CN101072154B - Loop Ethernet switching method - Google Patents
Loop Ethernet switching method Download PDFInfo
- Publication number
- CN101072154B CN101072154B CN2007101115534A CN200710111553A CN101072154B CN 101072154 B CN101072154 B CN 101072154B CN 2007101115534 A CN2007101115534 A CN 2007101115534A CN 200710111553 A CN200710111553 A CN 200710111553A CN 101072154 B CN101072154 B CN 101072154B
- Authority
- CN
- China
- Prior art keywords
- link
- time
- host node
- port
- state
- 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.)
- Active
Links
Images
Landscapes
- Small-Scale Networks (AREA)
Abstract
The invention discloses a loop Ethernet switching method. The method includes following steps: step S202, configuring delayed processing time in use for link recovery on main node of loop Ethernet; the delayed processing time is valid for main node only; step S204, when failure occurs on link, main node and transmission node registers state of link as state of failure, and corresponding process carries out; step S206, when link is recovered, main node processes link switching in delayed time based on delayed processing time. The method prevents multiple times of switching in EAPS short time cause by quick oscillation of link, and unsuccessful operation of EAPS on hardware, and descendent network performance. Moreover, the invention raises capability of resisting fault of Ethernet ring under complex link oscillation.
Description
Technical field
The present invention relates to the communications field, relate in particular to the ethernet ring network changing method.
Background technology
Along with the development of IP network towards the multiple service supporting direction; NGN (Next generationnetwork; next-generation communication network), IPTV (Internet protocol television; Web TV) etc. professional reliability for network, real-time require more and more higher; and traditional looped network protection STP of Access Network double layer network (Spanning-tree protocol, Spanning-Tree Protocol) technology can not satisfy the requirement of quick convergence, link switchover.RFC3619 has defined a kind of Ethernet automatic protection switching (EAPS; Ethernet Automatic ProtectionSwitching) method; this method has solved ethernet device slow problem of network failure convergence when the loop network topology, utilizes this method can make convergence time in 50ms.Ethernet automatic protection switching ring is connected to form by a plurality of nodes; on the ring a plurality of Ethernet automatic protection switchings territory (EAPS Domain) can be arranged; a territory can define host node and the transmission node of oneself; can define professional VLAN (VLAN, Virtual LocalArea Network) that one group of business that is used for user data that belongs to certain Ethernet automatic protection switching territory transmits and the control VLAN that is used for the forwarding of Ethernet automatic protection switching protocol massages.For a specific virtual LAN, host node blocks the professional VLAN forwarding capability of time port when link is intact; Decontrol the professional VLAN forwarding capability of time port when link occurs fault, the notice ring is gone up the node updates mac address table; The professional VLAN forwarding capability that when link-recovery, blocks time port again, and notify ring to go up the node updates mac address table, thus guaranteed that professional VLAN can not form loop, prevent because " broadcast storm " that loop causes.
The Ethernet automatic protection switching is in actual networking medium chain road meeting more complicated; the situation that has quick oscillation in the short time; Fig. 1 is the topological structure of EAPS networking; wherein; telephone net node S1, S2, S3, S4 have formed ethernet ring network, and wherein S2 is main (master) node, and all the other nodes are transmission (transit) node; the port one of S2 is a master port, and port 2 is time ports.When link was intact, the inferior port 2 of S2 got clogged; When link occurs fault, the inferior port 2 of S2 is decontroled, and notifies other node updates mac address tables; When link-recovery, the inferior port 2 of S2 gets clogged again, notifies other node updates mac address tables.But in actual networking, link is quick oscillation at short notice, and for example the link between S3 and the S4 takes place repeatedly logical and disconnected at short notice.According to Ethernet automatic protection switching agreement; Ethernet automatic protection switching territory is repeatedly handle link switching at short notice; the inferior port of host node also can repeatedly get clogged and decontrol at short notice, and node repeatedly upgrades mac address table at short notice on the whole ring.Because the disposal ability of hardware chip is limited, repeatedly hardware is operated at short notice, the situation of operation failure might occur, thereby it is obstructed to cause that the Ethernet automatic protection switching is handled wrong loop or the business datum of causing.In addition, repeatedly upgrade mac address table at short notice, business datum is all passed through broadcast transmitted, greatly reduces the efficient of system.
Summary of the invention
At the problems referred to above, the present invention proposes a kind of ethernet ring network changing method, this method can avoid the EAPS that causes owing to quick oscillation of link repeatedly to switch in the short time, and EAPS is to degradation problem under the unsuccessful and network performance of hardware operation, and can improve the fault-resistant ability of ethernet ring network under complicated link oscillation situation.
For achieving the above object, Ethernet changing method of the present invention may further comprise the steps: S202, and configuration is used for the delay process time of link-recovery on the host node in ethernet ring network, and it is only effective to host node that this states the delay process time; S204, under the situation of link occurs fault, host node and transmission node are the link failure state with link state records and carry out respective handling; And S206, under the situation of link-recovery, host node switches according to delay process time delays handle link.
Wherein, configuration is delayed time forwarding time on the transmission node in host node and ethernet ring network, and this forwarding time of delaying time is greater than the delay process time.
Wherein, S204 comprises: S2042, host node detect under the situation of link occurs fault, and the record chain line state is the link failure state, decontrol time port, send the link failure frame on ethernet ring network, and upgrade the mac address table of port; And S2044, transmission node receives under the situation of link failure frame, and the record chain line state is the link failure state, and upgrades the mac address table of port.
Wherein, S206 comprises S2062: be that host node is the link serviceable condition with link state records under 0 the situation in the delay process time, block time port, send the intact notification frame of link on ethernet ring network, and upgrade the mac address table of port.
Wherein, S206 comprises: S2064, the delay process time greater than 0 situation under, host node record chain line state is the link-recovery state, and will be recorded as the link-recovery time current time; And S2066, detecting at host node under the situation of link occurs fault, host node is the link failure state with link state records; Detect the current time at host node and deduct under the situation of link-recovery time more than or equal to the delay recovery time, host node and transmission node begin to carry out link switchover.
Wherein, the process that host node and described transmission node begin to carry out link switchover comprises: step 1, and host node is the link serviceable condition with link state records, blocks time port, on ethernet ring network, send the intact notification frame of link, and upgrade the mac address table of port; And step 2, transmission node is the link serviceable condition with link state records after receiving the intact notification frame of link of host node transmission, decontrols the port that blocks, and upgrades the mac address table of port.
The invention provides the method that a kind of delay process ethernet ring network switches, this method has been introduced notions such as returning to form of link, host node delay process link switchover when link-recovery, thereby avoid the EAPS that causes owing to quick oscillation of link repeatedly to switch in the short time, and EAPS is to degradation problem under the unsuccessful and network performance of hardware operation, and can improve the fault-resistant ability of ethernet ring network under complicated link oscillation situation, ensure professional reliable, stable operation.
Description of drawings
Shuo Ming accompanying drawing is used to provide further understanding of the present invention herein, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute to improper qualification of the present invention, in the accompanying drawings:
Fig. 1 is the topological structure of EAPS networking;
Fig. 2 is an Ethernet changing method flow chart of realizing the embodiment of the invention; And
Fig. 3 is that the host node delay process is recovered the flow chart that switches in the Ethernet changing method of the embodiment of the invention.
Embodiment
Below in conjunction with accompanying drawing technical scheme of the present invention is described in further detail.
In the present invention, description is divided into three kinds to Link State: serviceable condition (Link-Up-State), malfunction (Link-Down-State), return to form (Link-Preup-State), wherein Link-Preup-State is only effective for host node.When link was intact, the Link State of host node in the EAPS territory and transmission node record was Link-Up-State; When link occurs fault, the Link State of host node and transmission node record is Link-Down-State; When chain route fault recovery when being intact, can't illustrate that link is stable, host node is with the delay process link switchover, the Link State of host node record is Link-Preup-State, transmission node is not received the intact notification frame of Link State that host node sends, so the Link State of record remains Link-Down-State.
Fig. 1 is the topological structure of EAPS networking.As shown in Figure 1, node is in same EAPS territory on the ring, wherein S2 is a host node, S1, S3, S4 are transmission nodes, the port one of S2 is a master port, port 2 is time ports, and the pre-up-time time is set on the host node, and the pre-forward-time of each node setting is greater than the pre-up-time that is provided with on the host node.When the link occurs fault between S3 and the S4, node S3 and S4 send the chain circuit alarming frame to host node after detecting the link transmission fault of corresponding port, and the port block that link failure takes place.After host node S2 detected link failure, the record chain line state was Link-Down-State, decontroled time port 2, sent link failure frame (flush-down) from port one and 2, upgraded the mac address table of port then.After transmission node was received the link failure frame (flush-down) that host node sends, the record chain line state was Link-Down-State, upgraded the mac address table of port.
Below the idiographic flow of realizing method of the present invention is described in detail.With reference to Fig. 2, Ethernet changing method of the present invention may further comprise the steps:
A, the delay process time pre-up-time that configuration link recovers on host node, if the delay process time is 0, delay process not during link-recovery then; The Preforward state time pre-forward-time of configured port on host node and transmission node requires the delay process time pre-up-time (S202) of the pre-forward-time of each node greater than host node.
B, when link occurs fault, after host node detected fault, the record chain line state was Link-Down-State, decontroled time port, went up to ring and sent the link failure frame, upgraded the mac address table (S2042) of port then.
After c, transmission node were received the link failure frame that sends among the b, the record chain line state was Link-Down-State, upgraded the mac address table (S2044) of port.
D, when link-recovery, after host node detects link-recovery, configuring condition according to pre-up-time, two kinds of processing modes are arranged: when pre-up-time is 0, host node record chain line state is Link-Up-State, block time port, go up to ring and send the intact frame of link, upgrade the mac address table (S2062) of port then.When pre-up-time greater than 0 the time, processing scheme is as follows: d1, host node record chain line state are Link-Preup-State, the record current time is link-recovery time (S2064).D2, when the Link State of host node record is Link-Preup-State, if host node detects link occurs fault, then the host node recording status is Link-Down-State; If host node detects the current time and deducts the link-recovery time and be greater than or equal to pre-up-time, then host node begins handle link and switches (S2066).The record chain line state is Link-Up-State, blocks time port, goes up to ring and sends the intact frame of link, upgrades the mac address table of port then.
E, host node begin handle link and switch, and the record chain line state is Link-Up-State, block time port, go up to ring and send the intact frame of link, upgrade the mac address table of port then.After transmission node was received the intact frame of link that host node sends, the record chain line state was Link-Up-State, and the port that blocks is decontroled, and upgraded the mac address table of port.
Because the pre-forward-time time of node is greater than the delay process time pre-up-time of host node, so during host node is Link-Preup-State, node is in the port of Preforward state can overtimely not opened, and can not cause " broadcast storm " owing to loop yet.
The flow process of host node delay process in the Ethernet changing method of the embodiment of the invention being recovered to switch below with reference to Fig. 3 is described in detail.As shown in Figure 3, the method for delay process loop recovery switching of the present invention comprises step:
S301, host node record chain line state is Link-Down-State, also regularly sends health detection (hello) frame from master port to ring.
When S302, the inferior port of host node receive from the hello frame that master port sends, detect and think that link begins to recover.
S303, host node judge whether that the delay process loop recover to switch, if pre-up-time greater than 0, then delay process enters S304, otherwise delay process not enters S307.
S304, host node delay process loop recover to switch, and the record chain line state is Link-Preup-State, the record current time.
Whether S305, host node continue to detect link and break down again, if break down again, then enter S301, otherwise enter S306.
Whether S306, host node detect the Link-Preup-State state overtime, if do not have overtimely, then continues to wait for, otherwise enter S307.
S307, host node begin handle link and recover to switch, and are Link-Up-State with link state records, block time port, and send the intact frame of link (flush-up) to encircling to go up, and upgrade the port mac address table.
After transmission node was received the intact frame flush-up of the link frame that host node sends, the record chain line state was Link-Up-State, and S3 decontrols port one, and S4 decontrols port one, the mac address table of each node updates port.Because the pre-forward-time time of node is greater than the delay process time pre-up-time of host node, so during host node is Link-Preup-State, node is in the port of Preforward state can overtimely not opened, and can not cause " broadcast storm " owing to loop yet.
Ethernet changing method of the present invention is in the reality of certain Ethernet ring protection system is answered; the ring uplink is vibrated 10 times at short notice; be set to 0 when the delay process time of host node; promptly to link-recovery not during delay process; Ring protection system is to the hand-off process of link 20 times; operation obstruction and relieving operation to port will be handled more than 20 times; upgrading node M AC address table also will handle more than 20 times; during this period; business datum is substitute seriously; node is frequent to hardware operation, the probability of failure height.
Be set to 5 seconds when the delay process time of host node; when promptly the link-recovery time-delay being handled in 5 seconds; Ring protection system is to the hand-off process of link 2 times, and the operation of port is blocked and decontroled operation and will handle 2 times, upgrades node M AC address table and also will handle 2 times; during this period; the business datum packet loss time is 2 times link switchover time, is no more than 100ms, and business datum is stable; node is few to the hardware operation number of times, and probability of failure is low.
Those skilled in the art will readily understand additional advantages and modifications of the present invention.Therefore, above-mentioned at embodiment be described as the concrete Application Example of the present invention, the present invention aspect widely is not limited to the specific detail and the exemplary embodiments that illustrate and describe herein.Therefore, can under the condition that does not break away from by claim and the spirit or scope of the present invention that equivalent limited thereof, make various modifications.
Claims (5)
1. an ethernet ring network changing method is characterized in that, may further comprise the steps:
S202, configuration is used for the delay process time of link-recovery on the host node in ethernet ring network, and wherein, the described delay process time is only effective to described host node;
S204, under the situation of link occurs fault, described host node and transmission node are the link failure state with link state records and carry out respective handling; And
S206, under the situation of link-recovery, described host node switches according to described delay process time delays handle link;
Wherein, described S206 comprises: S2064, the described delay process time greater than 0 situation under, described host node record chain line state is the link-recovery state, and will be recorded as the link-recovery time current time; And S2066, detecting at described host node under the situation of link occurs fault, described host node is the link failure state with described link state records; Detect the current time at described host node and deduct under the situation of link-recovery time more than or equal to the described delay recovery time, described host node and described transmission node begin to carry out link switchover.
2. ethernet ring network changing method according to claim 1 is characterized in that, configuration time-delay forwarding time on the transmission node in described host node and ethernet ring network, wherein, described time-delay forwarding time is greater than the described delay process time.
3. ethernet ring network changing method according to claim 2 is characterized in that, described S204 comprises:
S2042, described host node detect under the situation of link occurs fault, and the record chain line state is the link failure state, decontrol time port, send the link failure frame on described ethernet ring network, and upgrade the mac address table of port; And
S2044, described transmission node receive under the situation of described link failure frame, and the record chain line state is the link failure state, and upgrade the mac address table of port.
4. ethernet ring network changing method according to claim 3, it is characterized in that, described S206 comprises S2062: be under 0 the situation in the described delay process time, described host node is the link serviceable condition with link state records, block time port, on described ethernet ring network, send the intact notification frame of link, and upgrade the mac address table of port.
5. ethernet ring network changing method according to claim 1 is characterized in that, the process that described host node and described transmission node begin to carry out link switchover comprises:
Step 1, described host node is the link serviceable condition with link state records, blocks time port, sends the intact notification frame of link on described ethernet ring network, and upgrades the mac address table of port; And
Step 2, described transmission node are the link serviceable condition with described link state records after receiving the intact notification frame of link of described host node transmission, decontrol the port that blocks, and upgrade the mac address table of port.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007101115534A CN101072154B (en) | 2007-03-02 | 2007-06-19 | Loop Ethernet switching method |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710080062.8 | 2007-03-02 | ||
| CN200710080062 | 2007-03-02 | ||
| CN2007101115534A CN101072154B (en) | 2007-03-02 | 2007-06-19 | Loop Ethernet switching method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101072154A CN101072154A (en) | 2007-11-14 |
| CN101072154B true CN101072154B (en) | 2011-09-21 |
Family
ID=38899152
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2007101115534A Active CN101072154B (en) | 2007-03-02 | 2007-06-19 | Loop Ethernet switching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101072154B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101599892A (en) * | 2008-06-03 | 2009-12-09 | 中兴通讯股份有限公司 | A kind of addresses forwarding table update method |
| CN101640622B (en) * | 2009-09-07 | 2012-05-23 | 中兴通讯股份有限公司 | Ethernet ring net protection method and transit node |
| CN101764731B (en) * | 2009-12-09 | 2011-11-30 | 熊伟 | Ethernet ring network algorithm switching method |
| CN101883038B (en) * | 2010-06-30 | 2015-08-12 | 中兴通讯股份有限公司 | Host node in the method for EAPS looped network protection switching and EAPS looped network |
| CN101997723A (en) * | 2010-11-18 | 2011-03-30 | 中兴通讯股份有限公司 | Service switching method and device |
| CN102055636B (en) * | 2010-12-13 | 2013-04-10 | 迈普通信技术股份有限公司 | Method for improving accelerated switching during EAPS looped network failure and EAPS looped network system |
| CN102739466B (en) * | 2011-04-13 | 2017-10-17 | 中兴通讯股份有限公司 | A kind of method and system for setting ethernet node to detect frame timeout duration |
| CN104081723A (en) * | 2012-01-17 | 2014-10-01 | 爱立信(中国)通信有限公司 | Methods and apparatus for network protection |
| CN104065546B (en) * | 2014-07-04 | 2018-01-19 | 引通通讯科技（上海）有限公司 | Ethernet protects switching method |
| CN106685783B (en) * | 2016-12-29 | 2019-06-28 | 瑞斯康达科技发展股份有限公司 | A kind of ring net protection method and system |
| CN107171840B (en) * | 2017-05-22 | 2019-12-06 | 杭州迪普科技股份有限公司 | Protection switching method and device based on ERPS protocol |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1812361A (en) * | 2006-01-23 | 2006-08-02 | 杭州华为三康技术有限公司 | Fast ring network protecting method and system |
| CN1870546A (en) * | 2006-01-21 | 2006-11-29 | 华为技术有限公司 | Method for obtaining communication system channel state |
-
2007
- 2007-06-19 CN CN2007101115534A patent/CN101072154B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1870546A (en) * | 2006-01-21 | 2006-11-29 | 华为技术有限公司 | Method for obtaining communication system channel state |
| CN1812361A (en) * | 2006-01-23 | 2006-08-02 | 杭州华为三康技术有限公司 | Fast ring network protecting method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101072154A (en) | 2007-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101072154B (en) | Loop Ethernet switching method | |
| CN100571180C (en) | Ensure the method and the device of ethernet ring network reliability service when transmission node fault | |
| CN101483570B (en) | Method, system and device for preventing looped network temporary loop circuit of relaying link | |
| RU2423008C2 (en) | METHOD AND SYSTEM FOR AUTOMATIC PROTECTION OF Ethernet NETWORK | |
| CN101677267B (en) | Method for retransmitting protocol frames in transannular way and sharing nodes in multiple rings of Ethernet | |
| CN101060485B (en) | Topology changed messages processing method and processing device | |
| CN101345683B (en) | Protocol packet transmission control method of Ethernet automatic protection switch system | |
| US20090207726A1 (en) | System and method for network recovery from multiple link failures | |
| CN101662421B (en) | Method and device for transmitting control message based on ethernet multi-ring network | |
| JP4790591B2 (en) | Ring node device | |
| CN100512128C (en) | Method and system for protecting multicast repeat path and service router | |
| CN101640622B (en) | Ethernet ring net protection method and transit node | |
| CN102255757B (en) | Link switching method and device | |
| CN101399737A (en) | Protection method and apparatus used for link convergence group of loop Ethernet | |
| CN101729426B (en) | Method and system for quickly switching between master device and standby device of virtual router redundancy protocol (VRRP) | |
| CN102035676A (en) | ARP (Address Resolution Protocol) interaction based method and equipment for detecting and recovering link fault | |
| CN103051536A (en) | Quick multicast switching method of two-layer redundant link | |
| CN101340346A (en) | Ring control method and apparatus of Ethernet looped network system | |
| EP2533470B1 (en) | Method and equipment for preventing repeated refreshing of ethernet ring node address table | |
| CN101714939A (en) | Fault treatment method for Ethernet ring network host node and corresponding Ethernet ring network | |
| CN102238067B (en) | Switching method and device on Rapid Ring Protection Protocol (RRPP) ring | |
| EP2466804A1 (en) | Protection switching method for ethernet automatic protection switching (eaps) ring network and master node in eaps ring network | |
| CN101237319B (en) | Time synchronization method in Ethernet ring network and Ethernet ring system | |
| CN101304359B (en) | Method and apparatus for RRPP looped network link recuperation | |
| CN108833247B (en) | Method and device for re-switching after link failure recovery |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |