WO1998049620A1 - Reprise par un serveur redondant, dans un environnement en reseau - Google Patents
Reprise par un serveur redondant, dans un environnement en reseau Download PDFInfo
- Publication number
- WO1998049620A1 WO1998049620A1 PCT/US1998/008142 US9808142W WO9849620A1 WO 1998049620 A1 WO1998049620 A1 WO 1998049620A1 US 9808142 W US9808142 W US 9808142W WO 9849620 A1 WO9849620 A1 WO 9849620A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- network
- server
- address
- physical address
- standby server
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
- G06F11/2023—Failover techniques
- G06F11/2028—Failover techniques eliminating a faulty processor or activating a spare
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
- G06F11/2035—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant without idle spare hardware
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
- G06F11/2038—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant with a single idle spare processing component
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/2097—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements maintaining the standby controller/processing unit updated
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/40—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L2101/00—Indexing scheme associated with group H04L61/00
- H04L2101/60—Types of network addresses
- H04L2101/618—Details of network addresses
- H04L2101/622—Layer-2 addresses, e.g. medium access control [MAC] addresses
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L61/00—Network arrangements, protocols or services for addressing or naming
- H04L61/09—Mapping addresses
- H04L61/10—Mapping addresses of different types
- H04L61/103—Mapping addresses of different types across network layers, e.g. resolution of network layer into physical layer addresses or address resolution protocol [ARP]
Definitions
- This invention relates to distributed computing environments and in particular to methods and associated apparatus for activating a standby server upon takeover from a failed server in a redundant server environment with minimal delays imposed in client services and applicable to a wide variety of network topologies.
- Distributed computing environments are those in which client processes request services from server processes.
- distributed computing environments are often used in conjunction with network communication media and protocols to enable distribution of the various communicating processes across physically remote nodes and locations.
- a server process may, for example be operable in a server computing node while a client process which requests services from the server process may be operable in the same node or in a remote node connected via communication networks.
- Computing nodes in a distributed computing environment are connect to the network communication medium via network adapters or network interface cards (also referred to herein as NICs).
- NICs network interface cards
- such a computing system may include general purpose computer systems (e.g., host systems) into which a NIC is inserted as well as peripheral devices with embedded NIC circuits which attach the peripheral to the network medium.
- a server or service process may be, for example, a file server providing coordinate access to files on behalf of a plurality of client processes, a print server providing printer functions to a plurality of client process, or any other function which provides services on behalf of a client process.
- a server process may therefore be operable within a general purpose computing node or may be embedded within a special purpose server device.
- a client process is therefore any process which requests such services from a server process whether operable in a general purpose computing environment or embedded in a special purpose device.
- Protocols used in network communication are often modeled as layers of modules.
- the lowest level layers manage the lowest levels of processing required to apply signals to the network communication medium destined to another node.
- Higher level modules manage correspondingly higher level functions relating to transferring information between two processes (e.g., routing of messages between a client node and a server node or error recovery retransmissions).
- the NIC has an address associated therewith often referred to as a media access address or MAC address (also referred to herein as a physical address).
- MAC address also referred to herein as a physical address.
- This address is programmed into the NIC typically at time of manufacture in accordance with industry standards which help assure a globally unique address is assigned each NIC.
- the MAC address is used to uniquely identify and distinguish the communicating nodes on the network medium.
- Blocks (packets) of information applied to the communication medium typically include address fields which identify the NIC which is the source of the packet and the NIC which is the destination of the packet.
- the MAC address of the NIC is therefore physical in nature in that it identifies the NIC globally regardless of the particular system in which it is operating.
- Higher level protocols utilize other addresses (also referred to herein as logical addresses) within their portions of the data packet to identify higher level components of the network.
- a packet includes a source and destination IP address to identify the communicating nodes. These addresses are logical in that they may be dynamically assigned to account for reconfiguration of the network topology and to logically group related nodes.
- Most protocols e.g., TCP/IP
- Most protocols therefore include a mapping of the higher level protocol specific logical addresses (e.g., IP addresses) to the corresponding lower level physical addresses (e.g., NIC MAC addresses).
- a request is addressed to a server based upon its logical address (e.g., the IP address). This logical address is translated into a corresponding lower level (physical) MAC address in order to transmit the request to the proper computing node.
- ARP packets an address resolution protocol (ARP packets) is defined for use in transferring such address mapping information throughout the network.
- a first node sends out a packet on the network including a higher level logical address (e.g., an IP address in TCP/IP) for which it requires the low level physical address (e.g., NIC MAC address).
- Nodes in the network typically cache such information after acquiring it a first time to reduce the volume of such network traffic.
- redundant server nodes are often utilized to help assure reliable access to the service(s) provided thereby.
- one server node or system provides a particular service while its redundant paired server node remains idle (with respect to provision of the same service).
- the idle second server node takes over the provision of the service when it senses that the first server has failed in some manner.
- the process of taking over service on behalf of a failed server is also referred to herein as "activating" the idle second server.
- Veritas® Firstwatch® provides for redundant server take over in networked distributed computing environments by permitting the redundant server to assume the IP and MAC address identity of its associated failed server.
- this method is not usable in certain networked topologies such as Token Ring or FDDI rings. In such ring topologies, the physical position of a particular NIC relative to its neighbors on the ring is vital to operation of the ring.
- Reconfiguration of the ring topology may be required where, as in Firstwatch®, a NIC suddenly assumes a different NIC address identity due to redundant take over procedures.
- the redundant server simply assumes the IP address of the failed server when taking over from the failed server thereby creating a new association (mapping) between an IP address and a lower level MAC address.
- client processes attempting to use the services of the redundant server will continue to use the cached MAC address associated with the failed server's IP address. Information routed to the failed server at its IP address will therefore not be received by the redundant server because the MAC address used in the transmission is not the MAC address used by the redundant server's NIC card.
- the client process node will timeout awaiting acknowledgment of information sent to the failed server's IP and MAC address.
- This timeout (often referred to in TCP/IP protocols as the ARP timeout) will cause the client to request an update of its cached IP/MAC address mapping.
- ARP messages are exchanged to update the clients ARP cache and processing then continues with the redundant server.
- the present invention solves the above and other problems, thereby advancing the state of the useful arts, by providing methods and associated apparatus for essentially immediate restoration of service to all clients in a wide variety of network topologies and applications.
- the methods of the present invention restore service to clients without need for the client processes to await error recovery timeout conditions and associated updates of their address mapping cache (e.g., ARP cache in TCP/IP protocols).
- the methods of the present invention are applicable in Ethernet as well as ring topologies (e.g., Token Ring and FDDI).
- methods of the present invention utilize multi-cast features of modern NICs to enable a NIC to receive information addressed to any of a plurality of MAC addresses.
- the NIC in a redundant server is configured to actively send and receive information using its own unique MAC address but is additionally configured to receive information addressed to the unique MAC address of the server for which it is the redundant mate.
- Multicast features are generally available on a number of commercially available NICs for enabling a NIC to "listen" to a plurality of MAC addresses.
- a multi-cast MAC address is used to broadcast information to a plurality of NICs all configured to receive information packets addressed to the multi-cast address as well as their own unique MAC address.
- the multi-cast address is programmed into the NIC to provide flexible configuration of the multi-cast addressing.
- the multi-cast address feature of the redundant server's NIC is programmed to receive packets addressed to the address of its paired server.
- the redundant server determines that it must take over provision of the service from its paired server, it assumes the IP address identity of the failed server and begins responding to packets addressed to the failed server's IP and MAC address.
- responses to such received packets are transmitted using the failed server's IP address but the MAC address uniquely assigned to the redundant server's NIC.
- Replies received at the client node using the failed server's IP address but a new MAC address, namely the MAC address of the redundant server's NIC may cause the requesting client to immediately update its address mapping cache (ARP cache) to reflect the new address mapping information.
- ARP cache address mapping cache
- the client need not await error recovery timeout conditions before updating its ARP cache.
- This feature of the present invention helps alleviate delay problems inherent in prior solutions.
- Other clients depending upon the system architecture of the client, may not update their ARP cache. Rather, they will continue to transmit messages to the physical address associated with the failed system. However, the redundant server will receive such messages and respond appropriately as above.
- methods of the present invention are applicable to a wider variety of network topologies than are previous techniques. Specifically, the methods of the present invention are applicable to, for example, Ethernet, Token Ring, and FDDI network topologies. Whereas prior techniques are inapplicable to Token Ring and FDDI (as well as other ring topologies), the methods of the present invention utilize the preconfigured MAC address uniquely assigned to the redundant server's NIC when responding to requests directed to the failed server's IP address.
- the redundant servers operate in a symmetric manner in that each may serve as a standby by for the other.
- the definition of standby and primary server as used herein, is therefore a relative one which describe a role and mode of operation rather than necessarily a physical designation of the servers.
- each server of a pair of redundant servers acts in the role of primary server with regard to services it provides to particular clients.
- each server of the redundant pair of servers acts as a standby server for the other.
- As a standby server it stands ready to provide services for the other server by assuming the identity of the primary server when the primary server has failed.
- both redundant servers continually monitor the physical address of the other so as to maintain a synchronized state in case a take over is required.
- the standby server in such an embodiment may therefore assume the identity of the failed server with complete knowledge of the present state of the failed server.
- Such an embodiment is preferred where the network protocols in use between the clients and servers are state based rather than state-less.
- a standby server need not constantly monitor messages addressed to the primary server. Rather, the standby server takes over from the failed primary server when the primary server is determined to have failed. At the time of such take over processing the standby server assumes the identity of the primary server. This relieves the standby server from the overhead of receiving messages destined for the primary server during normal operation which are ignored (other than maintaining state synchronization with the primary server as above). It is therefore an object of the present invention to provide methods and associated apparatus for activating a standby server in a redundant server network.
- NFS network file system protocols
- FIG. 1 is a block diagram of a redundant server network environment operable in accordance with known techniques devoid of the methods and apparatus of the present invention
- FIG. 2 is a block diagram of a redundant server environment operable in accordance with the improved methods and apparatus of the present invention
- FIG. 3 is a flowchart describing the operation of a redundant server while operable in a standby mode and while operable to take over from a failed server for a state based protocol requiring synchronization between the servers;
- FIG. 4 is a flowchart describing the operation of a redundant server while operable in a standby mode and while operable to take over from a failed server for a state-less protocol requiring no synchronization between the servers. 5. Detailed Description of the Preferred Embodiments
- FIG. 1 is a block diagram depicting a typical network distributed client/server distributed computing application 100.
- client/server model computing environments are applicable to networked as well as non-networked computing environments.
- client processes request and receive services by communicating with server processes.
- the various client and server processes may be operable within a single, standalone, non-networked computing environment or may be distributed over a plurality of networked computing nodes.
- the client/server distributed computing model covers such a broad range of computing environments, the problems addressed by the present invention stem from configurations in which redundant server processes are each operable on distinct computing nodes and a client process is operable on yet another computing node.
- a service is provided by system A 102 (also referred to herein as a server or server node) to client X 106 via communications over network 108.
- System B 104 also referred to herein as a standby server or a redundant server
- All nodes are attached to the network 108 via network interface circuits 112, 114, and 116 (also referred to herein as NICs), respectively.
- NIC 112 is associated with a physical address (MAC address) of MAC.A
- NIC 114 has a MAC address of MAC.B
- NIC 116 has a MAC address of MAC.X.
- many network protocols e.g., TCP/IP
- TCP/IP associate a higher level (logical) address with computing nodes or devices in which a NIC is embedded or inserted. In TCP/IP protocols, for example, this logical address is referred to as an IP address.
- server 102 is associated with a logical address (e.g., IP address) of IP.A
- server 104 is associated with a logical address of IP.B
- client 106 is associated with a logical address of IP.X.
- Service requests generated by client 106 include source and destination logical addresses indicating that the service request is from client 106 by its logical address IP.X and is destined to server 102 by its logical address IP.A.
- Network processing within client 106 eventually applies data representative of the service request to the communication medium of the network 108 via NIC 116.
- the data applied to the physical medium of network 108 includes the low level physical addresses of the source and destination of the information.
- MAC address MAC.X is identified as the physical address source of the information while MAC address MAC.A is identified as the physical address destination of the information on network 108.
- the request is received by server 102 via its NIC 112 and processed appropriately.
- a response is generated for transmission back to the requesting client 106 using the same logical and physical addresses supplied in the request (though the source and destination roles may be reversed as required by the particular protocols).
- Server 104 is idle (with respect to the particular service nominally provided by server 102) awaiting an indication that server 102 has failed in some manner.
- Such indications are provided and sensed by well known means outside the scope of the present invention.
- servers 102 and 104 may periodically communicate to determine each others respective operational status. Failure to appropriately respond to such an inquiry may suffice as an indication that the systems filing to respond has failed and that the redundant server should take over control of the service nominally provided by the failed server.
- Those skilled in the art will recognize many equivalent methods for recognizing indications that a main server has failed so as to initiate a take over by a redundant or standby server.
- Redundant server 104 takes over provision of the services nominally provided by server 102 when it senses a failure in the operation of server 102.
- redundant server 104 may adopt the identity of failed server 102 by assuming both the logical and physical address identity (IP.A and MAC.A) of failed server 102.
- IP.A and MAC.A logical and physical address identity
- the physical address identity of failed server 102 is not a viable option in certain common network topologies such as ring networks (e.g., Token Ring and FDDI).
- the physical address is tightly coupled to physical location of the device on the ring (i.e., the neighbors with which it is directly associated). Assumption of the physical address (MAC.A) in NIC 114 of server 104 may therefore cause problems in such ring topologies.
- redundant server 104 may adopt only the logical address of failed server 102 (e.g., IP.A). In so doing, it can process new service requests received through its NIC 114 having a logical address of failed server 102 (e.g., IP.A). However, client 106 is unaware of the take over by server 104 of failed server 102 and is therefore unaware of the mapping of the logical address of server 102 (IP.A) to the physical address of NIC 114 in sever 104 (MAC.B).
- IP.A the logical address of server 102
- MAC.B physical address of NIC 114 in sever 104
- client 106 will detect an error in transmission of a packet to the physical address of failed server 102. For example, a timeout condition is likely to occur after sufficient time has passed while client 106 waits for a response to its request.
- client 106 may invoke protocol messaging features to inquire as to a new mapping of the logical address of server 102 to a new physical address.
- protocol messaging In TCP/IP protocols such messaging is referred to as address resolution protocol or ARP.
- ARP messages are generated by client 106 in response to an error condition in communicating with server 102 and redundant server 104 responds with a new physical address (MAC.B) to be associated with the logical address of failed server 102 (IP.A).
- MAC.B new physical address
- Client 106 includes ARP cache 120 wherein it store address mapping information received in its initial configuration and as modified over time by address resolution protocol message exchange over network 108. Client 106 relies on the information stored in ARP cache 120 to avoid excessive ARP message traffic being generated on network 108 thereby using available bandwidth. Information in ARP cache 120 is updated when new mapping information is made available to client 106 or at certain time based events (e.g., timeouts corresponding to error conditions in network message traffic).
- these delays may be undesirable in certain high performance distributed applications. During these delay periods the service nominally provided by failed server 102 are simply unavailable to client 106.
- redundant server 104 can assume the logical address identity of failed server 102, client 106 remains unaware of the address mapping change until the ARP cache 120 is updated. Therefore, redundant server 104 receive none of the requests from client 106 and hence can do nothing to process such requests.
- FIG. 2 is a block diagram of a distributed computing networked application 200 similar in many respects to that of FIG. 1 but operable in accordance with the methods of the present invention.
- each server 202 and 204 is operable to take over operation of the other in case of failure.
- Each server 202 and 204 is configured with its own logical address, IP.A and IP.B, respectively.
- each server 202 and 204 includes a NIC 212 and 214, respectively.
- Each NIC 212 and 214 is configured with its own unique physical address, MAC.A and MAC.B, respectively.
- each server 202 and 204 is configured to remember the logical address of the alternate server, namely IP.B and IP.A, respectively.
- each NIC 212 and 214 is configured to additionally receive transmissions on network 208 destined for the alternate physical address, namely MAC.B and MAC.A, respectively.
- Each server 202 and 204 is therefore capable of receiving (e.g., snooping or monitoring) requests destined for the alternate server.
- Each server ignores requests received that are destined for the alternate server so long as the alternate server is operable to provide requisite services and a take over procedure is unnecessary.
- each server 202 and 204 receives information transmitted to either server but processes only those that are intended for its servicing under normal operating conditions.
- server 204 When server 204 detects a failure in the operation of server 202 it takes over processing of requests destined to server 202. Server 204 assumes the logical address identity (e.g., IP.A) of the failed server 202. Since server 204 was monitoring requests destined to failed server 202 it may immediately process any outstanding request and return an appropriate response to requesting client 106. The response from redundant server 204 to client 106 is generated with the assumed logical address identity of failed server 202 (e.g., IP.A) and the physical address of NIC 214 in redundant server 204 (e.g., MAC.B).
- IP.A the assumed logical address identity of failed server 202
- NIC 214 e.g., MAC.B
- client 106 may immediately recognize in the response packet that the mapping of a logical address for server 202 (e.g., IP.A) to a physical address has changed and should now be directed to the new physical address received as the source physical address in the response, namely MAC.B the physical address of redundant server 204. Further requests to logical address IP.A are therefore transferred via network 208 to server 204 rather than failed server 202. Other client architectures may not update their address mapping information based upon the response message. Rather, such clients will continue to use the existing mapping information which continues to direct requests to the logical address and physical address of the failed server. However, redundant (standby) server 204 receives, processes, and responds to such messages as described above during such take over mode processing of requests on behalf of the failed server.
- redundant (standby) server 204 receives, processes, and responds to such messages as described above during such take over mode processing of requests on behalf of the failed server.
- Client 106 receives its responses essentially without delay under this method of the present invention because redundant server 204 is prepared to respond to any monitored message destined to server 202 in response to detecting a failure in operation of server 202.
- This method of the present invention is usable in a wider variety of network topologies such as ring networks because transmissions from server 204 via its NIC 214 use the preconfigured nominal physical address associated therewith, namely MAC.B.
- Redundant server 204 continues to process requests addressed to the logical address of failed server 202 until server 204 detects that server 202 is again operable and prepared to resume normal processing of service requests.
- server 202 When so sensed as operable, server 202 will begin processing requests directed to its original logical address (IP.A) in a manner analogous to the take over processing above. Specifically, server 202 will monitor messages sent to its logical address but to the physical address of server 204 (MAC.B). Server 202 will process the next such request (in cooperation with redundant server 204). The response generated by such processing will use the logical address of server 202 (IP.A) and its nominal physical address (MAC.A). As above, client 106 will immediately detect the change in logical to physical address mapping reflected in the response received and update its ARP cache 120 accordingly.
- servers 202 and 204 act as a redundant service provider for the other. Each performs a take over process when it senses the other has failed. Each permits the other to resume processing of requests by reverting the address mapping when the other is sensed to be again operational.
- All address changes required for the take over and reversion processing between the two servers are performed as described above such that all information transmitted from a server uses the nominal physical address for that server.
- This feature allows the method of the present invention to be applied to ring topology networks such as Token Ring and FDDI in addition to Ethernet bus topologies.
- client 106 is notified immediately of the address mapping changes required in its ARP cache 120 by the first response returned from the redundant controller 204 following its take over from server 202. This feature eliminates delays incurred by previous techniques wherein the client would be unaware of the address change until potentially lengthy timeout delays when the client would naturally seek to update ARP cache information (e.g., in response to transmission and retry timeout errors).
- FIG. 3 is a flowchart describing the methods of the present invention operable within server 202 and 204 to perform the take over processing as described with respect to FIG. 2 above.
- the processing within each of a pair of redundant servers is (typically) symmetric.
- Each server provides its own particular services and serves as a redundant server to take over provision of services by the other in case of failure.
- the processing described in FIG. 3 is therefore applicable to either server.
- the method is preferably operable simultaneously within both servers 202 and 204 of FIG. 2 wherein each provides its own services to requesting clients and also provides a standby (redundant) service for the services of the other in case of failure.
- FIG. 3 describes the processing of the methods of the present invention with respect to a pair of server for simplicity of presentation. Specifically, FIG. 3 describes the operation of the methods of the present invention from the perspective of system A (server 202 of FIG.2) wherein it acts as a redundant server for system B (server 204 of FIG. 2).
- FIG. 3 describes a method of the present invention most applicable to state based network protocols in which messages are potentially interdependent upon one another.
- a state based protocol is the Microsoft® Windows ® networking protocol used for file and printer sharing among personal computers.
- the standby server needs to monitor message traffic destined to the primary server even during normal processing by the primary server.
- Messages received by the standby server but destined for the operational primary server are processed to the extent that any requisite state change information pertaining to the protocol is derived. No replies are generated by the standby server by virtue of such processing nor does the standby server generate any other side effects from the processing of the request. Only the state change information is gathered so as to maintain synchronization between the standby and primary servers.
- the method of FIG. 3 is required where state based protocols are applied to the network but is also operable in application where state-less protocols are utilized.
- Element 300 is first operable to determine the physical address (MAC.B) for the alternate system B. This may be determined by any form of communication between system A and system B. For example, the two servers may exchange configuration information at initialization via the common network 208. Alternatively, there may be a dedicated communication channel between systems A and B for purposes of such configuration communications. Element 302 then logically binds or associates the logical address (IP.A) for system A to the physical address for system A's NIC (MAC.A). This binding is a logical configuration which establishes the logical to physical address mapping for information to received or transmitted by system A via network 208.
- IP.A logical address
- MAC.A logical address mapping
- Element 304 then provides configuration information to system A's NIC to enable the NIC to receive information on either of the two physical addresses associated with the redundant pair of servers (MAC.A or MAC.B).
- MAC.A or MAC.B redundant pair of servers
- standard multi-cast features of most commercially available NICs may be used to receive information on either of two physical addresses.
- One exemplary such NIC device utilized in the preferred embodiment of the present invention is the Digital Semiconductor 21140A PCI Fast Ethernet LAN Controller made by Digital Equipment Corporation (DEC) of Maynard Massachusetts. This integrated circuit provides high speed Ethernet communications including programmable multi-cast features to enable operation of the methods of the present invention.
- element 306 is operable to determine if such failure indicia from system B has been sensed.
- any of several well known methods for detecting a failure may be utilized in conjunction with the methods of the present invention. For example, so called watchdog timer messages may be exchanged between the redundant servers on a frequent periodic basis. Failure to receive such a watchdog message from the alternate system is an indication that a take over process is required. Such messages may be exchanged via the common network 208 of FIG. 2 or over a dedicated communication channel among the redundant controllers.
- elements 308 and 310 are operable to process normal requests received by system A through its NIC via network 208 addressed to its nominal logical and physical address (e.g., IP.A and MAC.A). Specifically, element 308 processes normal requests destined for system A by performing the requested processing and returning normal responses resulting therefrom. Element 310 assures that information received via system A's NIC destined for system B's physical address (MAC.B) is processed to the extent necessary to synchronize with the state processing of the network protocol in use. Beyond such synchronization, the message is otherwise ignored in the sense that no reply is generated or transmitted to the requesting client by the standby server.
- MAC.B physical address
- element 310 may represent proactive steps to assure such packets are ignored or discarded within system A. In other environments, such messages may be discarded at lower levels of the computing environment (e.g., at the NIC or within various of the layers of the network processing modules - the protocol stack). Essentially, element 310 is shown to stress that packets received by system A which are destined for system B are processed differently than packets destined to system A during normal processing so that system B may process such requests in its normal course of processing. Normal processing continues iterativley with elements 306-310 until element 306 senses an indication of failure in the alternate system B.
- the computing environment e.g., operating system features and networking features
- elements 312-316 are iteratively operable to perform redundant mode processing within system thereby processing requests destined to the failed system B as well as those destined to system A for normal processing.
- element 312 is operable in like manner to element 308 to process any requests received by system A's NIC which are addressed to system A for normal processing. Such requests are processed by performing the requisite processing and returning any required messages to the requesting client. Such requests and responses use the nominal logical and physical addresses of system (IP.A and MAC.A).
- Element 312 is also operable to process requests received on system
- IP.B system B's logical address
- clients adjust their respective addressing mapping tables in response to particular responses generated by element 314. Requests generated by clients following such adjustments to their address mapping tables are directed to system B's logical address (IP.B) but systems A's physical address (MAC.A). Such requests are processed on behalf of the failed system B as though they were normally directed for processing to system A. As noted elsewhere, some client may not adjust their respective address mapping tables. Such clients will continue to address requests to the address of failed system B (IP.B and MAC.B). None the less, element 314 will process the requests and continue to return required responses to the requesting clients.
- Element 314 is operable within system A to process requests destined to system B and normally ignored by element 310 during normal processing.
- element 314 receives and processes requests addressed to the logical address of failed system B (IP.B). Initially such requests are addressed to the logical and physical address of system B (IP.B and MAC.B). However, in system A generates and returns any required response to the request using the logical address of system B but the physical address nominally assigned to system A (i.e., its own physical address MAC.A). As noted above, such a response which associates systems A's physical address (MAC.A) with failed system B's logical address (IP.B) will cause the requesting client to immediately adjust its address mapping table (ARP cache) to account for the revised mapping.
- ARP cache address mapping table
- watchdog timer messaging may provide both an indication of a failure as well as an indication of a return to operational status in a previously failed system.
- Those skilled in the art will recognize several equivalent methods for sensing indicia of restoration of a failed system to an operational state. If element 316 does not sense such an indication, elements 312-316 are iteratively operable to continue processing requests directed to either system A or failed system B.
- FIG. 4 is a flowchart describing a preferred embodiment of the methods of the present invention operable within server 202 and 204 to perform the take over processing as described with respect to FIG. 2 above.
- the processing within each of a pair of redundant servers is (typically) symmetric.
- Each server provides its own particular services and serves as a redundant server to take over provision of services by the other in case of failure.
- the processing described in FIG. 4 is therefore applicable to either server.
- the method is preferably operable simultaneously within both servers 202 and 204 of FIG. 2 wherein each provides its own services to requesting clients and also provides a standby (redundant) service for the services of the other in case of failure.
- FIG. 4 describes the processing of the methods of the present invention with respect to a pair of server for simplicity of presentation. Specifically, FIG. 4 describes the operation of the methods of the present invention from the perspective of system A (server 202 of FIG.2) wherein it acts as a redundant server (standby server) for system B (server 204 of FIG. 2).
- FIG. 4 describes a preferred embodiment of a method of the present invention wherein exclusively state-less protocols are used in communicating over the network.
- exemplary of such a state based protocol is the network file system (NFS) networking protocol used for file sharing among computers.
- NFS network file system
- a newly received message requires no state information derived from processing of preceding messages.
- the standby server need not monitor message traffic destined to the primary server during normal processing by the primary server.
- each server receives packets addressed only to its nominal physical and logical address during normal operation. Only when a failure is sensed in the primary server (system B as described in FIG.
- Element 402 is first operable to logically bind or associate the logical address (IP.A) for system A to the physical address for system A's NIC (MAC.A). This binding is a logical configuration which establishes the logical to physical address mapping for information to received or transmitted by system A via network 208.
- Element 404 then provides configuration information to system A's NIC to enable the NIC to receive information on its nominally assigned physical address (MAC.A).
- element 406 is operable to determine if such failure indicia from system B has been sensed.
- any of several well known methods for detecting a failure may be utilized in conjunction with the methods of the present invention.
- so called watchdog timer messages also referred to as heartbeat messages
- Such messages may be exchanged via the common network 208 of FIG. 2 or over a dedicated communication channel among the redundant controllers.
- element 408 is operable to process normal requests received by system A through its NIC via network 208 addressed to its nominal logical and physical address (e.g., IP.A and MAC.A). Specifically, element 408 processes normal requests destined for system A by performing the requested processing and returning normal responses resulting therefrom. Processing then continues iteratively performing elements 406 and 408 until a failure is sensed in system B.
- nominal logical and physical address e.g., IP.A and MAC.A
- elements 410-416 are iteratively operable to perform redundant mode processing within system thereby processing requests destined to the failed system B as well as those destined to system A for normal processing.
- element 410 and 412 determine the present physical address (MACADDR) assigned to system B (MAC.B) and reconfigure system A's NIC to receive messages address to either physical address (MAC.A or MAC.B).
- Element 412 is then operable in like manner to element 408 to process any requests received by system A's NIC which are addressed to either system A for normal processing or addressed to system B for redundant (take over) processing. Such requests are processed by performing the requisite processing and returning any required messages to the requesting client.
- Such requests and responses use the logical address supplied by the request (IP.A if directed to system A or IP.B if directed to system B) and the physical address of system A (MAC.A).
- clients may adjust their respective addressing mapping tables in response to particular responses generated by element 414. Requests generated by clients following such adjustments to their address mapping tables are directed to system B's logical address (IP.B) but systems A's physical address (MAC.A). Such requests are processed on behalf of the failed system B as though they were normally directed for processing to system A. Where clients do not adjust their address mapping information, requests will continue to be directed to the logical and physical address of the failed primary system B. Such requests will non the less be processed by the redundant (standby) processing of element 414.
- Element 416 is then operable to detect whether it is now possible to return to normal processing mode. Specifically, element 416 determines whether indicia of a return to normal processing by failed system B has been sensed. As noted above, watchdog timer messaging may provide both an indication of a failure as well as an indication of a return to operational status in a previously failed system. Those skilled in the art will recognize several equivalent methods for sensing indicia of restoration of a failed system to an operational state. If element 416 does not sense such an indication, received message continue to be processed by iterative performance of element 414. When element 416 determines that system B has been restored to an operation status, processing returns to element 404 to continue processing requests in normal mode wherein requests directed to system B are processed by system B. In particular, element 404 reconfigures system A's NIC to return to normal operation wherein only requests directed to system A (IP.A and MAC.A) are processed. Elements 406 and 408 then resume normal mode processing until another failure is detected.
- a key feature of both methods described above with respect to FIGS. 3 and 4 is that replies returned from the standby server when performing requests on behalf of the failed primary server are returned with a physical address corresponding to the nominal physical address of the standby server. This key feature enables the methods of the present invention to be utilized with a wider range of network topologies and media than was possible with prior approaches.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Hardware Redundancy (AREA)
- Small-Scale Networks (AREA)
Abstract
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU71494/98A AU7149498A (en) | 1997-04-25 | 1998-04-22 | Redundant server failover in networked environment |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US84571897A | 1997-04-25 | 1997-04-25 | |
US08/845,718 | 1997-04-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1998049620A1 true WO1998049620A1 (fr) | 1998-11-05 |
Family
ID=25295929
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US1998/008142 WO1998049620A1 (fr) | 1997-04-25 | 1998-04-22 | Reprise par un serveur redondant, dans un environnement en reseau |
Country Status (2)
Country | Link |
---|---|
AU (1) | AU7149498A (fr) |
WO (1) | WO1998049620A1 (fr) |
Cited By (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2000031597A2 (fr) * | 1998-11-24 | 2000-06-02 | Automationx Software For Industrial Automation Gmbh | Systeme d'automatisation pour la realisation de taches definies dans le cadre de la technique des processus, et procede associe |
WO2001030023A2 (fr) * | 1999-10-18 | 2001-04-26 | Vocaltec Communications Ltd. | Reprise automatique de composants de transmission ou de systeme de traitement logique au moyen d'une multidiffusion par protocole internet |
WO2002039198A2 (fr) * | 2000-11-07 | 2002-05-16 | Schneider Automation Inc. | Procede et appareil permettant d'assurer un systeme de commande auxiliaire actif sur un reseau |
WO2003017104A1 (fr) * | 2001-08-15 | 2003-02-27 | Spx Corporation | Procede et systeme d'exploitation de serveurs redondants |
US6657974B1 (en) * | 2000-04-14 | 2003-12-02 | International Business Machines Corporation | Method and apparatus for generating replies to address resolution protocol requests |
WO2004023305A1 (fr) * | 2002-09-05 | 2004-03-18 | International Business Machines Corporation | Mecanisme de commutation du controleur reseau active par l'acces direct memoire a distance et de commutation de retour a l'etat initial |
US6775278B1 (en) * | 2000-04-14 | 2004-08-10 | International Business Machines Corporation | Method and apparatus for generating replies to address resolution protocol requests |
EP1480404A2 (fr) * | 2003-05-19 | 2004-11-24 | Advance Fibre Communication, Inc. | Appareil et méthode pour utiliser un serveur primaire et un serveur de restauration qui prennent en charge un client radius et une adresse IP partagée |
EP1482712A1 (fr) * | 2003-05-30 | 2004-12-01 | Ixia | Adresse réseau virtuelle |
WO2005069189A1 (fr) * | 2004-01-15 | 2005-07-28 | Utstarcom Korea Limited | Serveur aaa duplique comprenant un procede de verification omp et procede de gestion associe |
US6973024B1 (en) * | 2001-02-28 | 2005-12-06 | Utstarcom, Inc. | Method for modem element switchover using multicast groups |
EP1758316A1 (fr) * | 2004-06-14 | 2007-02-28 | Huawei Technologies Co., Ltd. | Procede de comptage de quantite pour utilisateur de mbms |
WO2007097915A2 (fr) * | 2006-02-15 | 2007-08-30 | Sony Computer Entertainment America Inc. | Systèmes et méthodes de gestion de serveur |
US7340637B2 (en) * | 2001-03-26 | 2008-03-04 | Duaxes Corporation | Server duplexing method and duplexed server system |
US7468949B2 (en) * | 2003-02-06 | 2008-12-23 | Siemens Aktiengesellschaft | Method and device for medium-redundant operation of a terminal in a network |
US7596615B2 (en) | 2006-02-22 | 2009-09-29 | Microsoft Corporation | Multi-server automated redundant service configuration |
US7627694B2 (en) | 2000-03-16 | 2009-12-01 | Silicon Graphics, Inc. | Maintaining process group membership for node clusters in high availability computing systems |
CN101043330B (zh) * | 2006-06-22 | 2010-08-25 | 华为技术有限公司 | 一种防mac地址仿冒的装置及方法 |
US7853945B2 (en) | 2006-02-22 | 2010-12-14 | Michael Kramer | Integrated computer server imaging |
US7962509B2 (en) | 2006-02-15 | 2011-06-14 | Sony Computer Entertainment America Llc | Systems and methods for server management |
US8032791B2 (en) | 2009-07-07 | 2011-10-04 | International Business Machines Corporation | Diagnosis of and response to failure at reset in a data processing system |
EP2395713A1 (fr) * | 2009-02-09 | 2011-12-14 | Nec Corporation | Système de communication, dispositif de communication, dispositif de commande, procédé de commande et programme |
US8769132B2 (en) | 2000-03-16 | 2014-07-01 | Sony Computer Entertainment America Llc | Flexible failover policies in high availability computing systems |
US8966557B2 (en) | 2001-01-22 | 2015-02-24 | Sony Computer Entertainment Inc. | Delivery of digital content |
US20150113313A1 (en) * | 2013-10-23 | 2015-04-23 | Synology Incorporated | Method of operating a server system with high availability |
EP2936777A4 (fr) * | 2012-12-20 | 2016-07-27 | Ericsson Telefon Ab L M | Hôte actif et hôte de sauvegarde dans un système hôte pour des communications avec un terminal connecté à un réseau ip |
US9483405B2 (en) | 2007-09-20 | 2016-11-01 | Sony Interactive Entertainment Inc. | Simplified run-time program translation for emulating complex processor pipelines |
CN103346903B (zh) * | 2013-06-20 | 2017-02-15 | 北京捷成世纪科技股份有限公司 | 一种双机备份的方法和装置 |
US10558983B2 (en) * | 2003-06-26 | 2020-02-11 | International Business Machines Corporation | User access to a registry of business entity definitions |
CN111222100A (zh) * | 2018-11-27 | 2020-06-02 | 株式会社理光 | 功能管理系统、信息处理装置、计算机装置以及存储介质 |
CN112887133A (zh) * | 2021-01-21 | 2021-06-01 | 杭州康吉森自动化科技有限公司 | 一种用于工业网关的冗余切换方法、工业网关和存储介质 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1992018931A1 (fr) * | 1991-04-23 | 1992-10-29 | Eastman Kodak Company | Systeme de fichier pour reseau insensible aux defaillances |
US5513314A (en) * | 1995-01-27 | 1996-04-30 | Auspex Systems, Inc. | Fault tolerant NFS server system and mirroring protocol |
US5590285A (en) * | 1993-07-28 | 1996-12-31 | 3Com Corporation | Network station with multiple network addresses |
-
1998
- 1998-04-22 WO PCT/US1998/008142 patent/WO1998049620A1/fr active Application Filing
- 1998-04-22 AU AU71494/98A patent/AU7149498A/en not_active Abandoned
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1992018931A1 (fr) * | 1991-04-23 | 1992-10-29 | Eastman Kodak Company | Systeme de fichier pour reseau insensible aux defaillances |
US5590285A (en) * | 1993-07-28 | 1996-12-31 | 3Com Corporation | Network station with multiple network addresses |
US5513314A (en) * | 1995-01-27 | 1996-04-30 | Auspex Systems, Inc. | Fault tolerant NFS server system and mirroring protocol |
Cited By (49)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2000031597A3 (fr) * | 1998-11-24 | 2000-07-27 | M & R Automatisierung Von Indu | Systeme d'automatisation pour la realisation de taches definies dans le cadre de la technique des processus, et procede associe |
WO2000031597A2 (fr) * | 1998-11-24 | 2000-06-02 | Automationx Software For Industrial Automation Gmbh | Systeme d'automatisation pour la realisation de taches definies dans le cadre de la technique des processus, et procede associe |
WO2001030023A3 (fr) * | 1999-10-18 | 2002-06-06 | Vocaltec Comm Ltd | Reprise automatique de composants de transmission ou de systeme de traitement logique au moyen d'une multidiffusion par protocole internet |
WO2001030023A2 (fr) * | 1999-10-18 | 2001-04-26 | Vocaltec Communications Ltd. | Reprise automatique de composants de transmission ou de systeme de traitement logique au moyen d'une multidiffusion par protocole internet |
US8769132B2 (en) | 2000-03-16 | 2014-07-01 | Sony Computer Entertainment America Llc | Flexible failover policies in high availability computing systems |
US7627694B2 (en) | 2000-03-16 | 2009-12-01 | Silicon Graphics, Inc. | Maintaining process group membership for node clusters in high availability computing systems |
US9405640B2 (en) | 2000-03-16 | 2016-08-02 | Sony Interactive Entertainment America Llc | Flexible failover policies in high availability computing systems |
US6657974B1 (en) * | 2000-04-14 | 2003-12-02 | International Business Machines Corporation | Method and apparatus for generating replies to address resolution protocol requests |
US6775278B1 (en) * | 2000-04-14 | 2004-08-10 | International Business Machines Corporation | Method and apparatus for generating replies to address resolution protocol requests |
WO2002039198A3 (fr) * | 2000-11-07 | 2003-03-20 | Schneider Automation | Procede et appareil permettant d'assurer un systeme de commande auxiliaire actif sur un reseau |
WO2002039198A2 (fr) * | 2000-11-07 | 2002-05-16 | Schneider Automation Inc. | Procede et appareil permettant d'assurer un systeme de commande auxiliaire actif sur un reseau |
US8966557B2 (en) | 2001-01-22 | 2015-02-24 | Sony Computer Entertainment Inc. | Delivery of digital content |
US6973024B1 (en) * | 2001-02-28 | 2005-12-06 | Utstarcom, Inc. | Method for modem element switchover using multicast groups |
US7340637B2 (en) * | 2001-03-26 | 2008-03-04 | Duaxes Corporation | Server duplexing method and duplexed server system |
WO2003017104A1 (fr) * | 2001-08-15 | 2003-02-27 | Spx Corporation | Procede et systeme d'exploitation de serveurs redondants |
WO2004023305A1 (fr) * | 2002-09-05 | 2004-03-18 | International Business Machines Corporation | Mecanisme de commutation du controleur reseau active par l'acces direct memoire a distance et de commutation de retour a l'etat initial |
CN100380334C (zh) * | 2002-09-05 | 2008-04-09 | 国际商业机器公司 | 启用远程直接存储器存取的网络接口控制器切换方法 |
US7468949B2 (en) * | 2003-02-06 | 2008-12-23 | Siemens Aktiengesellschaft | Method and device for medium-redundant operation of a terminal in a network |
US7472313B2 (en) | 2003-05-19 | 2008-12-30 | Tellabs Petaluma, Inc. | Primary server and backup server that share an IP address and a limited number of message identifiers |
US7228459B2 (en) | 2003-05-19 | 2007-06-05 | Tellabs Petaluma, Inc. | Apparatus and method that provides a primary server and a backup server that both support a RADIUS client and share an IP address |
EP1480404A3 (fr) * | 2003-05-19 | 2005-08-03 | Tellabs Petaluma, Inc. | Appareil et méthode pour utiliser un serveur primaire et un serveur de restauration qui prennent en charge un client radius et une adresse IP partagée |
EP1480404A2 (fr) * | 2003-05-19 | 2004-11-24 | Advance Fibre Communication, Inc. | Appareil et méthode pour utiliser un serveur primaire et un serveur de restauration qui prennent en charge un client radius et une adresse IP partagée |
US7440415B2 (en) | 2003-05-30 | 2008-10-21 | Ixia | Virtual network addresses |
EP1482712A1 (fr) * | 2003-05-30 | 2004-12-01 | Ixia | Adresse réseau virtuelle |
US10650387B2 (en) | 2003-06-26 | 2020-05-12 | International Business Machines Corporation | User access to a registry of business entity definitions |
US10558983B2 (en) * | 2003-06-26 | 2020-02-11 | International Business Machines Corporation | User access to a registry of business entity definitions |
WO2005069189A1 (fr) * | 2004-01-15 | 2005-07-28 | Utstarcom Korea Limited | Serveur aaa duplique comprenant un procede de verification omp et procede de gestion associe |
US7953015B2 (en) | 2004-06-14 | 2011-05-31 | Huawei Technologies Co., Ltd. | Method for ensuring reliability in network |
EP1758316A1 (fr) * | 2004-06-14 | 2007-02-28 | Huawei Technologies Co., Ltd. | Procede de comptage de quantite pour utilisateur de mbms |
EP1758316A4 (fr) * | 2004-06-14 | 2008-02-27 | Huawei Tech Co Ltd | Procede de comptage de quantite pour utilisateur de mbms |
US7716238B2 (en) | 2006-02-15 | 2010-05-11 | Sony Computer Entertainment America Inc. | Systems and methods for server management |
US7962509B2 (en) | 2006-02-15 | 2011-06-14 | Sony Computer Entertainment America Llc | Systems and methods for server management |
US7979460B2 (en) | 2006-02-15 | 2011-07-12 | Sony Computer Entainment America Inc. | Systems and methods for server management |
US9886508B2 (en) | 2006-02-15 | 2018-02-06 | Sony Interactive Entertainment America Llc | Systems and methods for server management |
WO2007097915A2 (fr) * | 2006-02-15 | 2007-08-30 | Sony Computer Entertainment America Inc. | Systèmes et méthodes de gestion de serveur |
US8732162B2 (en) | 2006-02-15 | 2014-05-20 | Sony Computer Entertainment America Llc | Systems and methods for server management |
WO2007097915A3 (fr) * | 2006-02-15 | 2007-12-06 | Sony Comp Entertainment Us | Systèmes et méthodes de gestion de serveur |
US7596615B2 (en) | 2006-02-22 | 2009-09-29 | Microsoft Corporation | Multi-server automated redundant service configuration |
US7853945B2 (en) | 2006-02-22 | 2010-12-14 | Michael Kramer | Integrated computer server imaging |
CN101043330B (zh) * | 2006-06-22 | 2010-08-25 | 华为技术有限公司 | 一种防mac地址仿冒的装置及方法 |
US9483405B2 (en) | 2007-09-20 | 2016-11-01 | Sony Interactive Entertainment Inc. | Simplified run-time program translation for emulating complex processor pipelines |
EP2395713A4 (fr) * | 2009-02-09 | 2014-12-31 | Nec Corp | Système de communication, dispositif de communication, dispositif de commande, procédé de commande et programme |
EP2395713A1 (fr) * | 2009-02-09 | 2011-12-14 | Nec Corporation | Système de communication, dispositif de communication, dispositif de commande, procédé de commande et programme |
US8032791B2 (en) | 2009-07-07 | 2011-10-04 | International Business Machines Corporation | Diagnosis of and response to failure at reset in a data processing system |
EP2936777A4 (fr) * | 2012-12-20 | 2016-07-27 | Ericsson Telefon Ab L M | Hôte actif et hôte de sauvegarde dans un système hôte pour des communications avec un terminal connecté à un réseau ip |
CN103346903B (zh) * | 2013-06-20 | 2017-02-15 | 北京捷成世纪科技股份有限公司 | 一种双机备份的方法和装置 |
US20150113313A1 (en) * | 2013-10-23 | 2015-04-23 | Synology Incorporated | Method of operating a server system with high availability |
CN111222100A (zh) * | 2018-11-27 | 2020-06-02 | 株式会社理光 | 功能管理系统、信息处理装置、计算机装置以及存储介质 |
CN112887133A (zh) * | 2021-01-21 | 2021-06-01 | 杭州康吉森自动化科技有限公司 | 一种用于工业网关的冗余切换方法、工业网关和存储介质 |
Also Published As
Publication number | Publication date |
---|---|
AU7149498A (en) | 1998-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO1998049620A1 (fr) | Reprise par un serveur redondant, dans un environnement en reseau | |
JP4012545B2 (ja) | リモート・ダイレクト・メモリ・アクセス対応ネットワーク・インタフェース・コントローラのスイッチオーバーとスイッチバックのサポート | |
US7885180B2 (en) | Address resolution request mirroring | |
US6760859B1 (en) | Fault tolerant local area network connectivity | |
US7055056B2 (en) | System and method for ensuring the availability of a storage system | |
US6597700B2 (en) | System, device, and method for address management in a distributed communication environment | |
US7415535B1 (en) | Virtual MAC address system and method | |
US5983360A (en) | Information processing system with communication system and hot stand-by change-over function therefor | |
US20060206611A1 (en) | Method and system for managing programs with network address | |
US20060020854A1 (en) | Method and apparatus for high-speed network adapter failover | |
JP2004032224A (ja) | サーバ引継システムおよびその方法 | |
US20070223494A1 (en) | Method for the resolution of addresses in a communication system | |
JP2002057682A (ja) | ネットワークインタフェース切替え方法及びネットワークに接続可能なコンピュータ | |
CN108075971A (zh) | 一种主备切换方法及装置 | |
US10931601B2 (en) | Method and apparatus for conditional broadcasting of network configuration data | |
US6356985B1 (en) | Computer in multi-cluster system | |
US20080144634A1 (en) | Selective passive address resolution learning | |
KR100383490B1 (ko) | 고가용성 네트워크 구현방법 | |
JP4028627B2 (ja) | クライアントサーバシステムおよびクライアントサーバシステムの通信管理方法 | |
JP2001117899A (ja) | マルチサーバシステム | |
JP2003228527A (ja) | 冗長機能付きサーバシステムおよびそのサーバシステムの運用制御方法 | |
JP4258016B2 (ja) | ネットワーク構成制御システム | |
JP5170000B2 (ja) | 冗長化ペア検出方法、通信装置、冗長化ペア検出プログラム、記録媒体 | |
JP2001144806A (ja) | 二重化システムのネットワークアドレス引継方法およびネットワークアドレス引継機能を有する二重化システム並びに記録媒体 | |
WO1998049621A1 (fr) | Interfaçage de messages de commande dans un environnement a serveurs redondants |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AL AM AT AU AZ BA BB BG BR BY CA CH CN CU CZ DE DK EE ES FI GB GE GH GM GW HU ID IL IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT UA UG UZ VN YU ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GH GM KE LS MW SD SZ UG ZW AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN ML MR NE SN TD TG |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
REG | Reference to national code |
Ref country code: DE Ref legal event code: 8642 |
|
NENP | Non-entry into the national phase |
Ref country code: JP Ref document number: 1998547113 Format of ref document f/p: F |
|
122 | Ep: pct application non-entry in european phase | ||
NENP | Non-entry into the national phase |
Ref country code: CA |