CN101645812A - Protection switching method of Ethernet intersecting ring - Google Patents
Protection switching method of Ethernet intersecting ring Download PDFInfo
- Publication number
- CN101645812A CN101645812A CN200910171823A CN200910171823A CN101645812A CN 101645812 A CN101645812 A CN 101645812A CN 200910171823 A CN200910171823 A CN 200910171823A CN 200910171823 A CN200910171823 A CN 200910171823A CN 101645812 A CN101645812 A CN 101645812A
- Authority
- CN
- China
- Prior art keywords
- ring
- link
- port
- fault
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000004224 protection Effects 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000011084 recovery Methods 0.000 claims abstract description 57
- 238000001514 detection method Methods 0.000 claims abstract description 19
- 238000000638 solvent extraction Methods 0.000 claims abstract description 9
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 3
- 230000008569 process Effects 0.000 claims description 32
- 230000005540 biological transmission Effects 0.000 claims description 18
- 230000004044 response Effects 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 abstract description 17
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 230000003111 delayed effect Effects 0.000 abstract 1
- 230000001960 triggered effect Effects 0.000 abstract 1
- 101150007742 RING1 gene Proteins 0.000 description 58
- 208000035217 Ring chromosome 1 syndrome Diseases 0.000 description 58
- 238000010586 diagram Methods 0.000 description 51
- 208000032825 Ring chromosome 2 syndrome Diseases 0.000 description 41
- 230000002776 aggregation Effects 0.000 description 20
- 238000004220 aggregation Methods 0.000 description 20
- 238000012545 processing Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000001681 protective effect Effects 0.000 description 2
- 101150093240 Brd2 gene Proteins 0.000 description 1
- 208000032826 Ring chromosome 3 syndrome Diseases 0.000 description 1
- 208000002991 Ring chromosome 4 syndrome Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000009514 concussion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- -1 send DN (Ring0 Proteins 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
Images
Landscapes
- Small-Scale Networks (AREA)
Abstract
The invention discloses a protection switching method of an Ethernet intersecting ring, which is used for solving the problems of protection switching of link fault in complex network topology. The method comprises the following steps: introducing conceptions like basic ring, derivative ring, basic ring link, cyclopolymerization port and the like; carrying out logical partitioning on the intersecting ring topologies; assigning different priority levels of rings; carrying out bidirectional detection on the link by utilizing an Hello mechanism; realizing rapid and correct fault detection of thelink by combining port link status monitoring; adopting a fault detection mechanism based on the cyclopolymerization port status on the basic ring link or the logic basic ring link; and realizing thelayer upon layer protection switching and recovery of multiple ring topology according to the ring network principle in which the protection switching of the corresponding ring is triggered in link fault and the recovery delayed mechanism is introduced when the link fault is recovered. The adoption of the method in the invention can ensure that N rings intersect with at most N levels of link faultprotection switching, thus giving full play to the reliability of the multiple ring topology; in addition, the fault protection switching is within 50 ms, thus meeting the requirements of carrier-grade Ethernet.
Description
Technical field
The present invention relates to the link protection change technology of data communication field, relate in particular to a kind of Ethernet and intersect the ring protection reverse method.
Background technology
Ethernet ring network is a kind of metropolitan area network technology, and problems such as traditional data net protective capability is weak, failure recovery time is long that it has solved can provide the quick protective feature of 50ms in theory, progressively become one of trend of operation level metropolitan area network development.Looped network protocol core thought all is based on " standard MAC (Media AccessControl; medium access control) exchange+improved spanning tree algorithm+Ethernet fault detection mechanism+simple looped network control protocol "; by the looped network control protocol ring of physics is cracked into the chain of logic, and utilize improved Spanning-Tree Protocol and MAC exchange to finish the protection switching.
IETF (The Internet Engineering Task Force; the Internet engineering duty group) RFC3916 standard definition EAPS (Ethernet Automatic Protection Switching; the Ethernet APS) resist technology; EAPS be Extreme Networks propose be used for strengthening the reliability of ethernet ring network and a kind of technology of robustness; can solve the quick convergence problem of simple ring topological structure, its link switchover convergence time is usually in 50ms.Strictly speaking, EAPS is not the standard that IETF recommends, and is a memorandum, but since the implementation of its simple and flexible, the ethernet ring network technical standard on having come true.EAPS has done detailed regulation to the protection switching technique of monocycle; the equipment supplier has released looped network agreement separately on the EAPS basis; such as RRPP (Rapid Ring Protection Protocol; quick ring protection agreement); ZESR (ZTE Ethernet Smart Ring; in emerging intelligent ethernet ring network technology) etc.; processing and EAPS to the monocycle topology are identical; handling the notion of having used Edge node and assistant edge node when many rings intersect topology; adopt the subring model; the notion that does not have shared link in the subring model; constitute by the main ring of a sealing and several nonocclusive subrings but each many ring topology regarded as; each subring is connected with other ring or subring by the node at two ends, and this model can be supported the topological structure of any multi-ring networks.
ITU-T G.8032 (International Telecommunications Union communication standardization group Ethernet ring protection switching switch) standard definition ethernet ring network automatic protection switching mechanism; monocycle has been made regulation with the fault detect of cross ring/path switching mechanism mutually; be applicable to many ring/ladder networks, can protect in the Ethernet ring or interannular topological on point-to-point, put multiple spot, multiple spot interoperability to multiple spot.When handling multi-ring networks, adopt the shared link model; in the shared link model; each ring topology in encircling is designated as different priorities more; crossing link between ring and the ring is for sharing separately; the protection that can trigger this guilty culprit ring of breaking down of non-shared link is switched in the ring; and the fault of shared link is then switched by that ring protection of comparing the priority triggering high priority that shared link connected.This mechanism relies on the fixed allocation of many ring priority, guarantees that multiring structure correctly switches.
No matter be subring model or shared link model, all avoided many rings to intersect topological link failure trigger protection and switched the super ring of appearance.But, subring is not closed ring in the subring model, need to introduce auxiliary mechanism and carry out fault detect, increased the complexity of agreement, the subring model carries out logical level to phase cross ring topology to be divided, and the main ring link fault announcing can not directly be handled by subring, is that link failure appears in main ring under the guard mode once more if link failure appears in main ring continuously, then need to be detected by subring, often can not satisfy Carrier Ethernet 50ms and switch requirement by auxiliary mechanism; Shared link model strict regulations the order that ring protection is switched during the shared link fault; its purpose is intended to avoid protection to switch the super ring of appearance; and still need auxiliary mechanism to handle when the high priority ring link failure occurred continuously, and not ideal enough equally, be difficult to satisfy the Carrier Ethernet requirement.As seen, subring model and shared link model all fail to bring into play fully the topological reliability of many rings.In addition, during shared link models treated complex network topologies, the logical partitioning of ring is difficulty relatively.
Fig. 1 is existing complex network topologies schematic diagram; as shown in Figure 1; on directly perceived; all links of ring S1-S2-S4-S3 are all shared by other rings; at this moment just can not simply divide ring; need can not appear at the principle division ring of shared link in line with the ring protection port, can divide like this, encircle S7-S8-S2-S1, ring S7-S8-S2-S4-S3-S1, ring S7-S8-S2-S4-S6-S5-S3-S1, encircle S9-S1-S3-S10, ring S11-S12-S4-S2.As can be seen, such division, the relation between ring and the ring is very complicated, and very not directly perceived, and is same, also very complicated on processing logic.
Summary of the invention
In view of this; main purpose of the present invention is to provide a kind of Ethernet to intersect the ring protection reverse method; in order to logical partitioning and the processing complexity that reduces many ring topologys; strengthen its reliability; reduce the required time and the required time of recovery process of protection reversed process of many ring topologys simultaneously, make it satisfy the requirement of Carrier Ethernet.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of Ethernet intersects the ring protection reverse method, and this method comprises:
A, Ethernet phase cross ring topology is carried out logical partitioning, specify different priority, and between adjacent node, periodically mutually send out a Hello frame and set up session and carry out link bidirectional and detect, when detecting link occurs fault, the announcement fault, and send short sequence fault notification message DN;
B, block link during to fault when the node detection in the Ethernet phase cross ring, the announcement fault is to the ring Control Node of place ring, described ring Control Node open loop protection port, and trigger to transmit to upgrade and carry out ring protection and switch.
Wherein, described steps A further comprises: or with node the Link State of port is monitored, when finding that link failure then adopts interrupt mode notice CPU, triggers fault and also removes session.
The above carries out logical partitioning with Ethernet phase cross ring topology and specifically comprises:
To specify a ring in the phase cross ring topology to be divided into basic ring;
The ring of basic ring link and other links composition is divided into the ring of deriving; Wherein, regard the part of basic ring between crossed node as a redundant link and be called the basic ring link;
The cross ring mutually that the described basic ring and the ring of deriving are formed is divided into the logic basic ring.
Wherein, link failure detection specifically comprises:
When detecting link failure, send fault notification message DN, DN must carry the ring sign, DN only transmits on corresponding ring, up to the ring Control Node, when the cyclopolymerization port place node of deriving is received DN, check earlier and whether have this ring sign of deriving, if there is no, then in DN, add the ring sign of deriving, and put the member port Down of the cyclopolymerization port of receiving DN; If exist, then only do to transmit to handle.
The process of described announcement fault comprises:
When the member port of the cyclopolymerization port of deriving links link failure, put the member port Down of the cyclopolymerization port of the link that breaks down, and send fault notification message DN, in DN, add basic ring or logic basic ring sign and the ring of deriving and identify; And when all member ports of the cyclopolymerization port of deriving all were changed to Down, basic ring link or logic basic ring link failure, announcement DN switched for the ring Control Node trigger protection of deriving.
A kind of Ethernet phase cross ring based on the described protection reversed process of claim 1 is recovered reverse method, and this method comprises:
In a, the phase cross ring topology faulty link recovery process, the link point of link failure recovery sends short sequence fault recovery announcement RN message, and the RN message is transmitted in described ring, up to the ring Control Node; When receiving the RN message, the cyclopolymerization port place node of deriving checks wherein whether there is the ring sign of deriving earlier, if do not exist, then in DN, add the ring sign of deriving, put the member port Up of the cyclopolymerization port of the described RN of receiving message, and on the corresponding ring of deriving, announce fault recovery RN message; If exist, then only do to transmit to handle;
B, ring Control Node are received the RN message or detected link-recovery, then start the recovery back off timer of respective rings, and timer recovery delay time is set, and be overtime when described back off timer, then begins to recover to switch.
Wherein, described timer time is Wait-to-restore-time/ (2n+1) s; Wherein: Wait-to-restore-time is that topology is recovered delay time, need be configured according to the real network environment; N is the respective rings numbering.
Wherein, described recovery reversed process further comprises:
It is overtime that the ring Control Node is recovered back off timer; choke ring is protected port, is refreshed and transmit; and transmission fault recovery response message RR; and described RR carries and sends the ring sign; be higher than self and be on the ring of guard mode in respective rings and priority and transmit; high priority ring Control Node receives that RR will suppress the back off timer of self, and the node of receiving the RR message refreshes to be transmitted.
A kind of Ethernet provided by the present invention intersects the ring protection reverse method, has the following advantages:
The present invention adopts the logical partitioning mode of many ring topologys; look many crossing topology of the rings of complexity as a whole; and with all rings all as closed hoop; when break down continuously in the loop chain road; handle the problem of many ring topologys with the principle of monocycle; can give full play to the reliabilities of many ring topologys, making N ring intersect multipotency provides the protection of N level link failure to switch, and can handle the topology protection of any level and switch.Adopt the Ethernet multi-ring networks of the inventive method, protection is switched and can be finished in 50ms, can satisfy the requirement that the Carrier Ethernet error protection is switched fully.
Description of drawings
Fig. 1 is existing complex network topologies schematic diagram;
Fig. 2 is an embodiment of the invention phase cross ring topology schematic diagram;
Fig. 3 is the embodiment of the invention basic ring link and the ring logical schematic of deriving;
Fig. 4 is an embodiment of the invention primary link and from the link structure schematic diagram;
Fig. 5 is the embodiment of the invention complex network basic ring and the topological schematic diagram of ring of deriving;
Fig. 6 is an embodiment of the invention monocycle normal condition schematic diagram;
Fig. 7 is an embodiment of the invention monocycle link failure schematic diagram;
Fig. 8 is a single link failure protection reversed process schematic diagram in the embodiment of the invention;
Fig. 9 is that the many rings of embodiment of the invention complexity intersect the network topology schematic diagram;
Fig. 9 ' is that the many rings of embodiment of the invention complexity intersect the network work view;
Figure 10 is a link process fault detection schematic diagram in the embodiment of the invention;
Figure 11 is an embodiment of the invention link S4-S6 fault notification process schematic diagram;
Figure 12 causes protection reversed process schematic diagram for embodiment of the invention link S4-S6 fault;
Figure 13 is announcement process schematic diagram during link S5-S6 fault in the embodiment of the invention;
Figure 14 causes the process schematic diagram that protection is switched for embodiment of the invention link S5-S6 fault;
Figure 15 is an embodiment of the invention link S3-S4 fault notification process schematic diagram;
Figure 16 causes the process schematic diagram that protection is switched for embodiment of the invention link S3-S4 fault;
Figure 17 is an embodiment of the invention link S3-S4 fault recovery announcement process schematic diagram;
Figure 18 is that embodiment of the invention link S3-S4 fault recovery starts the timer schematic diagram;
Figure 19 is the overtime transmission of the Wait-to-restore RR process schematic diagram of embodiment of the invention node S7;
Figure 20 embodiment of the invention node is received the processing procedure schematic diagram of RR (Ring2);
Figure 21 is an embodiment of the invention link S5-S6 fault recovery announcement schematic diagram;
Figure 22 is the overtime transmission of an embodiment of the invention node S1 Wait-to-restore RR schematic diagram;
Figure 23 receives the processing procedure schematic diagram of RR (Ring1) for embodiment of the invention node;
Figure 25 is the overtime transmission of an embodiment of the invention node S3Wait-to-restore RR process schematic diagram;
Figure 26 receives the processing procedure schematic diagram of RR (Ring0) for embodiment of the invention node.
Embodiment
Below in conjunction with accompanying drawing and embodiments of the invention method of the present invention is described in further detail.
The present invention is according to the looped network protocol theory; when the many rings of processing are topological; introduce basic ring; the ring of deriving; the basic ring link; notions such as cyclopolymerization port; phase cross ring topology is carried out logical partitioning; specify the different priority of ring; carry out the two-way detection of link with Hello mechanism; link state monitoring in conjunction with port is realized link fault detect fast and accurately; basic ring link or logic basic ring link are adopted fault detection mechanism based on the cyclopolymerization port status; in conjunction with existing looped network principle; the protection that triggers respective rings during link failure is switched; introduce to recover time-delay mechanism during link failure recovery, thus the protections layer by layer that realize many ring topologys switch and recover, given full play to the reliabilities of many ring topologys; and error protection was switched at 50ms in the time, satisfied the Carrier Ethernet requirement.Described term and being explained as follows:
Basic ring: specifying a ring in the phase cross ring topology is basic ring, identifies with Ring0;
Basic ring link: basic ring is considered as a redundant link between the crossed node, is called the basic ring link;
The ring of deriving: the ring that basic ring link and other links are formed is the ring of deriving.Basic ring can derive a plurality of rings with other links, respectively with Ring1, Ring2 ..., sign such as RingN;
Logic basic ring and logic basic ring link: basic ring can be regarded a logic basic ring as with the cross ring mutually that the ring of deriving is formed, and is same, and the logic basic ring is considered as a redundant link between the crossed node, is called logic basic ring link;
The pure loop chain road of deriving: the part in the ring of deriving except that basic ring link or logic basic ring link;
The cyclopolymerization port: two basic ring ports on the link point of will deriving (can be logic basic ring ports, may be a plurality of) be considered as a logic port, be called the cyclopolymerization port, with Rag node identification (the ring sign of deriving, (member port 1, member port 1 state), (member port 2, member port 2 states) ...) expression;
Ring priority: with ring sign (ID) representative ring priority, the priority of basic ring (Ring0) is the highest, secondly is the ring Ring1 that derives, and the like;
The ring protection port: the ring control port, obstruction prevents that loop from producing during the ring operating state, opens during the ring protection state to finish to protect to switch;
Ring Control Node: the node of main decision-making and control on the ring.On each ring a Control Node must be arranged, and can only have one, the ring protection port is positioned on the ring Control Node.
The ring transmission node: other node on the ring except that Control Node can be called transmission node.
Whole looped network can be begun to develop from basic ring by basic ring, the ring of deriving, logic basic ring, it is whole that the final ring of deriving is actually whole looped network.
Fig. 2 is an embodiment of the invention phase cross ring topology schematic diagram, and as shown in Figure 2, this figure is the phase cross ring topology that is made of 6 nodes, and the ring of specifying S1, S2, S3, S4 to form is a basic ring, identifies with Ring0; The redundant link that basic ring is regarded as between S3, the S4 is the basic ring link, is made up of link S3-S1-S2-S4 and link S3-S4, and then the connection between S3, the S4 is the link of a high reliability.
Fig. 3 is the embodiment of the invention basic ring link and the ring logical schematic of deriving, and as shown in Figure 3, it is exactly the ring of deriving with the ring that node S5, S6 form that S3, the S4 basic ring link by is in logic connected the back, identifies with Ring1.Port one on the node S3, the 2nd, the port of basic ring Ring0, being considered as a logic port on the ring Ring1 that derives is the cyclopolymerization port, with Rag S3 (Ring1, (1, Up), (2, Up)) sign, the port one, 2 on the same S4 is considered as cyclopolymerization port Rag S4 (Ring1 on Ring1, (1, Up), (2, Up)).Deriving in the basic ring link on the ring, not have a link of port blocked be primary link, and other links that have a port blocked are for from link, and the port that connects primary link in the cyclopolymerization port is main member port, and connecting from the port of link is from member port.
Fig. 4 is an embodiment of the invention primary link and from the link structure schematic diagram, and as shown in Figure 4, the solid black circle is represented port blocked; Basic ring Ring0, the ring Ring1 that derives, node S1, S5 are configured to the Control Node of Ring0, Ring 1 respectively; The port 2 of node S1, the port 2 of node S5 are configured to encircle the ring protection port of Ring0, Ring1; Basic ring link primary link S3-S4, from link S3-S1-S2-S4, the ring Ring1 node S3 cyclopolymerization port Rag1 master port 2 of deriving, from port one, node S4 cyclopolymerization port Rag1 master port 1, from port 2.
Based on above-mentioned notion and example principle, specifying ring S1-S2-S4-S3 among Fig. 1 is basic ring Ring0, and whole topology derives ring Ring1 (Ring0-S5-S6), ring Ring2 (Ring1-S7-S8), ring Ring3 (Ring2-S9-S10) and ring Ring4 (Ring3-S11-S12) successively.
Fig. 5 is the embodiment of the invention complex network basic ring and the topological schematic diagram of ring of deriving, as shown in Figure 5, the basic principle of ethernet ring network is by a link on the looped network control protocol choke ring when Ethernet ring is complete, the ring of physics is cracked into the chain of logic, prevent data loopback, and when a link disconnects on the Ethernet ring, decontrol rapidly and block link to recover the communication path between each node on the looped network.
Fig. 6 is an embodiment of the invention monocycle normal condition schematic diagram, and as shown in Figure 6, Ring0 is made up of node S1, S2, S3, S4, configuration S1 is a Control Node, when all link normal conditions, Control Node S1 blocks its upper port 2, and whole ring is in normal condition.
Fig. 7 is an embodiment of the invention monocycle link failure schematic diagram, as shown in Figure 7, supposes link S3-S4 fault, and node S3, S4 block non-working port and the announcement fault detecting fault respectively.
Fig. 8 is a single link failure protection reversed process schematic diagram in the embodiment of the invention; as shown in Figure 8; Control Node S1 receives fault notification; decontrol the protection port; refresh and transmit; and send to refresh and transmit message Flush, transmission node is received that Flush refreshes and is transmitted and transmit, the communication path recovery on the ring between each node.
The present invention has introduced the notion of basic ring link and logic basic ring link, encircle message is received and need judge by the high priority Control Node at the protocol massages of its basic ring or logic basic ring intermediate ring road regulation low priority ring for preventing to derive, if receive and directly to abandon from port blocked, and can not transmit from port blocked, the message that aggregation port place node is received from the cyclopolymerization port can not be transmitted from the member port of cyclopolymerization port.When a plurality of rings intersect, trigger to transmit to refresh when preventing that error protection from switching and cause unnecessary concussion, regulation Flush message can not be transmitted on the ring of normal condition.
Fig. 9 is that the many rings of embodiment of the invention complexity intersect the network topology schematic diagram, as shown in Figure 9, have basic ring Ring0 (S3-S4-S6-S5) among the figure, derive ring Ring1 (Ring 0-S1-S2) and cyclopolymerization port RagS3 (Ring1, (2, Up), (3, Up)), cyclopolymerization port RagS4 (Ring1, (1, Up), (3, Up)), derive ring Ring 2 (Ring 1-S7-S8) and cyclopolymerization port RagS5 (Ring2, (1, Up), (2, Up)) and cyclopolymerization port RagS6 (Ring2, (1, Up), (2, Up)).
According to the ethernet ring network basic principle; by configuration basic ring Ring0 Control Node S3 and go up ring protection port 2, derive ring Ring1 Control Node S1 and go up ring protection port 2, ring Ring2 Control Node S7 and go up ring protection port 2 derives; the network shown in this figure then; topology when operate as normal is that a complexity is encircled operating state topology schematic diagram more shown in Fig. 9 '.
Be example with complicated many ring operating state topology schematic diagrames shown in Fig. 9 ' below, link failure detection mechanism of the present invention, wrap protection and Restoration Mechanism thereof are illustrated one by one.
Figure 10 is a link process fault detection schematic diagram in the embodiment of the invention, as shown in figure 10, carry out the bi-directional failure detection by sending out the Hello frame periodically mutually between adjacent node,, then represent link occurs fault if in a period of time, do not receive the Hello frame of opposite end.Detect when initial, node is not known the opposite end sign, as MAC, when sending the Hello frame by port, the opposite end is designated sky, when node is received the Hello frame of peer node, promptly begin session, and carry the Hello frame that the opposite end sign sends to the opposite end again, described peer node is received when having the Hello frame that oneself identifies, the expression link is normal, and session this moment is promptly set up.When node is not received the Hello frame of peer node in a period of time, then think link failure, promptly remove session this moment, in the Hello frame that the opposite end sends, the opposite end sign is put sky and announced fault, send a short sequence fault notification message DN (Defect-notify) then.Here, be that example describes with the link S1-S2 among Fig. 9 ', as can be seen from Figure 10, node S1, S2 send out Hello mutually and set up session.When link S1-S2 broke down, node S1, node S2 all can detect link failure and announce fault mutually.
In addition, except using Hello frame detection node fault, node also can be monitored the Link State of port, in case find fault, then adopts the mode of interrupting to notify CPU, triggers fault notification and also removes session.Be used in combination by this dual mode, can guarantee that link failure detection can carry out rapidly and accurately.
When faulty link recovered, node has only to set up with the other side just thought link-recovery after the session, and sends the fault recovery notification packet RN (Restore-notify) of a short sequence.
Below Ethernet of the present invention being intersected the ring protection reversed process is elaborated:
According to aforementioned ethernet ring network principle; when link breaks down in the ring; node detection is blocked the faulty link port when the link failure in the ring, and the announcement fault gives the ring Control Node, and ring Control Node open loop protection port and triggering is transmitted to upgrade and finished ring protection and switch.No matter the basic ring ring of still deriving is all deferred to this principle, and all links of basic ring and the pure loop chain road of deriving all are single-links, can be directly when breaking down by node detection to and announcement switch for corresponding ring Control Node triggering ring protection.And basic ring link or logic basic ring link in the ring of deriving are redundant links, thereby only with judging that this link failure trigger protection switches when the equal fault of its Member Link.Here, the described pure loop chain road of deriving is the part except that basic ring link or logic basic ring link in the ring of deriving.
When detecting basic ring link or logic basic ring link failure, make following regulation:
The first, node detection is to fault, send fault notification message DN, DN must carry the ring sign, DN only transmits on corresponding ring, up to the ring Control Node, when the cyclopolymerization port place node of deriving is received DN, will check whether there is this ring sign of deriving, if there is no, then in DN, add the ring sign of deriving, and put the member port Down of the cyclopolymerization port of receiving DN; If exist, then only do to transmit to handle;
The second, when the member port of the cyclopolymerization port of deriving links link failure, put the member port Down of the cyclopolymerization port of the link that breaks down, and send fault notification message DN, in DN, add basic ring or logic basic ring sign and the ring of deriving and identify;
Three, when all member ports of the cyclopolymerization port of deriving all are changed to Down, basic ring link or logic basic ring link failure, announcement DN switch for the ring Control Node trigger protection of deriving.
Be that example describes with Fig. 9 ' still, this moment, whole ethernet ring network was in normal operating conditions, and Figure 11 is an embodiment of the invention link S4-S6 fault notification process schematic diagram, as shown in figure 11, as node S4, during link S4-S6 fault between the S6, as the transmission node S4 of basic ring Ring0, S6 will detect fault, node S4 port blocked 3, put cyclopolymerization port RagS4 (Ring1, (1, Up), (3, Down)), do not trigger the ring Ring1 announcement fault of deriving, only on basic ring Ring0, announce fault, add basic ring sign Ring0 and derive ring sign Ring1, send DN (Ring0, Ring1), node S6 port blocked 2 is put cyclopolymerization port RagS6 (Ring2, (1, Up), (2, Down)), send DN (Ring0, Ring2) announcement fault.
Node S5 receives that (Ring0 Ring2), checks and finds to derive ring sign Ring2 fault notification DN, only does to transmit and handles.Node S3 receive fault notification DN (Ring0, Ring1), the ring sign Ring1 that check to find to derive, and inoperation aggregation port state, receive DN (Ring0, Ring2), put aggregation port state RagS3 (Ring1, (and 2, Up), (3, Down)).
Figure 12 as shown in figure 12, receives fault notification open port ring protection port 2 as the Control Node of Ring0 for embodiment of the invention link S4-S6 fault causes protection reversed process schematic diagram; send Flush; refresh and transmit, forward guard mode to, identify with * Ring0.Because Ring1 and Ring2 all are in normal condition, so Flush only transmits on Ring0, transmission node S4, S5, S6 receive Flush, refresh and transmit, and protection is switched and finished.
Figure 13 is announcement process schematic diagram during link S5-S6 fault in the embodiment of the invention, as shown in figure 13, and when link S5-S6 breaks down, node S5, S6 will detect fault, node S5 port blocked 2, announcement fault DN (Ring0), put cyclopolymerization port RagS5 (Ring2, (1, Up), (2, Down)), because cyclopolymerization port member port 1 still is in the Up state, do not trigger ring Ring 2 fault notification, node S6 port blocked 1, put cyclopolymerization port RagS6 (Ring2, (1, Down), (2, Down)), because member port all is in the Down state, the basic ring link failure of the ring Ring 2 that derives sends fault notification DN (Ring2).
Node S3 receives DN (Ring0); owing to be in guard mode, can not continue protection and switch, put aggregation port state RagS3 (Ring1; (2; Up), (3, Down)); because port 2 is Up still; do not trigger the fault notification of ring on the Ring1, node S8 receives that fault notification transmits, and Control Node S7 receives that the fault notification trigger protection switches.
Figure 14 causes the process schematic diagram switched of protection for embodiment of the invention link S5-S6 fault, and as shown in figure 14, Control Node S7 receives that fault notification protection switches; open port 2 sends Flush, refreshes and transmits; forward guard mode to, identify with * Ring2.Ring1 still is in normal condition, and the Flush message can not be transmitted on Ring1, and node S3, S4, S5, S6 and S8 receive to refresh behind the Flush and transmit, and finish protection and switch.
Figure 15 is an embodiment of the invention link S3-S4 fault notification process schematic diagram, as shown in figure 15, when link S3-S4 fault, node S3, S4 will detect fault, node S3 port blocked 2, put cyclopolymerization port RagS3 (Ring1, (2, Down), (3, Down)), Control Node as Ring0 need not fault notification, and Down falls because cyclopolymerization port member holds, triggers the fault notification on the ring Ring1, send DN (Ring1), node S4 port blocked 1 is put cyclopolymerization port RagS4 (Ring1, (1, Down), (3, Down)) are because member port all is in the Down state, derive and encircle the basic ring link failure of Ring1, send fault notification DN (Ring1).Node S2 receives that fault notification transmits, and Control Node S1 receives that the fault notification trigger protection switches.
Figure 16 causes the process schematic diagram switched of protection for embodiment of the invention link S3-S4 fault, and as shown in figure 16, Control Node S1 receives that fault notification protection switches; open port 2 sends Flush, refreshes and transmits; forward guard mode to, identify with * Ring1.The Flush message is transmitted on whole topology, and the node of receiving Flush refreshes to be transmitted, and finishes protection and switches.So far, the protection of this ring is switched and is finished, if other link failures are arranged again, will cause that isolated island node/chain (can not be called isolated island node/chain with the node/chain of other node communications) occurs, and the communication between part of nodes will be interrupted.
Certainly, link failure also can other order occur, but all can analyze according to top principle, does not repeat them here.
Recovery process to Ethernet phase cross ring Restoration Mechanism of the present invention describes below:
When phase cross ring topology faulty link recovers, the link point that detects link failure recovery sends RN announcement fault recovery, RN must carry the ring sign, RN only transmits on corresponding ring, up to the ring Control Node, when receiving RN, the cyclopolymerization port place node of deriving to check whether there is this ring sign of deriving, if there is no will in DN, add sign, and put the member port Up of the cyclopolymerization port of receiving RN, and on the corresponding ring of deriving announcement fault recovery RN message; If exist, then only do to transmit to handle.The ring Control Node is received RN or is detected link-recovery, then start the recovery back off timer Wait-to-restore of respective rings, this timer time is Wait-to-restore-time/ (2n+1) s, wherein: Wait-to-restore-time is that topology is recovered delay time, can be configured according to the real network environment; N is the respective rings numbering, as basic ring 0; Wait-to-restore is overtime when back off timer; then begin to recover to switch; choke ring protection port; refresh and transmit; send fault recovery response message RR (Restore-response), RR must carry and send the ring sign, and regulation RR can only be higher than own and be on the ring of guard mode and transmits in respective rings and priority; high priority ring Control Node receives that RR will suppress the back off timer Wait-to-restore of self, and the node of receiving the RR message refreshes to be transmitted.
Above-mentioned timer mechanism, the recovery delay time of high priority ring is longer than the low priority ring, and when making basic ring link or logic basic ring link-recovery, the ring of deriving of low priority will recover to switch prior to the logic basic ring or the basic ring of high priority.
Below, the process of recovering successively with 3 links among Figure 16 is that example describes recovery reversed process of the present invention.
It is 30s that the Wait-to-restore-time time is set, and then the Wait-to-restore timer time of Ring0, Ring1, Ring2 is respectively 30s, 10s, 6s.
Figure 17 is an embodiment of the invention link S3-S4 fault recovery announcement process schematic diagram, as shown in figure 17, when link S4-S6 among the figure recovers, node S4, S6 detects link-recovery, and S4 is the transmission node of basic ring Ring0, send fault recovery announcement RN (Ring0), S4 derives to encircle the aggregation port place node of Ring 1 simultaneously, puts aggregation port state RagS4 (Ring1, (1, Down), (3, Up)) send fault recovery announcement RN (Ring1), similarly, S6 is the transmission node of basic ring Ring0, sends fault recovery announcement RN (Ring0), and S6 derives to encircle the aggregation port place node of Ring2 simultaneously, put aggregation port state RagS6 (Ring2, (1, Down), (2, Up)), send fault recovery announcement RN (Ring0).
Figure 18 is that embodiment of the invention link S3-S4 fault recovery starts the timer schematic diagram, the Control Node S1 that encircles Ring1 that derives receives that RN (Ring1) starts Wait-to-restore (10s), the Control Node S7 that encircles Ring2 that derives receives that RN (Ring2) starts Wait-to-restore (6s), and the Control Node S3 of basic ring Ring0 receives that RN (Ring0) starts Wait-to-restore (30s).
Figure 19 is the overtime transmission of the Wait-to-restore RR process schematic diagram of embodiment of the invention node S7; the Wait-to-restore of ring Ring 2 Control Node S7 of deriving is at first overtime; choke ring protection port 2 refreshes and transmits, and the announcement ring recovers response RR (Ring2).
Figure 20 embodiment of the invention node is received the processing procedure schematic diagram of RR (Ring2), RR transmits on topology, receive to refresh behind the node of RR and transmit, ring Control Node S3, S1 find that the ring priority that RR carries is lower than oneself after receiving RR, to suppress the Wait-to-restore of oneself respectively and transmit, S4, S6 receive that RR opens port blocked.
Figure 21 is an embodiment of the invention link S5-S6 fault recovery announcement schematic diagram, as shown in figure 21, and when link S5-S6 recovers, node S5, S6 detects link-recovery, S5 is the transmission node of basic ring Ring0, sends fault recovery announcement RN (Ring0), and S5 derives to encircle the aggregation port place node of Ring2 simultaneously, then put aggregation port state RagS5 (Ring2, (1, Up), (3, Up)), because Ring2 is in normal condition, do not send fault recovery announcement RN, same, S6 is the transmission node of basic ring Ring0, send fault recovery announcement RN (Ring0), S6 derives to encircle the aggregation port place node of Ring2 simultaneously, puts aggregation port state RagS6 (Ring2, (1, Up), (2, Up)), because Ring2 is in normal condition, do not send fault recovery announcement RN.S3 receives that RN (Ring0) starts Wait-to-restore (30s), as the cyclopolymerization port place node of the ring Ring 1 that derives, puts aggregation port state RagS3 (Ring1, (2, Down), (3, Up)), send RN (Ring1), S4 receives RN (Ring0), as the cyclopolymerization port place node of the ring Ring1 that derives, put aggregation port state RagS4 (Ring1, (1, Down), (3, Up)), send RN (Ring1), S1 receives RN (Ring1), starts Wait-to-restore (10s).
Figure 22 is the overtime transmission of an embodiment of the invention node S1Wait-to-restore RR schematic diagram, and the Wait-to-restore of the ring Ring1 Control Node of deriving S1 is overtime, and choke ring protection port 2 refreshes and transmits, and the announcement ring recovers response RR (Ring1).
Figure 23 receives the processing procedure schematic diagram of RR (Ring1) for embodiment of the invention node, as shown in figure 23, RR transmits on topology, the node of receiving RR refreshes to be transmitted, ring Control Node S3 finds that the ring priority that RR carries is lower than oneself after receiving RR, will suppress the Wait-to-restore of oneself and transmit, and S5, S6 receive that RR opens port blocked, S5, S6 place derive the ring Ring2 be in normal condition, do not transmit RR.
Figure 24 is an embodiment of the invention link S3-S4 fault recovery announcement schematic diagram, as shown in figure 24, when link S3-S4 recovers, node S3, S4 detect link-recovery, S4 is the transmission node of basic ring Ring 0, send fault recovery announcement RN (Ring0), S4 derives to encircle the aggregation port place node of Ring1 simultaneously, puts aggregation port state RagS4 (Ring1, (1, Up), (3, Up)), because Ring1 is in normal condition, do not send fault recovery announcement RN, S3 is the Control Node of basic ring Ring0, starts Wait-to-restore (30s); Simultaneously S3 is again the aggregation port place node of ring Ring1 of deriving, put aggregation port state RagS3 (Ring1, (and 2, Up), (3, Up)); Because Ring1 is in normal condition, do not send fault recovery announcement RN.S6 receive RN (Ring0) as derive the ring Ring2 cyclopolymerization port place node, put aggregation port state RagS6 (Ring1, (and 1, Up), (2, Up)); Because Ring2 is in normal condition, does not send RN, S5 receives RN (Ring0), as the cyclopolymerization port place node of the ring Ring2 that derives, put aggregation port state RagS5 (Ring2, (and 1, Up), (2, Up)), because Ring2 is in normal condition, do not send RN.
Figure 25 is the overtime transmission of an embodiment of the invention node S3Wait-to-restore RR process schematic diagram, and the Wait-to-restore of basic ring Ring0 Control Node S3 is overtime, and choke ring protection port 2 refreshes and transmits, and the announcement ring recovers response RR (Ring0).
Figure 26 receives the processing procedure schematic diagram of RR (Ring0) for embodiment of the invention node, because derive ring Ring1, Ring2 all are in normal condition, so RR only transmits on Ring0, the node of receiving RR refreshes to be transmitted, and S4 receives that RR opens port blocked.
So far, all link failures all recover, and ring Ring0, ring Ring1, ring Ring2 also rework in normal condition.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.
Claims (8)
1, a kind of Ethernet intersects the ring protection reverse method, it is characterized in that this method comprises:
A, Ethernet phase cross ring topology is carried out logical partitioning, specify different priority, and between adjacent node, periodically mutually send out a Hello frame and set up session and carry out link bidirectional and detect, when detecting link occurs fault, the announcement fault, and send short sequence fault notification message DN;
B, block link during to fault when the node detection in the Ethernet phase cross ring, the announcement fault is to the ring Control Node of place ring, described ring Control Node open loop protection port, and trigger to transmit to upgrade and carry out ring protection and switch.
2, crossing ring protection reverse method according to claim 1; it is characterized in that; described steps A further comprises: or with node the Link State of port is monitored, when finding that link failure then adopts interrupt mode notice CPU, triggers fault and also removes session.
3, crossing ring protection reverse method according to claim 1 is characterized in that, Ethernet phase cross ring topology is carried out logical partitioning specifically comprise:
To specify a ring in the phase cross ring topology to be divided into basic ring;
The ring of basic ring link and other links composition is divided into the ring of deriving; Wherein, regard the part of basic ring between crossed node as a redundant link and be called the basic ring link;
The cross ring mutually that the described basic ring and the ring of deriving are formed is divided into the logic basic ring.
4, according to claim 1 or 3 described crossing ring protection reverse methods, it is characterized in that link failure detection specifically comprises:
When detecting link failure, send fault notification message DN, DN must carry the ring sign, DN only transmits on corresponding ring, up to the ring Control Node, when the cyclopolymerization port place node of deriving is received DN, check earlier and whether have this ring sign of deriving, if there is no, then in DN, add the ring sign of deriving, and put the member port Down of the cyclopolymerization port of receiving DN; If exist, then only do to transmit to handle.
5, crossing ring protection reverse method according to claim 4 is characterized in that the process of described announcement fault comprises:
When the member port of the cyclopolymerization port of deriving links link failure, put the member port Down of the cyclopolymerization port of the link that breaks down, and send fault notification message DN, in DN, add basic ring or logic basic ring sign and the ring of deriving and identify;
And when all member ports of the cyclopolymerization port of deriving all were changed to Down, basic ring link or logic basic ring link failure, announcement DN switched for the ring Control Node trigger protection of deriving.
6, a kind of Ethernet phase cross ring based on the described protection reversed process of claim 1 is recovered reverse method, it is characterized in that this method comprises:
In a, the phase cross ring topology faulty link recovery process, the link point of link failure recovery sends short sequence fault recovery announcement RN message, and the RN message is transmitted in described ring, up to the ring Control Node; When receiving the RN message, the cyclopolymerization port place node of deriving checks wherein whether there is the ring sign of deriving earlier, if do not exist, then in DN, add the ring sign of deriving, put the member port Up of the cyclopolymerization port of the described RN of receiving message, and on the corresponding ring of deriving, announce fault recovery RN message; If exist, then only do to transmit to handle;
B, ring Control Node are received the RN message or detected link-recovery, then start the recovery back off timer of respective rings, and timer recovery delay time is set, and be overtime when described back off timer, then begins to recover to switch.
7, phase cross ring according to claim 6 is recovered reverse method, it is characterized in that described timer time is Wait-to-restore-time/ (2n+1) s; Wherein: Wait-to-restore-time is that topology is recovered delay time, need be configured according to the real network environment; N is the respective rings numbering.
8, phase cross ring according to claim 6 is recovered reverse method, it is characterized in that described recovery reversed process further comprises:
It is overtime that the ring Control Node is recovered back off timer; choke ring is protected port, is refreshed and transmit; and transmission fault recovery response message RR; and described RR carries and sends the ring sign; be higher than self and be on the ring of guard mode in respective rings and priority and transmit; high priority ring Control Node receives that RR will suppress the back off timer of self, and the node of receiving the RR message refreshes to be transmitted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910171823A CN101645812A (en) | 2009-08-31 | 2009-08-31 | Protection switching method of Ethernet intersecting ring |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910171823A CN101645812A (en) | 2009-08-31 | 2009-08-31 | Protection switching method of Ethernet intersecting ring |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101645812A true CN101645812A (en) | 2010-02-10 |
Family
ID=41657537
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200910171823A Pending CN101645812A (en) | 2009-08-31 | 2009-08-31 | Protection switching method of Ethernet intersecting ring |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101645812A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101986615A (en) * | 2010-11-09 | 2011-03-16 | 中兴通讯股份有限公司 | Multiple-ring Ethernet and protection method thereof |
CN102143410A (en) * | 2010-07-09 | 2011-08-03 | 华为技术有限公司 | Path computing method and unit in optical network |
WO2011127736A1 (en) * | 2010-04-12 | 2011-10-20 | 中兴通讯股份有限公司 | Method and apparatus for maintaining saved information of node in ethernet ring network |
CN102546343A (en) * | 2010-12-21 | 2012-07-04 | 中兴通讯股份有限公司 | Protection switching method and device for intersected rings |
CN102664785A (en) * | 2012-04-23 | 2012-09-12 | 中兴通讯股份有限公司 | Method and system for protecting intersecting ring |
CN102724096A (en) * | 2012-06-26 | 2012-10-10 | 华为技术有限公司 | Mixed ring network protection method and network equipment |
CN105324959A (en) * | 2013-06-12 | 2016-02-10 | 三菱电机株式会社 | Communication system, communication device, and protection method |
CN111585860A (en) * | 2020-04-28 | 2020-08-25 | 杭州迪普科技股份有限公司 | Method, device, storage medium and equipment for loop network convergence |
CN111726291A (en) * | 2019-08-30 | 2020-09-29 | 新华三技术有限公司 | Path switching method and network equipment |
CN117938714A (en) * | 2024-02-01 | 2024-04-26 | 山东师范大学 | Network link failure detection method, device, equipment and readable storage medium |
-
2009
- 2009-08-31 CN CN200910171823A patent/CN101645812A/en active Pending
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011127736A1 (en) * | 2010-04-12 | 2011-10-20 | 中兴通讯股份有限公司 | Method and apparatus for maintaining saved information of node in ethernet ring network |
CN102143410B (en) * | 2010-07-09 | 2013-09-11 | 华为技术有限公司 | Path computing method and unit in optical network |
CN102143410A (en) * | 2010-07-09 | 2011-08-03 | 华为技术有限公司 | Path computing method and unit in optical network |
CN101986615A (en) * | 2010-11-09 | 2011-03-16 | 中兴通讯股份有限公司 | Multiple-ring Ethernet and protection method thereof |
CN102546343A (en) * | 2010-12-21 | 2012-07-04 | 中兴通讯股份有限公司 | Protection switching method and device for intersected rings |
CN102546343B (en) * | 2010-12-21 | 2014-12-10 | 中兴通讯股份有限公司 | Protection switching method and device for intersected rings |
CN102664785A (en) * | 2012-04-23 | 2012-09-12 | 中兴通讯股份有限公司 | Method and system for protecting intersecting ring |
CN102724096A (en) * | 2012-06-26 | 2012-10-10 | 华为技术有限公司 | Mixed ring network protection method and network equipment |
CN105324959A (en) * | 2013-06-12 | 2016-02-10 | 三菱电机株式会社 | Communication system, communication device, and protection method |
CN105324959B (en) * | 2013-06-12 | 2018-10-23 | 三菱电机株式会社 | Communication system, communication device and guard method |
CN111726291A (en) * | 2019-08-30 | 2020-09-29 | 新华三技术有限公司 | Path switching method and network equipment |
CN111585860A (en) * | 2020-04-28 | 2020-08-25 | 杭州迪普科技股份有限公司 | Method, device, storage medium and equipment for loop network convergence |
CN111585860B (en) * | 2020-04-28 | 2021-12-24 | 杭州迪普科技股份有限公司 | Method, device, storage medium and equipment for loop network convergence |
CN117938714A (en) * | 2024-02-01 | 2024-04-26 | 山东师范大学 | Network link failure detection method, device, equipment and readable storage medium |
CN117938714B (en) * | 2024-02-01 | 2024-11-29 | 山东师范大学 | Network link failure detection method, device, equipment and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101645812A (en) | Protection switching method of Ethernet intersecting ring | |
CN101841450B (en) | Method and system for realizing stable communication by forming multiple ring topologies into intersecting ring | |
CN101741673B (en) | Address refreshing method of Ethernet ring | |
CN101652963B (en) | Method for reconfiguring a communications network | |
CN100454880C (en) | A method and system for realizing ring network protection | |
CN101399737A (en) | Protection method and apparatus used for link convergence group of loop Ethernet | |
CN101252500A (en) | Intersect looped network, node and realizing method of random topology intersect looped network | |
CN102377605B (en) | Method for implementing double-returning protection of packet transport network (PTN) equipment by using combination of multiple protection modes | |
TWI455525B (en) | Recovery method for ring-based network | |
CN101127653A (en) | Ethernet loop device with backup and implementation method for master device backup | |
CN101714939A (en) | Fault treatment method for Ethernet ring network host node and corresponding Ethernet ring network | |
CN101227371A (en) | Method and apparatus of backup switch among same level switch equipments | |
CN101640622A (en) | Ethernet ring net protection method and transit node | |
CN101626335B (en) | Data protection method of dual-connection network | |
CN101674240A (en) | Method and device for realizing single-node dual home for distributed system | |
CN101686158A (en) | Control method for Ethernet loop guard fault recovery and Ethernet loop node | |
CN101127673A (en) | Ethernet Automatic Protection Switching Method | |
CN102158384A (en) | Novel MRing Ethernet ring network protection technology | |
CN101873244A (en) | Automatic multi-loop protection method | |
CN110635940B (en) | Main/standby switching method of EAPS Ethernet ring network | |
CN101686167A (en) | Method for controlling recovery of faults of Ethernet | |
CN101989930B (en) | Method and switching equipment for realizing Ethernet double loop | |
CN101222402B (en) | Ethernet ring protection method, system and device | |
CN104135417A (en) | Quick recovery method for ethernet ring network link interruption and corresponding ethernet ring network | |
CN101848128B (en) | Method, system and topological structure for realizing stabilized communication among multi-ring topologies |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100210 |