile node MN suffers from packet loss and disconnection. This can result in significant performance degradation in real time applications.

<Recovery>

If a failure is detected, on the current Mobile IPv6 specification, the mobile node (MN) attempts to register its secondary address with another home agent (HA) on the home link. For this purpose, the mobile node (MN) must know that other HAs are available on the home link. The mobile node (MN) may initiate the Dynamic Home Agent Discovery (DHAD) protocol and as a result may obtain a list of available home agents on the home link. The mobile node MN may select one HA_x from the list as a serviceable HA. The mobile node MN sends a binding update (BU) message to HA_x to set the home register (H) bit.

However, there is a problem with this recovery operation. If there is only one HA on the home link, according to the current Mobile IPv6 specification, even if the time of retransmission parameter MAX_BINACK_TIMEOUT (32 seconds) is reached, the mobile node (MN) continues to send a BU message to HA_1 to enable valid BA. I try to receive the message, but it doesn't happen to receive the BA message because it is already HA_1. As a result, the mobile node MN enters into an infinite loop.

Even when there are multiple HAs, in addition to fault detection, there is an additional burden on the mobile node (MN) to perform home registration for the new HA, and in some cases multiple home registrations occur due to unsuccessful attempts. Occurs. In addition, if there is no information on the available HA on the home link, the mobile node (MN) performs a DHAD. All these factors place the burden of sending additional messages on the air interface, causing service interruption and burdening the mobile node (MN).

In addition, when HA_1 fails, sequence number information on the binding cache of HA_1 is also lost. The new HA_x that the mobile node MN switches does not know the sequence number of the BU last sent by the mobile node MN. This will be a new security vulnerability on Mobile IPv6.

An object of the present invention is to provide a method and system for seamless mobility support using a hierarchical home agent in mobile IPv6.

Another object of the present invention is to provide a computer-readable recording medium having recorded thereon a program for executing the seamless mobility support method on a computer.

In mobile IPv6, mobile nodes rely on one home agent for seamless roaming on the Internet. The home agent is a router on the mobile node's home network, which tunnels datagrams to be sent to the mobile node when the mobile node is away from home, and has information about the mobile node's current location. The home agent HA may be responsible for the mobility of the plurality of mobile nodes MN on the home link. Failure of one HA can result in the loss of connectivity of many mobile nodes (MNs) located throughout the Internet. The mobile node cannot afford to lose its home agent (HA).

Accordingly, it is a primary object of the present invention to provide a system and method for detecting a failure of a home agent (HA) within a domain and handling failures for mobility support of all mobile nodes performed by the home agent (HA). . A new node called Home Agent Information Server (HIS) is used to achieve this goal.

The present invention proposes a method for handling the failure of a home agent within a hierarchical management domain and thus supporting the mobility of all mobile nodes handled by this home agent.

A new node, called the home agent information server, is added to the domain to detect and handle failures of home agents within the domain. The method according to the invention does not require any change in the function of the mobile node, but requires that all routers within the domain have home agent function.

Method for seamless mobility support using a hierarchical home agent according to the present invention for achieving the above technical problem, (a) adding a home agent information server that is a new server including a database for the home agent in the domain step; (b) When each home agent in the domain starts a new operation, when there is a change in information, or when the home agent information server requests the information, its IP address, list of current prefixes and gateway Sending detailed information to the home agent information server, the detailed information including addresses of all routers in the path to the home agent at; (c) detecting, at the home agent information server, a failure of a router acting as a home agent or a home agent, and substituting the failed home agent or router to routers in the path from the gateway to the failed home agent or router. Handling the failure by requesting to operate as a home agent; (d) detecting, at the home agent information server, the recovery of the failed home agent or router and stopping the home agent or router from operating as a home agent instead of the repaired home agent or router. Requesting to do; (e) transmitting the most recent binding cache information to the recovered home agent or router from the router which was handling the failure of the failed home agent or router; (f) detecting, at the home agent in the domain, the failure of the home agent information server, and stopping communication with the home agent information server; And (g) detecting, at the home agent and router in the domain, the recovery of the home agent information server and resuming communication with the home agent information server.

The method is preferably carried out in a communication system comprising a corporate domain comprising a border gateway, router and home agent used to connect to the Internet.

The database held by the home agent information server may include: an IP address for each home agent in the domain; A list of current prefixes; A list of IP addresses of all routers on the path from the gateway to the home agent; A flag indicating whether the home agent is running or not; And an IP address of a router to operate on behalf of the failed home agent when the home agent is not in operation, and for each router operating as a home agent on behalf of the failed home agent: an IP address of the router; A flag indicating whether the router is operating or not; A list of all failed home agents that the router is acting on behalf of; And an IP address of a router to be operated on behalf of the router when the router is not operating.

The detecting of the failure of the home agent or router may include: (i) at each home agent in the domain, after receiving a "home agent information notification" packet from a home agent information server in the domain, the home agent; Periodically sending a "hello" packet to the information server; (ii) a "hello request" packet requesting the home agent information server to transmit a "hello" packet to the specific home agent if it has not received the "hello" packet from a particular home agent for a configurable time Transmitting; (iii) during the configurable number of times the home agent information server transmits a message requesting the transmission of a "hello" packet to the specific home agent if the home agent information server still fails to receive a "hello" packet from the specific home agent. Repeating; And (iv) determining that the home agent has failed if the home agent information server does not receive a "hello" packet at all after repeatedly transmitting the specific request for the configurable number of times. The time interval at which the hello packet is transmitted is preferably very short so that the failure of the home agent can be detected immediately.

The act of operating the home agent and processing a failure of the home agent may include finding, at the home agent information server, a router closest to the failed home agent in a database, and replacing the failed home agent with the found router as a home agent. Sending a "Home Agent Proxy Request" message requesting to operate; At the router receiving the request, sending a message notifying the home agent information server that the request has been received, and starting to act as a home agent on behalf of the failed home agent; And when the home agent information server does not receive a response from the adjacent router to perform the function of the failed home agent, sending the request to the next level router in the hierarchy. desirable. All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, may observe the binding update (BU) message and the binding notification (Back) message by observing the binding cache of the router. It is desirable to keep an entry. All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, share security between the mobile node and the home agent and are transmitted from the mobile node to the home agent. It is desirable to be able to understand the packets being made. After the router assumes the task of acting as a home agent, it establishes security and communicates with the mobile node in a secure manner using Internet Key Exchange (IKE) with the mobile node of the failed home agent. It is good. If the failed home agent is not on a link to the router learned from the home agent's prefix, the router preferably does not perform a neighbor discovery (ND) proxy function on behalf of the failed home agent. .

In the router operating as a home agent to handle the failure of the home agent, when the home agent information server detects that a router operating as a home agent on behalf of the failed home agent has failed, Sending a "home agent proxy request" requesting to perform the home agent function of the failed home agent, wherein the request includes the current prefix list and internet protocol of all home agents being processed by the failed router; IP) addresses are preferred. If the router accepts the request, the home agent information server preferably updates the address of the new router that accepted the request in a handling router entry that handles the failure for each of the home agents. Detecting and processing a repair of the failed home agent and transmitting the most recent binding cache information to the repaired home agent may include transmitting a "home agent information" packet from the repaired home agent to the home agent information server. step; Sending, at the home agent information server, a "stop home agent proxy" message to the handling router that handles the failure of the home agent requesting to stop acting as a home agent on behalf of the restored home agent; Sending, at the handling router, a "binding cache information" message containing the most recent binding cache information to the recovered home agent; Sending, by the restored home agent, a "binding cache information notification" message to the router which sent the "binding cache information" message; Sending, at the handling router, a "home agent proxy stop notification" message to the home agent information server indicating the success or failure of the requested function; And updating, at the home agent information server, the contents of the database for the restored home agent in operation when receiving a message indicating success from the router.

Detecting recovery of the failed router includes: sending a "ready" message from the router to the home agent information server; Sending, at the home agent information server, a "ready notification" message confirming receipt of the "ready" message; Generating, at the home agent information server, a list of the failed home agent; Confirming that the recovered router is on a path to the gateway; Ascertaining whether the router currently processing the failed home agent is higher on the layer than the restored router; Sending a message from the home agent information server to a handling router that handles the failed home agent requesting to stop acting as a home agent on behalf of the home agent, and to the restored router the most recent of the home agent; Sending a binding cache of the to operate on behalf of the failed home agent; And transmitting, at the handling router, a response indicating success or failure of the requested operation to the home agent, after the restored router receives the binding cache information and, on the behalf of the failed home agent, It is desirable to start acting as an agent. The detecting of the failure of the home agent information server in the home agent may include: periodically transmitting a "home agent information server hello request" message from each home agent to the home agent information server; Transmitting, by the home agent information server, a "home agent information server hello" message in response to the request message; And if the home agent does not receive a response to the request from the home agent information server, determining that the home agent information server has failed and stopping communication with the home agent information server.

Detecting the recovery of the failed home agent information server at the home agent comprises sending a "home agent information request" message from the home agent information server requesting their information to all home agents in the domain; When the home agent receives the message, the home agent determines that the home agent information server has been restored, and transmits a "home agent information" message and a "hello" message to the home agent information server to communicate with the home agent information server. It is desirable to resume.

The detecting of the failed home agent information server at the router may include: transmitting a "home agent information server ready" message from the home agent information server to all routers in a domain; Transmitting a "home agent information server preparation response" message from the router to the home agent information server; At the home agent information server, upon receiving the response message, filling in a database with detailed information of routers acting as the home agent; And transmitting, from the home agent information server, a "home agent information server preparation response notification" message confirming receipt of the "home agent information server preparation response" message to routers operating as the home agent. And when the router operates as a home agent on behalf of the failed home agent, the response message is a message including an IP address and a current prefix list of each of the failed home agents, and the router homes on behalf of the failed home agent. If not acting as an agent, the response message is preferably an empty message confirming receipt of the " home agent information server ready " message to the home agent information server.

A system for seamless mobility support using a hierarchical home agent for achieving the technical problem comprises: (a) a home agent information server including a database for home agents in a domain and added as a new server; And (b) when each home agent in the domain starts a new operation, when there is a change in information, or when the home agent information server requests the information, its IP address, a list of current prefixes, and Each home agent sending detailed information including addresses of all routers on the path from the gateway to the home agent to the home agent information server, wherein the home agent information server operates as a home agent or home agent. Detects a failure of the router and handles the failure by requesting routers in the path from the gateway to the failed home agent or router to operate as a home agent instead of the failed home agent or router. The home agent information server is the broken home agency. Or detect a repair of the router and request the router that is handling the failure of the home agent or router to stop acting as a home agent instead of the restored home agent or router, and detect the failure of the failed home agent or router. The processed router sends the most recent binding cache information to the recovered home agent or router, and the home agent in the domain detects a failure of the home agent information server and communicates with the home agent information server. It is preferable that the home agent and the router within the domain detect the recovery of the home agent information server and resume communication with the home agent information server.

The system is preferably employed in a communication system that includes a corporate domain including border gateways, routers, and home agents used to connect to the Internet.

Preferred embodiments of the present invention are now described in detail with reference to the accompanying drawings. However, it is to be understood that the disclosed embodiments are merely examples of the invention and may be embodied in various forms. The following description and drawings are not to be construed as limiting the invention, and various details are provided to provide a thorough understanding of the invention, which are the basis of the claims and those of ordinary skill in the art to which the invention pertains. It is the basis for teaching the students how to make and / or use the invention. However, in order to avoid unnecessarily obscuring the details of the present invention, in some cases a detailed description of what is well known or common is not described.

1 shows the components on a mobile IPv6 network and how they are connected. In this figure, the home link is the link to which the home subnet prefix is assigned, and the mobile node obtains its home address using the home subnet prefix. The home agent is on the home link. In case of some HA failure, there are a plurality of HAs in the home link. In FIG. 1, HA1, HA2, HA3 are home agents. External links are links that are not home links of mobile nodes. When the mobile node is out of home, it registers its primary secondary address with the home agent.

2 is a reference scenario of the proposed method. The corporate domain is shown connected to the Internet 40 via a Border Gateway (BG) 50. Inside the domain there is shown a new node called routers, home agents and home agent information server (HIS) 60, which is intended to achieve the purpose of the present invention.

3 shows signal transmission between the HIS and the HA when the HIS starts to operate. The HIS starts the first operation or recovers from the failure, and then transmits a "HA information request" message to all HAs (S100). All HAs transmit a "home agent information" message to the HIS (S110). Upon receiving this information, the HIS transmits a "Home Agent Information Notification" message, which is a message confirming receipt, to corresponding HAs (H120). The HAs notified start periodically transmitting a "hello" message to the HIS (S130).

4 shows a sequence of operations during detection and processing of a failure of an HA. Before the HA fails, the HA transmits a "hello" message to the HIS (S200). If the HA fails, the HIS does not receive a "hello" message from the HA (S210). When the timeout elapses, the HIS sends a "hello request" message to the HA to request a "hello" message (S220). Since the HA has failed, the HIS does not receive any response from the HA (S230). The HIS now transmits an "HA proxy request" message to the router closest to the failed HA and requests to operate as an HA instead of the broken HA (S240). When the router receives this request, the router transmits an "HA proxy notification" message to the HIS (S250), and starts to operate as an HA on behalf of the failed HA, and transmits a "hello" message to the HIS (S260).

5 is a diagram illustrating a sequence of various operations that occur when HA starts operation. When HA starts for the first time or recovers from a failure, it always transmits a "HA information" message to the HIS (S300). The HIS transmits a "HA proxy stop" message to the router which was acting as an HA instead of the repaired HA when it failed, so as to stop operating as HA (S310). Upon receiving this message, the router transmits a "binding cache information (Bcache) information" message including corresponding binding cache information to the recovered HA (S320). When the HA receives this information, the HA transmits a confirmation message of "binding cache information (Bcache) information notification" (S330). When the router receives the notification message from the HA, the router transmits an "HA proxy stop notification" message to the HIS (S340).

6 is a diagram illustrating a sequence of operations that occur when a router starts operation. When the router starts operation for the first time or recovers from a failure, it always transmits a "ready" message to the HIS (S400). The HIS transmits a "ready notice" message to the router indicating that it has received a "ready" message (S410).

After sending the notification message, the HIS generates a list of all failed HAs that the recovered router must handle. The HIS transmits an "HA proxy stop" message to a router (Handling Router) currently handling the failed HAs, and requests to delegate the task of acting as an HA to a recovered router (S420). The handling router (Handling Router) transmits a "binding cache information (Bcache)" message including the binding cache information for the failed HA to the recovered router (S430). When the restored router receives the binding cache information, it transmits a confirmation notification message to the handling router (S440). The handling router transmits an "HA proxy stop notification" message to the HIS (S450).

FIG. 7 is a diagram illustrating signal exchange between a home agent information server (HIS) and a router when the home agent information server (HIS) starts to operate. When the HIS starts a new operation or recovers from a failure, the HIS transmits a "HIS ready" message to all routers in the domain (S500 and S505). The router sends a "router ready response" message to the HIS, which includes information on whether the router is operating as an HA on behalf of the failed HA (S510 and S520). Upon receiving this message, the HIS updates its database for the router operating as HA and transmits a confirmation notification message to the router operating as HA (S530). When the HA operating as the HA receives this notification message, it starts to transmit a "hello" message to the HIS (S540).

The operation of the present invention will be described in more detail with reference to the drawings.

2 shows an example scenario in which the proposed method is applicable.

The system shown in FIG. 2 includes a hierarchical management domain connected to the Internet using a border gateway 50. Inside the management domain, a router, a home agent, and a home agent information server 60 are shown. The home agent information server 60 is for achieving the object of the present invention.

The operation of the present invention is as follows.

○ First, a new server called Home Agent Information Server must be added to the domain. This node handles the failure of the home agent inside the domain.

The HIS has a database of information about home agents in the domain, which includes the IP address for each home agent, the current prefix list, and the addresses of all routers from the HA to the border gateway of the domain.

Each HA should transmit the information to the HIS when it first starts its operation, when there is a change in the above information, or when a "HA information request" message is received from the HIS.

○ The information is included in the "HA information" message and delivered.

○ Upon receipt of the "HA Information" message, the HIS shall send a "HA Information Notice" message as a receipt confirmation notice.

If the HA does not receive the notification message, it may retransmit the information as many times as possible and if it does not receive the notification message until the end, it may determine that the HIS is missing or has failed. At this time, the HA stops communicating with the HIS.

When the HIS starts to operate inside the domain, the HIS can resume communication with the HA by sending a "Request for HA Information" message to all HAs within the domain. For this purpose, the HIS should be configured to have the addresses of all HAs in the domain.

<Home Agent Failure Detection>

After each home agent in the domain receives a "HA Information Notification" message packet from the HIS, it starts periodically sending a "hello" packet to the HIS located inside the domain.

If the HIS does not receive a "hello" packet for a configurable time from a particular home agent, it sends a "hello request" packet to the HA to request that the "hello" packet be sent.

If the HIS does not receive a "hello" packet, it repeats the process of sending a "hello request" packet to the HA for a configurable number of times.

If the HIS does not receive a "hello" packet after every attempt, it determines that the HA has failed and proceeds with the actions necessary to handle it.

○ The time interval during which hello packet is transmitted by HA should be set very short so that failure of HA can be detected immediately.

<Handling Home Agent Failure>

By observing BU and Back packets sent to or from the HA, all routers on the path to the HA inside the domain have binding cache entries for each HA.

In addition, the security policies and security policies maintained by the HA are shared by all routers along the path from the gateway to the HA.

If the HIS detects a failure of the HA (assuming, for example, HA1 in Figure 2 is broken), it indicates that the failure is on the entry for that HA in the database, and finds the nearest router of that HA in the database. (R1 in the example of FIG. 2), the router R1 sends a " HA Proxy Request " message requesting to operate as HA instead of a failed HA.

The "HA Proxy Request" message contains the current prefix list and IP address of the failed HA. This information is stored in the HA information database maintained by the HIS.

O The router R1 receiving the request sends a " HA Proxy Notification " message to the HIS as a confirmation notification.

If the HIS does not receive a response from the router, it resends the request as many times as possible.

If no response is received despite all attempts, the HIS finds the next higher router (in the example R2 in Figure 2) for HA (in the hierarchy) and sends the request to that router.

This process is repeated for all routers on the path to the HA within the domain and ends when the notification message is received or when no notification message is received for all routers.

If a router (eg R1) accepts a request to act as an HA, the router has security policies or policies for the already bound HA's binding cache entries and all mobile nodes that were handled by the failed HA. Therefore, it will act as HA instead of a failed HA.

Before starting to communicate with the mobile node MN, the router R1 establishes security matters with the mobile node MN using IKE (Internet key Exchange). After establishing the security matters, the router R1 can securely send packets to the mobile node MN.

Since the router R1 already knows the security policy between the failed HA and the MN, the router R1 can receive and understand the packet transmitted from the mobile node MN.

Although the failed HA is not on the link to the router learned from the HA's prefix, the router does not need to perform the neighbor discovery (ND) proxy function on behalf of the failed HA.

Since router R1 is operating on behalf of HA, the router should start sending "hello" packets to the HIS.

After receiving an "HA Proxy Notification" from the router (R1, R2, or BG), the HIS shall update the record of the failed HA by adding the address of the router that handles the failure in the database.

The HIS must also add a new record to the HA information database for router R1 that the router is operating as HA. HIS adds an IP address field to the database and marks it in the field that indicates it is running.

The HIS should now periodically receive a "hello" packet from the router R1.

<Handling Router Failure>

If the HIS detects a failure of a router (for example, R1) operating as HA on behalf of the failed HA (HA1, HA2), it takes over the HA function of the failed HA (HA1, HA2) to the adjacent router (R2). Send an "HA proxy request" to ask to give.

This request message contains the current prefix list and IP address for all HAs (HA1, HA2) that were processed by the failed router (R1).

When router R2 accepts the request, the HIS updates the handling router entry for each HA HA1, HA2 with the address of the new router R2.

<Home Agent Database Entries>

The HA information database entries maintained by the HIS are as follows.

O For each home agent in the domain includes:

● IP address

● list of current prefixes

A list of IP addresses of all routers on the route from the gateway to the home agent

A flag indicating whether the home agent is running or not

IP address of a router to act on (handling) if the home agent is not running

O For each router acting as a home agent on behalf of the failed home agent, includes:

● IP address of the router

A flag indicating whether the router is operating or not

A list of all failed home agents that are acting as home agents on behalf of the router;

● IP address of the router to act on behalf of the router if it is not running

<Repair of a Failed Home Agent>

When a home agent HA1 starts to function again, the HA sends a "HA Information" packet to the HIS.

On receipt of this packet, the HIS checks to see if the packet came from a broken HA. If the packet is from a failed HA, it sends a "HA Proxy Stop" request to the router (e.g. R1) that is handling the failure of that HA to stop acting as a HA on behalf of the recovered HA. The request includes the IP address and prefix list of the recovered HA.

Upon receipt of this request, router R1 sends a " Bcache Information " packet containing the binding cache information corresponding to the HA to the HA.

When the HA receives this information, it sends a "Bcache Information Notification" message confirming this to Router R1 and starts operating as HA.

Upon receipt of the notification message from HA1, Router R1 sends a "HA Proxy Suspension Notification" message to the HIS indicating the failure or success of sending the information to the HA.

When the HIS receives a message indicating success, it deletes HA1 from the list of failed HAs being processed by router R1. If the list of failed HAs being processed by the router is empty and it is not HA itself, then the entry for the router is deleted from the database.

The HIS also deletes the handling router entry for the recovered HA and indicates that the HA is active.

<Repair of a Failed Router>

When a failed router (eg R1) starts working again, it sends a "ready" message to the HIS.

○ The HIS shall send a "Ready Notice" message to Router R1 confirming that it has received a "Ready" message.

After the HIS sends a "Ready Notify" message, for a failed HA in the database, the router (R1) has not been on the path from the HA to the gateway, and the router that handles the currently broken HA has been recovered. Check that is not the ancestor side (higher on the hierarchy).

The HIS creates a list of failed HAs (HA1, HA2) and sends a "HA Proxy Stop" request to the handling routers of the failed HAs (eg R2). This message contains the IP address and current prefix list of the failed HAs HA1 and HA2, and the address of the recovered router R1.

The handling router R2 sends binding cache information for the HAs HA1 and HA2 included in the request to the restored router R1.

Upon receiving the notification, the handling router R2 sends a "HA proxy suspension notification" to the HIS.

After receiving this notification, the HIS deletes the list of the created broken HAs HA1 and HA2 from the list of broken HAs processed by this router R2. If the list of failed HAs handled by router R2 is empty and it is not HA itself, then the entry for router R2 is removed from the database.

HIS also updates the address of the handling router for the failed HAs HA1, HA2 with the address of the recovered router R1.

<Failure detection of home agent information server (HIS)>

In order to detect a failure of the HIS, each HA shall periodically send a special packet to the HIS called a "HIS Hello Request".

○ In response, the HIS shall send a message of "HIS Hello" to the HA.

Upon receipt of this response, the HA knows that the HIS is present and operating normally.

If the HA does not receive a "HIS Hello" response, it resends the request to the HIS a configurable number of times.

If the HA does not receive a response after every attempt, it determines that the HIS is broken and stops communicating with the HIS.

When the HA receives the "HA Information Request" packet from the HIS, it resumes communication with the HIS.

Recovery Detection of Home Agent Information Server (HIS) by Router>

When the HIS is restored or first started, it sends a "HIS Ready" message to all routers in the domain.

If the router is acting as a HA on behalf of the failed HA, send a "HIS Ready Response" message containing the IP address and current prefix list for each failed HA.

If the router is not a router acting as an HA on behalf of a failed HA, it simply sends a "HIS Ready Response" message, a null message confirming that it has received a "HIS Ready" message.

When this response is received, the HIS populates the HA information database with the details of the routers acting as HA.

After receiving the response from the router, the HIS sends a "HIS Ready Response Notification" message to the router.

The present invention can be embodied as code that can be read by a computer (including all devices having an information processing function) in a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of computer-readable recording devices include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like.

The descriptions given above are for the best mode contemplated for carrying out the invention. The manners and processes of making and using the same are described in abundant, clear, concise and accurate terms, allowing anyone skilled in the art to make and use the invention. In particular new embodiments may be created within the scope of the invention as other examples of ways intentionally combined with others.

However, the present invention is exposed to modifications or alternative arrangements that are sufficiently equivalent to the above disclosure. Accordingly, the particular embodiments disclosed are not intended to limit the invention. On the contrary, the intention is to cover all modifications and alternative arrangements within the spirit and scope of the present invention as expressed generally by the following claims, which particularly point out and clearly claim the technical features of the present invention.

Although the foregoing description has been focused on the novel features of the invention as applied to various embodiments, those skilled in the art will appreciate that the apparatus and method described above without departing from the scope of the invention. It will be understood that various deletions, substitutions, and changes in form and detail of the invention are possible. Accordingly, the scope of the present invention is defined by the appended claims rather than the foregoing description. All variations within the scope of the appended claims are embraced within the scope of the present invention.

a. The failure of the home agent HA is transparent to the MN, thus relieving the mobile node MN of the burden of detecting and recovering from the failure of its home agent HA.

b. Minimize the period of service interruption.

c. No change is required on the mobile node (MN) side.

d. It is not necessary to have a plurality of home agents HA in one subnet in order to deal with the failure of the home agents HA.

Claims (31)

In the method for seamless mobility support using a hierarchical home agent, (a) adding a home agent information server, which is a new server containing a database for home agents within a domain; (b) When each home agent in the domain starts a new operation, when there is a change in information, or when the home agent information server requests the information, its IP address, list of current prefixes and gateway Sending detailed information to the home agent information server, the detailed information including addresses of all routers in the path to the home agent at; (c) detecting, at the home agent information server, a failure of a router acting as a home agent or a home agent, and substituting the failed home agent or router to routers in the path from the gateway to the failed home agent or router. Handling the failure by requesting to operate as a home agent; (d) detecting, at the home agent information server, the recovery of the failed home agent or router and stopping the home agent or router from operating as a home agent instead of the repaired home agent or router. Requesting to do; (e) transmitting the most recent binding cache information to the recovered home agent or router from the router which was handling the failure of the failed home agent or router; (f) detecting, at the home agent in the domain, the failure of the home agent information server, and stopping communication with the home agent information server; And (g) detecting, at the home agent and router in the domain, the recovery of the home agent information server and resuming communication with the home agent information server. The database of claim 1, wherein the database held by the home agent information server comprises: For each home agent in the domain: IP address; A list of current prefixes; A list of IP addresses of all routers on the path from the gateway to the home agent; A flag indicating whether the home agent is running or not; And If the home agent is not running and includes the IP address of the router to operate on behalf of, For each router acting as a home agent on behalf of the failed home agent: An IP address of the router; A flag indicating whether the router is operating or not; A list of all failed home agents that the router is acting on behalf of; And Seamless mobility support method characterized in that it comprises the IP address of the router to operate in place if the router is not in operation. The method of claim 1, wherein the detecting of the failure of the home agent or the router comprises: (i) at each home agent in the domain, periodically receiving a "home agent information notification" packet from a home agent information server in the domain, and then periodically sending a "hello" packet to the home agent information server ; (ii) a "hello request" packet requesting the home agent information server to transmit a "hello" packet to the specific home agent if it has not received the "hello" packet from a particular home agent for a configurable time Transmitting; (iii) during the configurable number of times the home agent information server transmits a message requesting the transmission of a "hello" packet to the specific home agent if the home agent information server still fails to receive a "hello" packet from the specific home agent. Repeating; And (iv) determining that the home agent has failed when the home agent information server does not receive a "hello" packet at all even after repeatedly transmitting the specific request for the configurable number of times; And a time interval at which the home agent transmits the hello packet enables instant detection of a home agent failure. The method of claim 1, wherein the acting as the home agent to handle the failure of the home agent, At the home agent information server, finding a router closest to the failed home agent in a database, and sending a "home agent proxy request" message to the found router requesting to act as a home agent on behalf of the failed home agent; ; At the router receiving the request, sending a message notifying the home agent information server that the request has been received, and starting to act as a home agent on behalf of the failed home agent; And At the home agent information server, if a response is not received from the adjacent router for performing the function of the failed home agent, sending the request to the next level router in the hierarchy. Seamless mobility support method. 5. The method of claim 4, All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, may observe the binding update (BU) message and the binding notification (Back) message by observing the binding cache of the router. A method of seamless mobility support, characterized by maintaining an entry. 5. The method of claim 4, All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, share security between the mobile node and the home agent and are transmitted from the mobile node to the home agent. Seamless mobility support method, characterized in that the packet can be understood. 5. The router of claim 4, wherein after the router assumes the task of acting as a home agent, it establishes security matters using Internet Key Exchange (IKE) with the mobile node of the failed home agent and in a secure manner. Seamless mobility support method, characterized in that the communication with the node. 5. The router of claim 4, wherein if the failed home agent is not on a link to a router learned from the home agent's prefix, the router performs neighbor discovery (ND) proxy functionality on behalf of the failed home agent. Seamless mobility support method, characterized in that not performed. The method of claim 1, wherein acting as a home agent in the router to handle failure of the home agent comprises: When the home agent information server detects that a router operating as a home agent on behalf of the failed home agent has failed, a "home agent proxy request" requesting an adjacent router to perform a home agent function of the failed home agent. Sending a message, And the request includes a current prefix list and internet protocol (IP) addresses of all home agents being processed by the failed router. 10. The method of claim 9, wherein when the router accepts the request, the home agent information server updates the address of the new router that accepted the request in a handling router entry that handles the failure for each of the home agents. Seamless mobility support method. The method of claim 1, wherein detecting and processing the repair of the failed home agent and sending the most recent binding cache information to the restored home agent comprises: Transmitting a "home agent information" packet from the recovered home agent to the home agent information server; Sending, at the home agent information server, a "stop home agent proxy" message to the handling router that handles the failure of the home agent requesting to stop acting as a home agent on behalf of the restored home agent; Sending, at the handling router, a "binding cache information" message containing the most recent binding cache information to the recovered home agent; Sending, by the restored home agent, a "binding cache information notification" message to the router which sent the "binding cache information" message; Sending, at the handling router, a "home agent proxy stop notification" message to the home agent information server indicating the success or failure of the requested function; And And at the home agent information server, updating the contents of the database for the restored home agent in operation upon receiving a message indicating success from the router. The method of claim 1, wherein the detecting of the failed router comprises: Sending a "preparation" message from the router to the home agent information server; Sending, at the home agent information server, a "ready notification" message confirming receipt of the "ready" message; Generating, at the home agent information server, a list of the failed home agent; Confirming that the recovered router is on a path to the gateway; Ascertaining whether the router currently processing the failed home agent is higher on the layer than the restored router; Sending a message from the home agent information server to a handling router that handles the failed home agent requesting to stop acting as a home agent on behalf of the home agent, and to the restored router the most recent of the home agent; Sending a binding cache of the to operate on behalf of the failed home agent; And Sending, at the handling router, a response indicating success or failure of the requested operation to the home agent, And after the restored router receives the binding cache information, it starts to act as a home agent on behalf of the failed home agent. The method of claim 1, wherein the detecting of the failure of the home agent information server in the home agent comprises: Periodically transmitting a "home agent information server hello request" message from each home agent to the home agent information server; Transmitting, by the home agent information server, a "home agent information server hello" message in response to the request message; And If the home agent does not receive a response to the request from the home agent information server, determining that the home agent information server has failed and stopping communication with the home agent information server; How to apply. The method of claim 1, wherein the detecting of the failed home agent information server by the home agent comprises: Sending, from the home agent information server, a "home agent information request" message requesting their information to all home agents in the domain, When the home agent receives the message, the home agent determines that the home agent information server has been restored, and transmits a "home agent information" message and a "hello" message to the home agent information server to communicate with the home agent information server. Seamless mobility support method, characterized in that resume. The method of claim 1, wherein the detecting of the failed home agent information server by the router comprises: Sending a "home agent information server ready" message from the home agent information server to all routers in the domain; Transmitting a "home agent information server preparation response" message from the router to the home agent information server; At the home agent information server, upon receiving the response message, filling in a database with detailed information of routers acting as the home agent; And Sending, from the home agent information server, a "home agent information server preparation response notification" message confirming receipt of the "home agent information server preparation response" message to routers operating as the home agent, If the router operates as a home agent on behalf of the failed home agent, the response message is a message including an IP address and a current prefix list of each of the failed home agents, If the router does not act as a home agent on behalf of the failed home agent, the response message is an empty message confirming receipt of the "home agent information server ready" message only to the home agent information server. How to support seamless mobility. In a system for seamless mobility support using a hierarchical home agent, (a) a home agent information server containing a database of home agents within the domain and added as a new server; And (b) When each home agent in the domain starts a new operation, when there is a change in information, or when the home agent information server requests the information, its IP address, list of current prefixes and gateway Each home agent transmits detailed information including the addresses of all routers in the path to the home agent to the home agent information server in, The home agent information server detects a failure of a home agent or a router acting as a home agent and operates as a home agent instead of the failed home agent or router to routers on the path from the gateway to the failed home agent or router. To handle the failure by requesting The home agent information server detects the repair of the failed home agent or router and requests the router that is handling the failure of the home agent or router to stop acting as a home agent instead of the restored home agent or router. , The router which handled the failure of the failed home agent or router transmits the most recent binding cache information to the recovered home agent or router. The home agent in the domain detects a failure of the home agent information server, stops communication with the home agent information server, And a home agent and a router in the domain detect a recovery of the home agent information server and resume communication with the home agent information server. The database of claim 16, wherein the database held by the home agent information server comprises: For each home agent in the domain: IP address; A list of current prefixes; A list of IP addresses of all routers on the path from the gateway to the home agent; A flag indicating whether the home agent is running or not; And If the home agent is not running and includes the IP address of the router to operate on behalf of, For each router acting as a home agent on behalf of the failed home agent: An IP address of the router; A flag indicating whether the router is operating or not; A list of all failed home agents that the router is acting on behalf of; And Seamless mobility support system, characterized in that it includes the IP address of the router to operate in place if the router is not in operation. 17. The method of claim 16, At each home agent in the domain, after receiving a "home agent information notification" packet from a home agent information server in the domain, periodically sending a "hello" packet to the home agent information server, If the home agent information server does not receive the "hello" packet from a specific home agent for a configurable time, it transmits a "hello request" packet requesting to transmit a "hello" packet to the specific home agent. , If the home agent information server fails to continuously receive a "hello" packet from a specific home agent, repeating the transmission of a message requesting to transmit a "hello" packet to the specific home agent for a set number of times, If the home agent information server does not receive a "hello" packet at all even after repeatedly transmitting the specific request for the set number of times, it is determined that the home agent has failed. And a time interval during which the home agent transmits the hello packet enables instant detection of a failure of the home agent. 17. The method of claim 16, In the home agent information server, find a router closest to the failed home agent in a database, and send a "home agent proxy request" message to the found router requesting to act as a home agent on behalf of the failed home agent, At the router receiving the request, sending a message notifying the home agent information server that the request has been received, and starting to act as a home agent on behalf of the failed home agent; If the home agent information server does not receive a response from the adjacent router to perform the function of the failed home agent, the process of sending the request to the next level router in the hierarchy is seamless. Mobility Support System. 20. The method of claim 19, All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, bind the router's binding cache by observing binding update (BU) messages and binding notification (Back) messages. A seamless mobility support system characterized by maintaining an entry. 20. The method of claim 19, All routers on the path from the gateway to the failed home agent, including routers operating on behalf of the failed home agent, share security between the mobile node and the home agent and are transmitted from the mobile node to the home agent. A seamless mobility support system, characterized by being able to understand the packets being made. 20. The method of claim 19, wherein after the router assumes the task of acting as a home agent, it establishes security matters using Internet Key Exchange (IKE) with the mobile node of the failed home agent and in a secure manner. Seamless mobility support system, characterized in that the communication with the node. 20. The router of claim 19, wherein if the failed home agent is not on a link to a router learned from the home agent's prefix, the router performs neighbor discovery (ND) proxy functionality on behalf of the failed home agent. Seamless mobility support system, characterized in that it does not perform. 17. The method of claim 16, When the home agent information server detects that a router operating as a home agent on behalf of the failed home agent has failed, a "home agent proxy request" requesting an adjacent router to perform a home agent function of the failed home agent. , And the request includes a current prefix list and internet protocol (IP) addresses of all home agents being processed by the failed router.  17. The method of claim 16, wherein when the router accepts the request, the home agent information server updates the address of the new router that accepted the request in a handling router entry that handles the failure for each of the home agents. Seamless mobility support system. 17. The method of claim 16, Transmitting a "home agent information" packet from the recovered home agent to the home agent information server, Sending, from the home agent information server, a "stop home agent proxy" message to the handling router handling the failure of the home agent requesting to stop acting as a home agent on behalf of the restored home agent; Sending, from the handling router, a "binding cache information" message containing the most recent binding cache information to the recovered home agent, In the repaired home agent, transmit a "binding cache information notification" message to the router that sent the "binding cache information" message, Sending, from the handling router, a "home agent proxy stop notification" message to the home agent information server indicating the success or failure of the requested function, And at the home agent information server, updating the contents of the database for the restored home agent in operation upon receiving a message indicating success from the router. 17. The method of claim 16, Send a "ready" message from the router to the home agent information server, Sending, from the home agent information server, a "ready notification" message confirming receipt of the "ready" message, Generating a list of the failed home agent in the home agent information server, Confirm that the recovered router is on the path to the gateway, Confirm that the router currently processing the failed home agent is higher on the layer than the recovered router, Sending a message from the home agent information server to a handling router that handles the failed home agent requesting to stop acting as a home agent on behalf of the home agent, and to the restored router the most recent of the home agent; Sends a binding cache of to operate on behalf of the failed home agent, At the handling router, send a response indicating the success or failure of the requested operation to the home agent, And the restored router begins to act as a home agent on behalf of the failed home agent after receiving the binding cache information. 17. The method of claim 16, Each home agent periodically sends a "home agent information server hello request" message to the home agent information server, The home agent information server transmits a "home agent information server hello" message in response to the request message, And if the home agent does not receive a response to the request from the home agent information server, determines that the home agent information server has failed, and stops communication with the home agent information server. 17. The method of claim 16, The home agent information server sends a "home agent information request" message requesting their information to all home agents in the domain, When the home agent receives the message, the home agent determines that the home agent information server has been restored, and transmits a "home agent information" message and a "hello" message to the home agent information server to communicate with the home agent information server. Seamless mobility support system, characterized in that resume. 17. The method of claim 16, The home agent information server transmits a "home agent information server ready" message to all routers in the domain, Sending a "home agent information server preparation response" message from the router to the home agent information server, Upon receiving the response message, the home agent information server fills in a database with detailed information of routers acting as the home agent, Sending, from the home agent information server, a "home agent information server preparation response notification" message confirming reception of the "home agent information server preparation response" message to routers operating as the home agent, If the router operates as a home agent on behalf of the failed home agent, the response message is a message including an IP address and a current prefix list of each of the failed home agents, If the router does not act as a home agent on behalf of the failed home agent, the response message is an empty message confirming receipt of the "home agent information server ready" message only to the home agent information server. Seamless mobility support system. (a) adding a home agent information server, which is a new server that includes a database for home agents within a domain; (b) When each home agent in the domain starts a new operation, when there is a change in information, or when the home agent information server requests the information, its IP address, list of current prefixes and gateway Sending detailed information to the home agent information server, the detailed information including addresses of all routers in the path to the home agent at; (c) detecting, at the home agent information server, a failure of a router acting as a home agent or a home agent, and substituting the failed home agent or router to routers in the path from the gateway to the failed home agent or router. Handling the failure by requesting to operate as a home agent; (d) detecting, at the home agent information server, the recovery of the failed home agent or router and stopping the home agent or router from operating as a home agent instead of the repaired home agent or router. Requesting to do; (e) transmitting the most recent binding cache information to the recovered home agent or router from the router which was handling the failure of the failed home agent or router; (f) detecting, at the home agent in the domain, the failure of the home agent information server, and stopping communication with the home agent information server; And (g) at a home agent and a router in the domain, detecting a recovery of the home agent information server and resuming communication with the home agent information server may be read by a computer having recorded a program for executing on the computer. Recording media.

Images