CROSS REFERENCE TO RELATED APPLICATIONThis is a continuation of U.S. application Ser. No. 13/594,075, filed Aug. 24, 2012, which is a continuation of U.S. application Ser. No. 11/833,320, filed Aug. 3, 2007 (now, U.S. Pat. No. 8,255,575). This application relates to and claims priority to Japanese Patent Application No. 2006-321021, filed Nov. 29, 2006. The entirety of the contents and subject matter of all of the above is incorporated herein by reference.
BACKGROUND OF THE INVENTION1. Field of the Invention
The present invention generally relates to an apparatus which provides redundancy of communication routes in a ring network. More specifically, the present invention is directed to such an apparatus capable of realizing high-speed switching of routes, and capable of simply coupling a plurality of ring networks, and also capable of independently performing operations in each of these ring networks so as to realize constructions of more flexible networks, and further, capable of reducing loads given to a CPU.
2. Description of the Related Art
While networks are constructed, there are some possibilities that a plurality of apparatuses are connected to each other in ring structures in order to make communication routes redundant. Networks having ring structures have a feature that necessary amounts of transfer routes and the like may be reduced, as compared with those of networks having mesh structures. On the other hand, since the networks are structured in ring shapes, when broadcast packets flow through the networks having the ring structures, transfer operations of these broadcast packets are repeatedly carried out in an infinite time, so that such a loop is produced in which the broadcast packets are continuously circulated in the same route for a long time. As a result, traffic loads on the networks caused by broadcast streams are increased, or CPU loads caused by unstable conditions of learning are increased, which may give adverse influences to communications. In order to avoid the occurrence of this loop, various methods capable of logically cutting off loop structures have been proposed in networks having ring structures.
In the spanning tree protocol described in IEEE Std 802.1D, 1998 Edition, such a control frame called as “BPDU” is periodically exchanged among apparatuses which construct a network in order to cut off a loop structure. While a plurality of protocol information used to determine operations of the protocol have been superimposed on the BPDU, each of these apparatuses which receive this BPDU changes logical statuses in ports which constitute a loop into blocking statuses based upon the protocol information of the received BPDU, and constructs such a network having a tree structure, so that the loop is logically cut off.
While the redundant protocol described in JP-A-2004-201009 is specifically used in a ring network, one structural apparatus among apparatuses which constitutes the ring network is defined as a monitoring apparatus, whereas other structural apparatuses are defined as relay apparatuses. In the monitoring apparatus, a control frame is periodically transmitted from an one-sided port thereof, and a reception of this transmitted is monitored at an opposite-sided port thereof. While the control frame is being received at the opposite-sided port, the monitoring apparatus transmits another control frame from this opposite-sided port, and sets the opposite-sided port to such a condition that a user frame cannot be relayed. As a consequence, the loop structure is logically cut off. If the control frame cannot be received at the opposite-sided port, then the monitoring apparatus stops the transmission of the control frame from this opposite-sided port, and permits the relay of the user frame. As a result, the communication routes can be made redundant.
Within the above-described conventional techniques, since the spanning tree protocol disclosed in IEEE Std 802.1D 1988 Edition may also be directed to complex network structures, the tree structure must be determined based upon the plurality of information superimposed on BPDU exchanged among the apparatuses, and thus, the protocol operations become complex. As a result, even when a structure of a ring network is made simple, the below-mentioned problems may occur. That is, while the structure of such a simple ring network is changed when a failure occurs and the failure is recovered within this simple ring network, and also while apparatuses are additionally conducted to the simple ring network, there are some possibilities that several tens of seconds are required until communications are stabilized, during which the communications are stopped. Also, since it is practically difficult to realize the complex protocol operations by employing hardware, these complex protocol operations are realized by employing software. As a result, if loads given to a CPU are increased, then there is another problem that an adverse influence may be given to the complex protocol operations.
SUMMARY OF THE INVENTIONFurther, as to the redundant protocol described in JP-A-2004-201009 within the conventional techniques, the problems occurred in the spanning tree protocol have been solved by specifying the redundant protocol to the ring network. However, this redundant protocol has the below-mentioned problems to be solved. That is, since the control frame is transmitted from only one-sided port (namely, only one direction) of the monitoring apparatus provided in the ring network, in such a case that a failure of one direction happens to occur in a link which constitutes the ring network, the monitoring apparatus judges that a ring status is under abnormal condition, and thus, permits the opposite-sided port to relay a user frame, so that a ring status of the other direction is brought into a loop status. Furthermore, in another network structure that a plurality of ring networks are coupled to each other and coupled apparatuses become redundant, there is such a limitation that an apparatus at the coupling portion must be a monitoring apparatus. As a consequence, there is another problem to be solved that the network structure cannot have a flexible network structure. Also, as a problem specific to a ring network, in the case that the plurality of ring networks are coupled to each other and apparatuses of coupling portions are made redundant, if a failure happens to occur in a link between the apparatuses made redundant, then the following problem occurs. That is, in the coupled ring networks, there are certain possibilities that a large loop structure may be formed.
An object of the present invention has been made to solve the above-described problems, and therefore, is to provide an apparatus capable of realizing high-speed switching of communication routes, capable of preventing an occurrence of a loop which is caused by an one-directional failure of a link within a ring network, and capable of realizing a simple and flexible structure constructed by coupling a plurality of ring networks, and further, capable of avoiding a construction of a loop caused by a failure of a specific link.
To achieve the above-described object, an apparatus according to the present invention is featured as follows: That is, as apparatuses which construct a ring network, such a structure made of a monitoring apparatus and other relay apparatuses is employed. The monitoring apparatus monitors only a ring status of such a ring network to which the own monitoring apparatus belongs. The monitoring apparatus periodically transmits health check frames from ring ports thereof respectively so as to monitor whether or not the transmitted health check frame has been received by the ring port located opposite thereto. Also, in a structure of a multi-ring network having a shared link, auxiliary health check frames are periodically transmitted from two sets of shared apparatuses respectively with respect to a monitoring apparatus of a shared link non-monitored ring network. The monitoring apparatus also monitors the auxiliary health check frames in combination with the health check frames in order to avoid an occurrence of a loop which bridges a plurality of ring networks. Since the above-described apparatus is provided, the above-described object may be achieved.
In accordance with the present invention, the monitoring apparatus which constructs the ring network judges whether or not relaying of a user frame is permitted by checking whether or not the periodically transmitted health check frame is received. As a consequence, the high-speed switching of the routes can be realized.
BRIEF DESCRIPTION OF THE DRAWINGSFIG. 1A andFIG. 1B are diagrams for schematically showing a monitoring apparatus employed in a ring network constructing apparatus according to an embodiment of the present invention.
FIG. 2A andFIG. 2B are diagrams for indicating a relay apparatus employed in the ring network constructing apparatus of the embodiment.
FIG. 3A andFIG. 3B are diagrams for representing a monitoring apparatus in a shared link non-monitored ring network of the ring network constructing apparatus of the embodiment.
FIG. 4A andFIG. 4B are diagrams for showing a relay apparatus (which is commonly operable as shared apparatus) in the shared link non-monitored ring network of the ring network constructing apparatus of the embodiment.
FIG. 5A andFIG. 5B are diagrams for showing a relay apparatus in the shared link non-monitored ring network of the ring network constructing apparatus of the embodiment.
FIG. 6 is a structural diagram of a single ring network with employment of the ring network constructing apparatus of the embodiment.
FIG. 7 is another structural diagram of the single ring network with employment of the ring network constructing apparatus of the embodiment.
FIG. 8 is a structural diagram of a multi-ring network (having no shared link) with employment of the ring network constructing apparatus of the embodiment.
FIG. 9 is another structural diagram of the multi-ring network (having no shared link) with employment of the ring network constructing apparatus of the embodiment.
FIG. 10 is a structural diagram of a multi-ring network (having shared link) with employment of the ring network constructing apparatus of the embodiment.
FIG. 11 is another structural diagram of the multi-ring network (having shared link) with employment of the rink network constructing apparatus of the embodiment.
FIG. 12 is a structural diagram of a multi-ring network (having no shared link) with employment of the ring network constructing apparatus of the embodiment.
FIG. 13 is a structural diagram of a multi-ring network (having no shared link) with employment of the ring network constructing apparatus of the embodiment.
FIG. 14 is a structural diagram of a multi-ring network (having no shared link) with employment of the ring network constructing apparatus of the embodiment.
DETAILED DESCRIPTION OF THE INVENTIONReferring now to drawings, a detailed description is made of a ring network constructing apparatus according to an embodiment of the present invention.
FIG. 1A andFIG. 1B schematically show an apparatus101 (will be referred to as “monitoring apparatus” hereinafter) which monitors a ring network employed in a ring network constructing apparatus according to an embodiment of the present invention. Themonitoring apparatus101 constitutes a portion of a ring network indicated, for example, inFIG. 6, and corresponds to anapparatus601 which monitors this ring network. The ring network is arranged by themonitoring apparatus601 andother relay apparatuses602,603, and604. Although a detailed description as toFIG. 6 will be made later, in this example, for instance, operations of themonitoring apparatus601 in the ring network ofFIG. 6 are described with reference toFIG. 1A andFIG. 1B. Themonitoring apparatus101 contains two pieces of ports (will be referred to as “ring ports” hereinafter)102 and103, which constitute the same ring network. Themonitoring apparatus101 periodically transmits acontrol frame111 from thering port102 so as to monitor whether or not ahealth check frame113 via other apparatuses which constitute the ring network has been received at thering port103. The control frame111 (will be referred to as “health check frame” hereinafter) monitors such a status (will be referred to as “ring status” hereinafter) as to whether or not the ring network has performed a communicating operation under normal condition, and after thehealth check frame113 has been received, thishealth check frame113 is discarded. Similarly, themonitoring apparatus101 periodically transmits ahealth check frame112 from thering port103 so as to monitor whether or not thishealth check frame114 has been received at thering port102, and after thishealth check frame114 has been received, the receivedhealth check frame114 is discarded. While any one of thehealth check frame113 and thehealth check frame114 is being received, themonitoring apparatus101 judges that the ring status is under normal condition, and brings thering port103 into such a condition that a user frame is not transmitted/received in order not generate a loop (FIG. 1A). When both the health check frames113 and114 cannot be received, themonitoring apparatus101 judges that the ring status is under abnormal condition, and transfers thering port103 to such a condition that the user frame is transmitted/received (FIG. 1B). In this example, thering port103 controls whether or not the transmission/reception of the user frame are permitted in response to the ring status. Alternatively, thering port102 may control whether or not the transmission/reception of the user frame are permitted in response to the ring status. A ring port for controlling whether or not the transmission/reception of the user frame are permitted may be automatically determined based upon the numbers of the two ring ports, or may be alternatively designated based upon a configuration by a user.
It should also be noted that in themonitoring apparatus101, the health check frames111 and112 are periodically transmitted from the tworing ports102 and103, and the reception of the health check frames111 and112 are monitored at the ring ports which are different from those used in the frame transmissions. Alternatively, the health check frames111 and112 may be periodically transmitted only from any one of thering ports102 and103, and the received health check frames111 and112 may be monitored only at the other ring port thereof. Also, at this time, while the health check frames111 and112 can be received at the other ring port, themonitoring apparatus101 may judge that the ring status is under normal condition, and defines this ring status as such a status (which is painted in solid condition in drawings) that a user frame is not transmitted/received at any one of these tworing ports102 and103. Also, another status that the user frame is transmitted/received at the other ring port thereof is indicated in a blank condition. When the health check frames112 and113 cannot be received at the other ring port, themonitoring apparatus101 judges that the ring status is under abnormal condition, and changes such a ring port from which the transmissions/reception of the user frame are not permitted into a condition under which the user frame is transmitted/received. Thereafter, themonitoring apparatus102 clears MAC address information and transmits flush request frames from the tworing ports102 and103, and when this flush request frame is received at the opposite ring port, the received flush request frame may be discarded.
Further, in the case that in themonitoring apparatus101, the ring status is changed (ring status is changed from normal condition to abnormal condition, or from abnormal condition to normal condition) by permitting, or not permitting the receptions of the health frames113 and114, themonitoring apparatus101 transmits a control frame131 (will be referred to as “flush request frame” hereinafter) from thering port102 with respect to other apparatus (will be referred to as “relay apparatus” hereinafter) which constitutes the ring network, and when aflush request frame132 is received at thering port103, themonitoring apparatus101 discards the receivedflush request frame132. The above-describedcontrol frame131 instructs to clear MAC address information which has been studded by receiving a user frame with respect to the relay apparatus. Similarly, in such a case that themonitoring apparatus101 transmits aflush request frame133 from thering port103 and receives aflush request frame134 at thering port102, themonitoring apparatus101 discards the receivedflush request frame134. In themonitoring apparatus101, when the ring status is changed, thismonitoring apparatus101 also clears the MAC address information which has been learned by receiving the user frame. Generally speaking, MAC address information which has been learned by receiving a user frame implies, for example, such MAC address information that port numbers for transferring combinations between MAC address and VLAN information to a key have been held as a table.
FIG. 2A andFIG. 2B show anotherrelay apparatus201 which constitutes the ring network. Therelay apparatus201 corresponds to, for example,relay apparatuses602,603, and604 other than amonitoring apparatus601, which constitute a ring network shown inFIG. 6. Therelay apparatus201 contains tworing ports202 and203, which constitute the same ring network. When ahealth check frame211 transmitted from a monitoring apparatus is received at thering port202, therelay apparatus201 relays ahealth check frame212 to thering port203. Similarly, when ahealth check frame213 is received at thering port203, therelay apparatus201 relays ahealth check frame214 to thering port202 provided on the opposite side (FIG. 2A). In therelay apparatus201, only health check frames are relayed, but are not monitored.
Also, for example, in such a case that due to a failure of the ring network, a ring port to which a received health check frame is transferred is brought into a link down condition, the health check frame cannot be transferred to a succeeding apparatus from a relay apparatus brought into the link down condition. As a result, the health check frame is not transferred up to a monitoring apparatus. In this case, the monitoring apparatus judges that the ring status is under abnormal condition. However, if the network failure is recovered and the link down condition is changed into a link up condition, then a health check frame can be again transferred. As a result, the monitoring apparatus judges that the ring status becomes a normal condition.
When aflush request frame221 transmitted from a monitoring apparatus is received at thering port202 in therelay apparatus201, therelay apparatus201 relays aflush request frame222 to thering port203, and at the same time, clears MAC address information which has been learned by receiving a user frame. Similarly, when aflush request frame223 transmitted from a monitoring apparatus is received at thering port203 in therelay apparatus201, therelay apparatus201 relays aflush request frame224 to thering port202 provided on the opposite side, and at the same time, clears the MAC address information which has been learned by receiving the user frame (FIG. 2B). It should also be noted that when therelay apparatus201 has already cleared the MAC address information by receiving the previousflush request frame221, therelay apparatus201 does not clear the MAC address information by receiving theflush request frame223.
When either thering port202 or thering port203 has been brought into the link down condition, the condition of therelay apparatus201 is brought into such a condition that a control frame and a user frame are not relayed. However, when this link down condition is recovered to a link up condition, the condition of therelay apparatus201 is changed into such a condition that the control frame is relayed, and the user frame is not relayed. This reason is given as follows: That is, when the condition of therelay apparatus201 is brought into such a condition that therelay apparatus201 relays the user frame just after the link down condition is recovered to the link up condition, if the monitoring apparatus has still set to such a condition that one of thering ports202 and203 is used to transmit/receive the user frame, then a loop may occur. Then, under this condition, when therelay apparatus201 receives any one of the flush request frames221 and223 transmitted from the monitoring apparatus, the condition of therelay apparatus201 is changed into such a condition that the user frame is relayed. Otherwise, even when a flush request frame has not yet been received, after a predetermined time has passed, the condition of therelay apparatus201 is changed into the condition that the user frame is relayed.
Next, a description is made of a network structure in which a plurality of ring networks have been coupled to each other, which will be referred to as a multi-ring network hereinafter. The multi-ring network has been arranged in such a manner that, for instance, as represented inFIG. 10, two sets ofring networks1061 and1062 have been coupled to each other by employing sharedapparatuses1002 and1003. In this case, 2 sets, or more sets of ring networks to be coupled to each other may be alternatively employed, and 2 pieces, or more pieces of shared apparatuses may be alternatively employed. In the tworing networks1061 and1062,monitoring apparatuses1001 and1005 are present respectively. While control frames which are transmitted by themonitoring apparatuses1001 and1005 are transferred only within therespective ring networks1061 and1062, both the control frames of themonitoring apparatuses1001 and1005 are transferred between the sharedapparatus1002 and the shared apparatus1003 (will be referred to as “shared link” hereinafter). In the present embodiment, while themonitoring apparatuses1001 and1005 monitor ring status of therespective ring networks1061 and1062, as to the shared link, only themonitoring apparatus1001 monitors, whereas themonitoring apparatus1005 monitors ring status of the ring network other than the shared link within thering network1062. As a consequence, these ring networks are divided into a ring network (will be referred to as “shared link monitored ring network” hereinafter)1061 which monitors the ring status containing also the shared link, and another ring network (will be referred to as “shared link non-monitored ring network” hereinafter)1062 which monitors the ring status which do not contain the shared link. At this time, the shared link monitoredring network1061 and the shared linknon-monitored ring network1062 correspond to independent ring networks respectively. It should also be noted that a detailed description as toFIG. 10 will be discussed later.
Next, a description is made of operations of the respective apparatuses which constitute the multi-ring network. Since the shared link monitoredring network1061 is identical to such a ring network shown inFIG. 6, themonitoring apparatus1001 performs the same operation as that of themonitoring apparatus101 described inFIG. 1A andFIG. 1B, whereas the sharedapparatuses1002 and1003, and therelay apparatus1004 perform the same operations as those of therelay apparatus201 explained inFIG. 2A andFIG. 2B. As a result, the shared link monitoredring network1061 is monitored.
Next, a description is made of operations of the respective apparatuses employed in the shared linknon-monitored ring network1062.
FIG. 3A andFIG. 3B indicate amonitoring apparatus301 provided in the shared linknon-monitored ring network1062. Themonitoring apparatus301 contains two pieces ofring ports302 and303, which constitute the same ring network. Themonitoring apparatus301 periodically transmits a control frame311 (will be referred to as “health check frame” hereinafter) for monitoring a ring status from thering port302 via a relay apparatus so as to monitor whether or not ahealth check frame312 has been received at thering port302, and after thehealth check frame312 has been received, thishealth check frame312 is discarded. Similarly, themonitoring apparatus301 periodically transmits ahealth check frame313 from thering port303 so as to monitor whether or not thishealth check frame314 has been received at thering port302, and after thishealth check frame314 has been received, the receivedhealth check frame314 is discarded. Also, in combination with this operation, themonitoring apparatus301 monitors both an auxiliary health check frame (will be referred to as “auxiliary health check frame” hereinafter)315, and another auxiliaryhealth check frame316, and after these auxiliary health check frames315 and316 have been received, these received auxiliary health check frames315 and316 are discarded. The auxiliary health check frames315 and316 are periodically transmitted from a sharedapparatus401 which will be discussed later inFIG. 4A andFIG. 4B with respect to thering port302 and to thering port303 respectively. While any one of thehealth check frame312 and thehealth check frame314 is being received, themonitoring apparatus301 judges that the ring status is under normal condition, and sets thering port302 into such a condition that a user frame is not transmitted/received (FIG. 3A). Although both the health check frames312 and314 cannot be received, while both the auxiliary health check frames315 and316 are being received, themonitoring apparatus301 judges that the ring status is under normal condition, and sets thering port303 to such a condition321 that the user frame is not transmitted/received, since it is conceivable that the apparatuses other than the shared link which should be monitored are under normal conditions. When both the health check frames312 and314 cannot be received, but also, any one of the auxiliary health check frames315 and316 cannot be received, themonitoring apparatus301 judges that the ring status is under abnormal condition, and thus, transfers thering port303 to such a condition322 that the user frame is transmitted/received (FIG. 3B). In this case, although thering port303 controls whether or not the transmission/reception of the user frame are permitted in response to the ring status, thering port302 may control whether or not the transmission/reception of the user frame are permitted in response to the ring status. A ring port for controlling whether or not the transmission/reception of the user frame are permitted may be automatically determined based upon the numbers of the two ring ports, or may be alternatively designated based upon a configuration by a user.
Further, in the case that in themonitoring apparatus301, the ring status is changed (ring status is changed from normal condition to abnormal condition, or from abnormal condition to normal condition) by permitting, or not permitting the receptions of the health frames312 and314, themonitoring apparatus301 transmits a control frame (will be referred to as “flush request frame” hereinafter)331 from thering port302 with respect to other apparatuses which constitute the ring network, and when aflush request frame332 is received at thering port303, themonitoring apparatus301 discards the receivedflush request frame332. The above-describedcontrol frame331 instructs to clear MAC address information which has been learned by receiving a user frame with respect to other apparatuses. Similarly, in such a case that themonitoring apparatus301 transmits aflush request frame333 from thering port303 and receives aflush request frame334 at thering port302, themonitoring apparatus301 discards the receivedflush request frame334. In themonitoring apparatus301, thismonitoring apparatus301 also clears the MAC address information which has been learned by receiving the user frame.
FIG. 4A andFIG. 4B show ashared apparatus401 employed in the shared linknon-monitored ring network1062. The sharedapparatus401 has tworing ports402 and403, which constitute the same ring network. Thering port403 corresponds to a port which is used as a shared link with another ring network. The sharedapparatus401 periodically transmits an auxiliaryhealth check frame415 from thering port402 toward themonitoring apparatus301 of the ring network, which does not monitor the shared link. Also, since the sharedapparatus401 also constitutes the relay apparatus, when ahealth check frame411 transmitted from themonitoring apparatus301 is received at thering port402, the sharedapparatus401 relays ahealth check frame412 to thering port403. Also, similarly, when ahealth check frame413 is received at thering port403, the sharedapparatus401 relays ahealth check frame414 to thering port402 provided on the opposite side (FIG. 4A).
It should also be noted that such a fact whether or not a health check frame transmitted from themonitoring apparatus301 of the shared linknon-monitored ring network1062 has been received may be alternatively superimposed on the auxiliaryhealth check frame415 which is transmitted from the sharedapparatus401. Also, while the sharedapparatus401 does not periodically transmit the auxiliaryhealth check frame415 which is transmitted from the sharedapparatus401, the sharedapparatus401 may alternatively transfer such a health check frame transmitted from themonitoring apparatus301 to the next apparatus, and also may fold back this transferred health check frame, and then, may alternatively transmit the folded health check frame to themonitoring apparatus301 as the auxiliaryhealth check frame415.
Also, when the sharedapparatus401 receives aflush request frame421 at thering port402, which is transmitted from themonitoring apparatus301, the sharedapparatus401 relays aflush request frame422 to thering port403, and at the same time, clears MAC address information which has been learned by receiving a user frame. Similarly, when the sharedapparatus401 receives aflush request frame423 at thering port403, the sharedapparatus401 relays aflush request frame424 to thering port402 provided on the opposite side, and at the same time, clears the MAC address information. It should be noted that when the sharedapparatus401 has already cleared the MAC address information by receiving the previousflush request frame421, the sharedapparatus401 does not clear the MAC address information by receiving theflush request frame423.
When either thering port402 or thering port403 has been brought into the link down condition, the condition of the sharedapparatus401 is brought into such a condition that a control frame and a user frame are not relayed. However, when this link down condition is recovered to a link up condition, the condition of the sharedapparatus401 is changed into such a condition that the control frame is relayed, and the user frame is not relayed. This reason is given as follows: That is, when the condition of the sharedapparatus401 is brought into such a condition that the sharedapparatus401 relays the user frame just after the link down condition is recovered to the link up condition, if the monitoring apparatus has still set to such a condition that one of thering ports402 and403 is used to transmit/receive the user frame, then a loop may occur. Then, under this condition, when the sharedapparatus401 receives any one of the flush request frames421 and423 transmitted from the monitoring apparatus, the condition of the sharedapparatus401 is changed into such a condition that the user frame is relayed. Otherwise, even when a flush request frame has not yet been received, after a predetermined time has passed, the condition of the sharedapparatus401 is changed into the condition that the user frame is relayed.
FIG. 5A andFIG. 5B indicate arelay apparatus501 provided in the shared linknon-monitored ring network1062. Therelay apparatus501 contains tworing ports502 and503, which constitute the same ring network. When therelay apparatus501 receives either ahealth check frame511 transmitted from a monitoring apparatus or an auxiliaryhealth check frame515 transmitted from a shared apparatus, therelay apparatus501 relays either ahealth check frame512 or ahealth check frame516 to thering port503. Similarly, when therelay apparatus501 receives thehealth check frame513 at thering port503, therelay apparatus501 relays ahealth check frame514 to thering port202 located opposite side thereof (FIG. 5A).
Also, when therelay apparatus501 receives aflush request frame521 at thering port502, which is transmitted from themonitoring apparatus501, therelay apparatus501 relays aflush request frame522 to thering port503, and at the same time, clears MAC address information which has been learned by receiving a user frame. Similarly, when therelay apparatus501 receives aflush request frame523 at thering port503, therelay apparatus501 relays aflush request frame524 to thering port502 provided on the opposite side, and at the same time, clears the MAC address information. It should be noted that when therelay apparatus501 has already cleared the MAC address information by receiving the previousflush request frame521, therelay apparatus501 does not clear the MAC address information by receiving theflush request frame523.
When either thering port502 or thering port503 has been brought into the link down condition, the condition of therelay apparatus501 is brought into such a condition that a control frame and a user frame are not relayed. However, when this link down condition is recovered to a link up condition, the condition of therelay apparatus501 is changed into such a condition that the control frame is relayed, and the user frame is not relayed. This reason is given as follows: That is, when therelay apparatus501 is brought into such a condition that therelay apparatus501 relays the user frame just after the link down condition is recovered to the link up condition, if the monitoring apparatus has still set to such a condition that one of thering ports502 and503 is used to transmit/receive the user frame, then a loop may occur. Then, under this condition, when therelay apparatus501 receives any one of the flush request frames521 and523 transmitted from the monitoring apparatus, the condition of therelay apparatus501 is changed into such a condition that the user frame is relayed. Otherwise, even when a flush request frame has not yet been received, after a predetermined time has passed, the condition of therelay apparatus501 is changed into the condition that the user frame is relayed.
In therelay apparatus501 and the sharedapparatus401, when a link condition of a ring port is changed, a control frame (will be referred to as “link condition change notification frame” hereinafter) may be alternatively transmitted from a ring port under normal condition, while this control frame notifies the change in the link condition. Also, themonitoring apparatus301 may not only monitor the ring status based upon either the health check frame or the auxiliary health check frame, but also may monitor the ring status, while including the reception of the link condition change notification frame which is transmitted from either therelay apparatus501 or theshred apparatus401 in addition to these health check frames. It is so assumed that the transmission time periods as to the health check frames and the auxiliary health check frames, and the time durations required for judging that the ring status are under the abnormal conditions since the health check frames are not yet received may be changed based upon a configuration by a user. These transmission time periods and time durations have been described in themonitoring apparatuses101 and301 shown inFIG. 1A,FIG. 1B,FIG. 3A,FIG. 3B; and also in the sharedapparatus401 shown inFIG. 4A,FIG. 4B.
While themonitoring apparatuses101 and301 explained inFIG. 1A,FIG. 1B,FIG. 3A, andFIG. 3B are equipped with hardware circuits, these hardware circuits transmit/receive various sorts of control frames and user frames from ring ports, and also, monitor whether or not either a health check frame or an auxiliary health check frame is received within a predetermined time at a ring port. In such a case that either the health check frame or the auxiliary health check frame has not yet been received for the predetermined time, the monitoringapparatuses101 and301 judge that a ring status is brought into an abnormal condition. Also, while themonitoring apparatuses101 and301 monitor the receptions of the health check frame and the auxiliary health check frame under such a condition that the ring status is under abnormal condition, themonitoring apparatus101 and301 judge that the ring status is under normal condition by receiving these health check frames. It should also be noted that the above-described functions may be alternatively realized by employing software.
While therelay apparatuses201 and501 shown inFIG. 2A,FIG. 2B,FIG. 5A, andFIG. 5B, and also, the sharedapparatus401 shown inFIG. 4A andFIG. 4B are equipped with hardware circuits, these hardware circuits transmit/receive various sorts of control frames and user frames, and also, judge that the ring status is changed.
It should also be noted that the above-described functions may be alternatively realized by employing software.
Based uponFIG. 6, a description is made of operations as to respective apparatuses employed in a single ring network (will be referred to as “single ring network” hereinafter). InFIG. 6, amongapparatuses601 to604, according to an embodiment of the present invention, which constitute the single ring network, a monitoring apparatus is defined as601, and relay apparatuses are defined as602 to604. Although the single ring network is constituted by employing4 sets of theapparatuses601 to604, there is no limitation as to a total number of these apparatuses.
Themonitoring apparatus601 periodically transmits health check frames611 and615 from two ring ports thereof. These health check frames611 and615 are relayed in therelay apparatuses602 to604, and then, the health check frames611 and615 are received at ring ports which are different from the above-described ring ports for transmitting these health check frames611 and615. In the case that thehealth check frame611, or615 is being received at any one of these two ring ports, themonitoring apparatus601 judges that a ring status is under normal condition, and sets such a condition that a user frame is not transmitted/received at aring port621 thereof.
FIG. 7 indicates a condition that afailure751 occurs in the single ring network ofFIG. 6. When thefailure751 happens to occur, ahealth check frame711 transmitted from amonitoring apparatus701 cannot be relayed from arelay apparatus702 to a succeeding apparatus. Similarly, ahealth check frame715 cannot be also relayed from arelay apparatus703 to a succeeding apparatus. As a result, themonitoring apparatus701 cannot receive the health check frames711 and715 at the two ring ports thereof, so that themonitoring apparatus701 judges that a ring status is under abnormal condition, and changes aring port721 thereof into such a condition that a user frame can be transmitted/received. Also, themonitoring apparatus701 clears MAC address information which has been learned by receiving the user frame, and transmits flush request frames731 and735 from the two ring ports thereof with respect toother relay apparatuses702 to704. The relay apparatuses702 to704 which receive either theflush request frame731 or theflush request frame735 clear the MAC address information which has been learned by receiving the user frame.
When the failure is recovered from the condition where the failure has occurred inFIG. 7 and then the failure condition is returned to the normal condition ofFIG. 6, themonitoring apparatus601 receives any one of the health check frames611 and615 which are transmitted by themonitoring apparatus601 at the ring port. As a result, themonitoring apparatus601 judges that the ring status becomes the normal condition, and again changes such a ring port where the transmission/reception of the user frame have been permitted into another condition where the user frame is not transmitted/received. Also, themonitoring apparatus601 clears the MAC address information which has been learned by receiving the user frame, and also, transmits a flush request frame from the two ring ports with respect toother relay apparatuses602 to604 in a similar manner when thefailure751 occurs inFIG. 7. The relay apparatuses602 to604 which receive this flush request frame clear the MAC address information which has been learned by receiving the user frame.
Also, the ring ports of therelay apparatuses602 and603 where a failure of a link has occurred can transmit/receive such a control frame as a health check frame and a flush request frame when the failure is recovered, and are brought into such a condition that a user frame cannot be transmitted/received. When therelay apparatuses602 and603 receive the flush request frame, therelay apparatuses602 and603 change the ring ports where the transmission/receptions of the user frames cannot be permitted into such a condition that the user frames can be transmitted/received. Otherwise, even when the flush request frame is not received, after a predetermined time has passed, therelay apparatuses602 and603 changes the present condition into such a condition that the user frames can be transmitted/received.
Referring now toFIG. 8, a description is made of operations as to respective apparatuses provided in such a structure that a plurality of ring networks are coupled to each other, and furthermore, a coupling apparatus has not been made redundant. This structure corresponds to such a network structure that there is no link (namely, shared link) which is shared by the plural ring networks. InFIG. 8, although a total number of the coupled ring networks is selected to be 2, there is no limitation with respect to a total quantity of these ring networks to be coupled to each other.
InFIG. 8, amongapparatuses801 to803 which constitute aring network841, a monitoring apparatus is defined as801, and relay apparatuses are defined as802 and803. Also, amongapparatuses802,804, and805, which constitute anotherring network842, a monitoring apparatus is defined as804, and relay apparatuses are defined as802 and805. While therelay apparatus802 constitutes a coupling apparatus of thering networks841 and842, therelay apparatus802 has constructed thering networks841 and842 by two different ring ports with respect to each of thering networks841 and842. In this coupling apparatus, control frames which are transmitted/received in therespective ring networks841 and842 are relayed by being closed in therespective ring networks841 and842, but are not transferred to the other ring network. Themonitoring apparatus801 monitors a ring status of thering network841, and sets aring port831 thereof to such a condition that a user frame is not transmitted and received. Themonitoring apparatus804 monitors a ring status of thering network842, and sets aring port832 thereof to such a condition that a user frame is not transmitted and received. As a result, thering networks841 and842 can be independently operated.
FIG. 9 indicates a structure made in the case that an apparatus for coupling two ring networks is used as a monitoring apparatus. InFIG. 9, amongapparatuses901 to903 which constitute aring network941, a monitoring apparatus is defined as901, and relay apparatuses are defined as902 and903. Also, amongapparatuses902,904, and905, which constitute anotherring network942, a monitoring apparatus is defined as901, and relay apparatuses are defined as904 and905. While the coupled apparatus becomes themonitoring apparatus901, this coupled apparatus constitutes thering networks941 and942 at two different ring ports with respect to each of thering networks941 and942, and a control frame is not transferred to theother ring network941, or942. Themonitoring apparatus901 monitors thering network941, and sets aring port931 thereof to such a condition that a user frame is not transmitted and received. At the same time, themonitoring apparatus901 monitors thering network942, and sets aring port932 thereof to such a condition that a user frame is not transmitted and received. As a result, thering networks941 and942 can be independently operated.
Referring now toFIG. 10, a description is made of operations as to respective apparatuses provided in such a structure that a plurality of ring networks are coupled to each other, and furthermore, a coupling apparatus has been made redundant. InFIG. 10, although a total number of the coupled ring networks is selected to be2, there is no limitation with respect to a total quantity of these ring networks to be coupled to each other.
InFIG. 10, amongapparatuses1001 to1004 which constitute aring network1061, a monitoring apparatus is defined as1001, and relay apparatuses are defined as1002 to1004. Also, amongapparatuses1002,1003,1005, and1006, which constitute anotherring network1062, a monitoring apparatus is defined as1005, and relay apparatuses are defined as1002,1003, and1006. While therelay apparatus1002 and1003 constitute coupling apparatuses of thering networks1061 and1062, therelay apparatuses1002 and1003 have been constructed as having redundant structures (will be referred to as “shared apparatus” hereinafter). Also, a link between the sharedapparatuses1002 and1003 constitutes such a link (will be referred to as “shared link” hereinafter) which is shared by thering networks1061 and1062. In the sharedapparatuses1062 and1063, control frames which are transmitted/received in therespective ring networks1061 and1062 are relayed by being closed in therespective ring networks1061 and1062, but are not transferred to the other ring network. Within the two ring networks, thering network1061 is defined as such a ring network (will be referred to as “shared link monitored ring network” hereinafter) which monitors a ring status containing also the shared link, whereas thering network1062 is defined as such a ring network (will be referred to as “shared link non-monitored ring network” hereinafter) which monitors a ring status which does not contain the shared link. Themonitoring apparatus1001 of the shared link monitoredring network1061 periodically transmitshealth check frames1011 and1015 from two ring ports thereof. Thesehealth check frames1011 and1015 are relayed via therelay apparatuses1002 to1004, and themonitoring apparatus1002 receives thehealth check frames1011 and1015 at different ring ports from the ring ports for transmitting thesehealth check frames1011 and1015. Within the two ring ports, while the health check frame is being received at any one of these ring ports, themonitoring apparatus1002 judges that the ring status is under normal condition, and sets such a condition that a user frame is not transmitted/received at aring port1051.
Themonitoring apparatus1005 of the shared linknon-monitored ring network1062 periodically transmitshealth check frames1021 and1025 from two ring ports thereof. Also, since the sharednodes1002 and1003 do not monitor the shared link, the sharednodes1002 and1003 periodically transmit auxiliaryhealth check frames1031 and1032 from such ring ports which are located opposite to the ring ports which become the shred links toward themonitoring apparatus1005. The auxiliaryhealth check frames1031 and1032 are not transmitted to the shared link monitoredring network1061. Thesehealth check frames1021 and1025, and these auxiliaryhealth check frames1031 and1032 are relayed in the relay apparatuses (including shared apparatus)1002,1003, and1006, and then, themonitoring apparatus1005 receives thesehealth check frames1021 and1025, and these auxiliaryhealth check frames1031 and1032 at the two ring ports thereof. Within the two ring ports, while thehealth check frames1021 and1025 are being received at any one of these two ring ports, or while the auxiliaryhealth check frames1031 and1032 transmitted from the sharedapparatuses1002 and1003 are being received at the two ring ports, themonitor apparatus1005 judges that the ring status is under normal condition, and sets such a condition that the user frame is not transmitted and received at thering port1052.
FIG. 11 indicates a condition that afailure1171 occurs in the shared link monitoredring network1061 ofFIG. 10. When thefailure1171 happens to occur, ahealth check frame1111 transmitted from amonitoring apparatus1101 cannot be relayed from arelay apparatus1103 to a succeeding apparatus. Similarly, ahealth check frame1115 cannot be also relayed from arelay apparatus1104 to a succeeding apparatus. As a result, themonitoring apparatus1101 cannot receive thehealth check frames1111 and1115 at the two ring ports thereof, so that themonitoring apparatus1101 judges that a ring status is under abnormal condition, and changes aring port1151 thereof into such a condition that a user frame can be transmitted/received. Also, themonitoring apparatus1101 clears MAC address information which has been learned by receiving the user frame, and transmits flush request frames1111 and1115 from the two ring ports thereof with respect toother relay apparatuses1102 to1104. Therelay apparatuses1102 to1104 which receive the flush request frames111 and1115 clear the MAC address information which has been learned by receiving the user frame. At this time, since themonitoring apparatus1105 of the shared linknon-monitored ring network1162 has already received thehealth check frames1111 and1115 at the two ring ports thereof, thismonitoring apparatus1105 judges that there is no change in the ring status, and is continuously operated.
When the failure is recovered from the condition where the failure has occurred inFIG. 11 and then the failure condition is returned to the normal condition ofFIG. 10, themonitoring apparatus1001 receives any one of thehealth check frames1011 and1015 which are transmitted by themonitoring apparatus1001 at the ring port. As a result, themonitoring apparatus1001 judges that the ring status becomes the normal condition, and again changes such a ring port where the transmission/reception of the user frame have been permitted into another condition where the user frame is not transmitted/received. Also, themonitoring apparatus1001 clears the MAC address information which has been learned by receiving the user frame, and also, transmits a flush request frame from the two ring ports with respect toother relay apparatuses1002 to1004 in a similar manner when thefailure751 occurs inFIG. 7. Therelay apparatuses1002 to1004 which receive this flush request frame clear the MAC address information which has been learned by receiving the user frame. At this time, since themonitoring apparatus1005 of the shared linknon-monitored ring network1062 has continuously received thehealth check frames1124 and1125, thismonitoring apparatus1005 is continuously operated.
FIG. 12 represents such a condition that a sharedlink failure1271 ofFIG. 10 happens to occur. When the sharedlink failure1271 occurs, ahealth check frame1211 transmitted from amonitoring apparatus1201 of the shared link monitorednetwork1261 cannot be relayed from arelay apparatus1202 to a succeeding apparatus. Similarly, ahealth check frame1215 cannot be also relayed from arelay apparatus1203 to a succeeding apparatus. As a result, themonitoring apparatus1201 cannot receive thehealth check frames1211 and1215 at the two ring ports thereof, so that themonitoring apparatus1201 judges that a ring status is under abnormal condition, and changes aring port1251 thereof into such a condition that a user frame can be transmitted/received. Also, themonitoring apparatus1201 clears MAC address information which has been learned by receiving the user frame, and transmits flush request frames1211 and1215 from the two ring ports thereof with respect toother relay apparatuses1202 to1204. Therelay apparatuses1202 to1204 which receive the flush request frames1211 and1215 clear the MAC address information which has been learned by receiving the user frame. At this time, ahealth check frame1221 transmitted from amonitoring apparatus1205 of a shared linknon-monitored network1262 cannot be relayed from therelay apparatus1202 to a succeeding apparatus. Similarly, ahealth check frame1225 cannot be also relayed from therelay apparatus1203 to a succeeding apparatus. However, auxiliaryhealth check frames1231 and1232 transmitted from the sharedapparatuses1202 and1203 can be received at the two ring ports of themonitoring apparatus1205, so that themonitoring apparatus1205 judges that a ring status except for the shred link is under normal condition, and is continuously operated. In accordance with the above-described operations, even when the failure happens to occur in the shared link, it is possible to avoid an occurrence of a large loop which bridges the plurality of ring networks.
When the failure is recovered from the condition where the failure has occurred inFIG. 12 and then the failure condition is returned to the normal condition ofFIG. 10, themonitoring apparatus1001 of the shared link monitoredring network1061 receives any one of thehealth check frames1011 and1015 which are transmitted by themonitoring apparatus1001 at the ring port. As a result, themonitoring apparatus1001 judges that the ring status becomes the normal condition, and again changes such a ring port where the transmission/reception of the user frame have been permitted into another condition that the user frame is not transmitted/received. Also, themonitoring apparatus1001 clears the MAC address information which has been learned by receiving the user frame, and also, transmits a flush request frame from the two ring ports with respect toother relay apparatuses1002 to1004 in a similar manner when thefailure1271 occurs inFIG. 12. Therelay apparatuses1002 to1004 which receive this flush request frame clear the MAC address information which has been learned by receiving the user frame.
At this time, themonitoring apparatus1005 of the shared linknon-monitored ring network1062 also receives thehealth check frames1021 and1025 transmitted from themonitoring apparatus1005 at the two ring ports thereof. It should also be noted that since the ring status remains under the normal condition, the operation is merely continued.
FIG. 13 indicates a condition that afailure1371 occurs in the shared linknon-monitored ring network1062 ofFIG. 10. When thefailure1371 happens to occur, ahealth check frame1321 transmitted from amonitoring apparatus1305 cannot be relayed from arelay apparatus1306 to a succeeding apparatus. Similarly, ahealth check frame1325 cannot be also relayed from arelay apparatus1303 to a succeeding apparatus. As a result, themonitoring apparatus1305 cannot receive thehealth check frames1321 and1325 at the two ring ports thereof. Also, an auxiliary health check frame transmitted from the sharedapparatus1303 cannot also be relayed from therelay apparatus1306 to a succeeding apparatus. An auxiliaryhealth check frame1331 transmitted from the sharedapparatus1302 is received at the ring port of themonitoring apparatus1305. As a result, since thehealth check frames1321 and1325 transmitted from themonitoring apparatus1305 cannot be received at the two ring ports, and further, the auxiliary health check frame transmitted from the sharedapparatus1303 cannot be received at any one of the two ring ports, themonitoring apparatus1305 judges that a ring status is under abnormal condition, and changes aring port1352 thereof into such a condition that a user frame can be transmitted/received. Also, themonitoring apparatus1305 clears MAC address information which has been learned by receiving the user frame, and transmits a flush request frame from the two ring ports thereof with respect toother relay apparatuses1302,1303, and1306. Therelay apparatuses1302,1303, and1306 which receive the flush request frame clear the MAC address information which has been learned by receiving the user frame. At this time, since themonitoring apparatus1301 of the shared link monitoredring network1361 has received the health check frame at the two ring ports, themonitoring apparatus1301 judges that there is no change in the ring status, and the operation thereof is continued.
When thefailure1371 is recovered from the condition where thefailure1371 has occurred inFIG. 13 and then the failure condition is returned to the normal condition ofFIG. 10, themonitoring apparatus1005 receives any one of thehealth check frames1021 and1025 which are transmitted by themonitoring apparatus1005 at the ring port. Otherwise, themonitoring apparatus1005 receives the auxiliaryhealth check frames1031 and1033 which are transmitted from the sharedapparatuses1002 and1002 at the two ring ports respectively. As a result, themonitoring apparatus1005 judges that the ring status becomes the normal condition, and again changes such a ring port where the transmission/reception of the user frame have been permitted into another condition that the user frame is not transmitted/received. Also, themonitoring apparatus1005 clears the MAC address information which has been learned by receiving the user frame, and also, transmits a flush request frame from the two ring ports with respect toother relay apparatuses1002,1003, and1006 in a similar manner when thefailure1371 occurs inFIG. 13. Therelay apparatuses1002,1003 and1006 which receive this flush request frame clear the MAC address information which has been learned by receiving the user frame. At this time, since themonitoring apparatus1005 of the shared link monitoredring network1061 continuously receives the health check frames at the two ring ports thereof, the monitoring apparatus105 continues the operation thereof.
FIG. 14 shows a structure made in the case that a shared apparatus is used as a monitoring apparatus. Amongapparatuses1401 to1404 of an embodiment of the present invention, which constitute a shared link monitoredring network1461, a monitoring apparatus is defined as1401, and relay apparatuses are defined as1402 to1404. Also, amongapparatuses1401, and1404 to1406 of the embodiment of the present invention, which constitute a shared linknon-monitored ring network1462, a monitoring apparatus is defined as1401, and relay apparatuses are defined as1404 to1406. While theapparatus1401 corresponding to the monitoring apparatus of each of thering networks1461 and1462 constitutes one shared apparatus, thismonitoring apparatus1401 independently monitors therespective ring networks1461 and1462. Themonitoring apparatus1401 monitors a ring status as the monitoring apparatus for the shred link monitorednetwork1461, and sets aring port1451 to such a condition that a user frame is not transmitted and received. At the same time, themonitoring apparatus1401 monitors a ring status as the monitoring apparatus for the shared linknon-monitored network1462, and sets aring port1452 to such a condition that a user frame is not transmitted and received. At this time, thering ports1451 and1452 cannot be constituted as ring ports on the shared link side, but must be constituted as ring ports which are located opposite to the ring ports corresponding to the shared link. Also, since themonitoring apparatus1401 of the shared linknon-monitored ring network1462 is commonly employed as the shared apparatus, themonitoring apparatus1401 does not transmit such an auxiliary health check frame so as not to monitor the shared link, but monitors another auxiliaryhealth check frame1431 transmitted from another sharedapparatus1404 in combination with thehealth check frames1421 and1425 which are transmitted by the own sharedapparatus1401. As a result, similar toFIG. 10, thering networks1461 and1462 can be independently operated.
In accordance with the monitoring apparatuses in the present embodiments, the shared apparatuses, and the relay apparatuses of the present invention, the below-mentioned effects can be achieved.
(1) In a monitoring apparatus which constitutes a ring network, the monitoring apparatus judges whether or not a relay of a user frame is permitted by determining whether or not a health check frame is received which is periodically transmitted. As a result, routes can be switched in a high speed.
(2) As previously described, since the monitoring apparatus may merely monitor a reception of a control frame, the system may be easily realized in a hardware fashion. As a result, while there is no unwanted protocol operation, loads given to a CPU can be reduced, so that a stable protocol operation can be realized.
(3) Health check frames are transmitted from two ring ports of a monitoring apparatus which constitutes a ring network, and these health check frames are monitored at the two ring ports (namely, health check frames are monitored in bidirectional manner). As a result, even when a failure happens to occur along one direction in a portion of the link which constitutes the ring network, it is possible to prevent an occurrence of a loop status in connection with the failure.
(4) In the case that a multi-ring network is constructed, the following conditions may be merely set: any one of a monitoring apparatus, a shared apparatus, and a relay apparatus may be merely set with respect to this apparatus, and which port may be merely set to a ring port. As a result, these monitoring apparatus, shared apparatus, and relay apparatus may be simply coupled to the multi-ring network. Also, since operations in the respective rings are independently carried out, a ring network may be readily added to be coupled to the existing ring network without giving an adverse influence to the existing ring network.
(5) In such a case that a multi-ring network is constructed, a structural position of a monitoring apparatus may be freely determined irrespective of such a fact whether or not an apparatus to be coupled takes a redundant structure, so that apparatuses may be arranged in a flexible manner. Also, the apparatus to be coupled is not necessarily a monitoring apparatus.
(6) In the case where a multi-ring network is constructed and an apparatus to be coupled takes a redundant structure, an auxiliary health check frame is transmitted from this apparatus having the redundant structure, and a monitoring apparatus monitors a health check frame in combination with this auxiliary health check frame. As a result, even when a failure happens to occur in a link between apparatuses which are made redundant, it is possible to avoid that a loop is constructed.
It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claim.