WO2014139175A1 - 双上行相切环收敛的方法、设备和系统 - Google Patents

双上行相切环收敛的方法、设备和系统 Download PDF

Info

Publication number
WO2014139175A1
WO2014139175A1 PCT/CN2013/072759 CN2013072759W WO2014139175A1 WO 2014139175 A1 WO2014139175 A1 WO 2014139175A1 CN 2013072759 W CN2013072759 W CN 2013072759W WO 2014139175 A1 WO2014139175 A1 WO 2014139175A1
Authority
WO
WIPO (PCT)
Prior art keywords
network device
port
conjugate
arp
entry
Prior art date
Application number
PCT/CN2013/072759
Other languages
English (en)
French (fr)
Inventor
李毅
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to ES13861532.3T priority Critical patent/ES2586941T3/es
Priority to EP13861532.3A priority patent/EP2802105B1/en
Priority to PCT/CN2013/072759 priority patent/WO2014139175A1/zh
Priority to CN201380000223.1A priority patent/CN104160667B/zh
Priority to US14/315,541 priority patent/US9553760B2/en
Publication of WO2014139175A1 publication Critical patent/WO2014139175A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06Management of faults, events, alarms or notifications
    • H04L41/0654Management of faults, events, alarms or notifications using network fault recovery
    • H04L41/0659Management of faults, events, alarms or notifications using network fault recovery by isolating or reconfiguring faulty entities
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/08Arrangements for detecting or preventing errors in the information received by repeating transmission, e.g. Verdan system
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/22Arrangements for detecting or preventing errors in the information received using redundant apparatus to increase reliability

Definitions

  • Embodiments of the present invention relate to the field of computer communications, and in particular, to a method, device, and system for convergence of dual uplink tangent loops. Background technique
  • dual uplink mode is usually used, that is, one downstream device is connected to two upstream devices at the same time to avoid single point of failure and improve network reliability.
  • the dual uplink mode can improve the network reliability, it introduces a loop problem, which requires a broken ring protocol, such as spanning tree protocol (STP)/multiple spanning tree protocol (MSTP). / Rapid spanning tree protocol (MSTP), rapid ring protection protocol (RRPP), and flexible link (SmartLink) protocol to eliminate loops.
  • STP spanning tree protocol
  • MSTP multiple spanning tree protocol
  • RRPP rapid spanning tree protocol
  • SmartLink flexible link
  • the multiple uplink network devices are dual-homed to a pair of identical upstream network devices, and the links between the upstream network devices form a plurality of dual uplink tangential ring networks, and the virtual local area network (virtual local area) Network, abbreviated as VLAN)
  • VLAN virtual local area network
  • PSPV virtual local area network
  • all links of multiple dual uplink tangent rings share one service VLAN, although the deployment of the broken ring protocol can prevent traffic. Looping in the broadcast domain, but the operation of the existing broken ring protocol is to perform the clear/refresh operation of related entries at the granularity of VLAN.
  • any double-uplink tangent ring fault will cause other non-faulty dual-uplink tangent ring related entries to be cleared/refreshed, resulting in traffic interruption.
  • the user volume is large, a large number of entry updates will be triggered. , resulting in long-term business interruption.
  • the embodiment of the present invention provides a method, a device, and a system for convening a dual uplink tangent ring, which can solve the problem that a double uplink tangential ring fault triggers the refresh of other non-fault dual uplink tangential ring entries. / Clearing, causing traffic interruption problems, which can solve the problem of a large number of entries being updated when a large amount of users is large, resulting in long-term business interruption.
  • the present invention provides a method for convergence of dual uplink tangent rings, where one or more downstream devices are dual-homed to a first network device and a second network device, and the first network device and the first The link between the two network devices constitutes one or more dual uplink tangential rings, and the method includes:
  • the first port of the device is a conjugate port, and in the medium access control MAC table and the address resolution protocol ARP table of the first network device, the first port of the first network device is cleared.
  • the MAC address entry and the ARP entry of the interface are used to clear all MAC entries that are outbound interfaces of the first network device.
  • the conjugate port is an access side port of the first network device.
  • the medium access control MAC of the first network device is cleared.
  • the first port of the first network device is an entry of the outbound interface, and all the non-commons of the first network device in the MAC table of the first network device are cleared.
  • the yoke port is an entry of the outbound interface, it also includes:
  • the entry update message includes virtual local area network (VLAN) VLAN information; correspondingly, the In the media access control MAC address table and the address resolution protocol ARP table of the device, the MAC address entry and the ARP entry that are the outbound interface of the first network device are cleared, and all the first network devices are cleared.
  • the non-conjugate port is the MAC address entry of the outbound interface.
  • the location of the first network device is cleared.
  • the first port is a MAC address entry and an ARP entry of the outbound interface, and all MAC entries that are outbound interfaces of the first network device are removed.
  • the ARP request is performed by the first port and all the non-conjugated ports of the first network device, and the ARP request is performed on the ARP entry that is the outbound interface of the first network device.
  • ARP request is made for the ARP entry that is the outbound interface of the outbound interface of all the VLANs identified by the VLAN information from the first port and all the non-conjugated ports of the first network device.
  • the entry update message is a flexible link SmartLink An extended Flush packet of the protocol, where the extended Flush packet includes a first control virtual local area network identifier; and before determining that the first port of the first network device is a conjugate port, the method further includes:
  • the first network device obtains the first control virtual local area network identifier in the extended Flush message, and determines to determine the first control virtual local area network identifier and the first network setting
  • the identifiers of the control virtual local area networks configured on the first port are the same.
  • the method further includes: the first network device copying and forwarding The entry updates the message to the second port of the second network device.
  • the first port is a non-conjugate port, and the entry update message is discarded.
  • the present invention provides a network device, including:
  • a receiving module configured to receive an entry update message from the first port of the network device, and a determining module, configured to determine whether the first port of the network device is a conjugate port;
  • An entry update module configured to: when determining that the first port of the network device is a conjugate port, in the media access control MAC table and the address resolution protocol ARP table of the network device, clear all The first port of a network device is a MAC address entry and an ARP entry of the outbound interface, and all MAC entries that are outbound interfaces of the first network device are removed.
  • the conjugate port is an access side port of the first network device.
  • the method further includes: an ARP requesting module, configured to: when determining that the first port of the network device is a conjugate port, from the network device The first port and all the non-conjugate ports perform an ARP request for all ARP entries that are outbound interfaces of the network device.
  • the entry update message includes virtual local area network VLAN information, where the entry update module is specifically configured to: when determining that the first port of the network device is a conjugate port, in the network In the MAC address table and the ARP table of the device, in the MAC address entry and the ARP entry in all the VLANs identified by the VLAN information, all MAC entries and ARPs that are outbound interfaces of the first port of the network device are cleared.
  • the entry, the MAC address entry that is the outbound interface of the network device is removed, and the ARP request module is specifically configured to: when determining that the first port of the network device is a conjugate port, ARP request is made to the ARP entry of the outbound interface that is the non-conjugated port in all the VLANs identified by the VLAN information from the first port and all the non-conjugated ports of the network device.
  • the entry update message is a flexible link SmartLink
  • the extended Flush packet of the protocol where the extended Flush packet includes a first control virtual local area network identifier, and the network device further includes:
  • a second determining module configured to obtain the first control virtual local area network identifier in the entry update message, and determine the first control virtual local area network identifier and the identifier of the control virtual local area network configured on the first port Whether they are the same;
  • the first determining module is specifically configured to: when the determining result of the second determining module is that the first control virtual local area network identifier is the same as the identifier of the control virtual local area network configured on the first port, Whether the first port of the network device is a conjugate port.
  • the method further includes:
  • the present invention provides a network device, including:
  • a state sensing module configured to detect a state change of the first port of the network device, and a determining module, configured to determine whether the first port of the network device is a conjugate port;
  • An entry update module configured to: when determining that the first port of the network device is a conjugate port, clearing, in the MAC table and the ARP table of the network device, all the devices that are the first network device.
  • the first port is the MAC address entry and the ARP entry of the outbound interface, and all the MAC entries that are the outbound interfaces of the first network device are removed.
  • the conjugate port is the first network device. Access side port.
  • the method further includes: an ARP requesting module, configured to: when determining that the first port of the network device is a conjugate port, from the network device The first port and all the non-conjugate ports perform an ARP request for all ARP entries that are outbound interfaces of the network device.
  • the method further includes:
  • a packet generation and sending module configured to generate a first entry update message, and send the first entry update message to the second network device
  • the method further includes a receiving module, configured to use the network device The second port receives the second entry update message;
  • the determining module is further configured to determine whether the second port of the network device is a conjugate port.
  • the present invention provides a dual uplink tangential ring system, including a first upstream network device, a second upstream network device, and one or more downstream network devices; wherein one or more downstream network devices are dual uplink Linking to the first upstream network device and the second upstream network device, and the link between the first upstream network device and the second upstream network device constitutes one or more dual uplink tangential rings;
  • the access side port in the port associated with the one or more dual uplink tangent rings on the first upstream network device and the second upstream network device is a conjugate port
  • the first upstream network device is configured to receive a first entry update message from the first port of the first upstream network device, or to detect that a state of the first port of the first upstream network device changes; Whether the first port of the first upstream network device is a conjugate port, and when determining that the first port of the first upstream network device is a conjugate port, the MAC of the first upstream network device In the table and the ARP table, all the MAC entries and the ARP entries that are the outgoing interfaces of the first upstream network device are cleared, and all the non-conjugated ports of the first upstream network device are cleared.
  • the item performs an ARP request to refresh the related entries of the dual uplink tangent ring where the first port is located.
  • the first upstream network device is further configured to receive a first entry update report from a first port of the first upstream network device After the text, the first entry update message is copied and forwarded to the second upstream network device; or, the state of the first port of the first upstream network device is changed, and the second entry update message is generated. And sent to the second upstream network device;
  • the second upstream network device is configured to use the second end of the second upstream network device
  • the port receives the first entry update message or the second entry update message, and determines whether the second port of the second upstream network device is a conjugate port.
  • the first downstream network device is configured to send the first entry update message to the first port of the first upstream network device.
  • all ports related to multiple dual uplink tangential rings on the network device are divided into a conjugate port and a non-conjugate port, and the network device receives the entry update report from the first port.
  • the first network device determines whether the first port is a conjugate port, and when determining that the first port is a conjugate port, the MAC table and the ARP table of the network device Clear all MAC entries and ARP entries that take the first port as the outgoing interface, clear all MAC entries with the non-conjugated port as the outbound interface, and keep the MAC table with all the conjugated ports as the outgoing interfaces.
  • the fault of one dual uplink tangential ring may be caused to cause the refresh/purge of the other non-faulty dual uplink tangential ring entries, which may cause the traffic to be interrupted, and the large number of entries may be updated when the user volume is large. Time business interruption problem.
  • FIG. 1 is a schematic diagram of a networking diagram of multiple dual uplink tangent rings according to an embodiment of the present invention
  • FIG. 2 is a flow chart of a method for convergence of dual uplink tangent loops according to an embodiment of the present invention
  • FIG. 3 is a flowchart of another method for convergence of dual uplink tangent loops according to an embodiment of the present invention.
  • FIG. 4 is a flow chart of another method for convergence of dual uplink tangent rings according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a network device according to an embodiment of the present invention
  • FIG. 7 is a schematic structural diagram of a network device according to an embodiment of the present invention
  • FIG. 8 is a schematic diagram of another embodiment of the present invention.
  • FIG. 10 is a schematic diagram of another hardware structure of a network device according to an embodiment of the present invention
  • FIG. 11 is a schematic diagram of a dual uplink tangential ring system according to an embodiment of the present invention.
  • a plurality of dual uplink tangential ring related ports are divided into a conjugate port and a non-conjugate port; wherein, the side port; the non-conjugate port represents a port between network devices that constitute a common tangent edge of the dual uplink tangential ring, That is, the port of the common tangent edge of the dual uplink tangent ring.
  • Network devices can maintain a set of conjugated ports.
  • N is a positive integer
  • FIG. 1 is a schematic diagram of networking of multiple dual uplink tangent rings according to an embodiment of the present invention
  • two downstream network devices namely, switch 1 and switch 2
  • Upstream network equipment - router 1 and router 2 with router
  • the link between 1 and router 2 constitutes two dual uplink tangent rings, wherein router 1 and the network device one router 1 and router 2 serve as service gateways, and the PSPV deployment mode and two dual uplink tangent rings are used.
  • a service VLAN is shared, and a broken ring protocol, such as SmartLink, RRPP, STP, RSTP, or MSTP, is deployed to prevent traffic from being looped in the broadcast domain.
  • SmartLink such as SmartLink, RRPP, STP, RSTP, or MSTP
  • ports 11, 12, and 15 in router 1 are ports on multiple dual uplink tangent rings, where ports 11 and 12 are access side ports, which are both conjugate ports, and port 15 is dual uplink.
  • the port of the common tangent edge of the tangent ring is a non-conjugated port; likewise, ports 21, 22 and 25 of router 2 are ports on multiple dual uplink tangent rings, where ports 21 and 22 are access side ports Both are conjugate ports, and port 25 is a common tangent edge port of the dual uplink tangential ring, which is a non-conjugated port.
  • FIG. 2 is a schematic diagram of networking of another dual uplink tangent ring according to an embodiment of the present invention; as shown in FIG. 2, three downstream network devices, namely, switch 1, switch 2, and switch 3, wherein the switch 1.
  • Switch 2 dual uplink belongs to a pair of upstream network devices, router 1 and router 2, and the link between router 1 and router 2 constitutes two dual uplink tangent rings, where router 1 and router 2 The link constitutes a common tangent edge of two dual uplink tangent rings; and the switch 3 dual uplink belongs to a pair of upstream network devices - router 1 and router 3, and the link between router 1 and router 3 constitutes 1 Dual uplink tangent ring, in which the link between router 1 and router 3 constitutes a common tangent edge of the dual uplink tangent ring; Router 1, Router 2 and Router 3 serve as service gateways, and use PSPV deployment mode, 3 A dual uplink tangent ring shares a service VLAN and a broken ring protocol is deployed
  • ports 11, 12, 13, 15, and 16 in router 1 are ports on multiple dual uplink tangential rings, and ports 11, 12, and 13 are access side ports, which are conjugate ports.
  • port 15 and 16 is the port of the common tangent edge of the dual uplink tangential ring, which is a non-conjugated port;
  • ports 21, 22 and 25 of router 2 are ports on multiple dual uplink tangential rings, where ports 21 and 22 are The access side ports are all conjugate ports
  • the port 25 is a common tangent edge port of the dual uplink tangential ring, which is a non-conjugated port;
  • the ports 33 and 36 of the router 3 are dual uplink tangential rings.
  • Ports, where port 33 is an access side port, is a conjugate port
  • port 36 is a common tangent edge port of the dual uplink tangential ring, which is a non-conjugated port.
  • the technical solution of the present invention is used in a scenario where multiple downstream network devices are dual-homed to a pair of upstream network devices, and the same service VLAN is deployed to form multiple dual uplink tangential rings in the same service VLAN.
  • the technical solution of the present invention is applicable when only one downstream network device is dual-homed to a pair of upstream network devices, that is, a dual uplink tangential ring.
  • conjugate port and the non-conjugate port are for customizing the access side port and the non-access side port, and the custom naming is given; of course, other naming can also be made. Not limited.
  • An embodiment of the present invention provides a method for convergence of dual uplink tangent rings. As shown in FIG. 2, the method includes:
  • the first network device receives an entry update message from the first port of the first network device, or detects that the state of the first port of the first network device changes.
  • the item update message may be an extended Flush message or a topology change (TC) message.
  • the entry update message is an extended Flush message, for example, the control type (Control Type) or the control version of the Flush message in the SmartLink protocol may be extended.
  • the control version of the field obtains the extended Flush packet, which is not limited by the present invention; if the deployed broken protocol is RSTP or MSTP, etc.
  • the item update message is an extended TC message.
  • the protocol version identifier field of the TC message may be extended.
  • the extended TC message is obtained by using an identifier value that is not used in the current protocol.
  • the invention is not limited thereto.
  • the entry update message can also be customized according to the corresponding broken protocol.
  • the first network device perceives a state change of the first port of the first network device, where the first network device senses that the state of the first port of the first network device is blocked.
  • the first network device determines whether the first port of the first network device is a conjugate port.
  • the first network device is in a media access control (MAC) table and (address) of the first network device.
  • the resolution protocol (abbreviated as ARP) table clears all MAC entries and ARP entries that are the outbound interface of the first network device, and clears all non-conjugated ports of the first network device.
  • the MAC address entry of the outbound interface is a media access control (abbreviated as ARP).
  • an ARP request is performed on all ARP entries that are outbound interfaces of the first network device.
  • a related entry of the dual uplink tangential ring of the first port; the ARP request is initiated to speed up the update of the related entries, so that the service can be quickly restored.
  • the first network device only clears/refreshes the entry related to the dual uplink tangential ring of the first port in the MAC table of the first network device and the ARP table, and retains the first network.
  • the MAC table of the device and the ARP table are all in common.
  • the yoke port is an entry of the outbound interface, that is, the entries related to other dual uplink tangential rings are reserved, and are cleared/refreshed.
  • the first network device does not process the update packet of the entry, and specifically, the entry update packet may be discarded. Or no processing of the state change of the first port.
  • the first network device copies the entry update message and forwards the packet to the second network device; specifically, the first network device copies the entry update message, and the copied The entry update message is forwarded from the second port of the first network device to the link connected to the second network device by the first network device, that is, the common tangent edge of the dual uplink tangent ring, to the The second network device.
  • the second network device receives, from the second port of the second network device, the entry update message that is copied and forwarded by the first network device, where the second network device is The second port is a port of the common tangent edge of the dual uplink tangential ring on the second network device, and is a non-conjugated port. Therefore, the second network device discards the entry update message.
  • the dual uplink tangential ring convergence method provided in this embodiment is to divide all ports related to multiple dual uplink tangential rings on the first network device into conjugate ports and non-conjugate ports, and the first network device is from the first The port receives the entry update message, or detects the state change of the first port, the first network device determines whether the first port is a conjugate port, and when determining that the first port is a conjugate port, In the MAC address table and the ARP table of a network device, clear all MAC entries and ARP entries that take the first port as the outbound interface, and clear all MAC entries with the non-conjugated port as the outbound interface.
  • the conjugate port is the MAC address entry and the ARP entry of the outbound interface; and the ARP request is performed on the ARP entry that is the outbound interface of the non-conjugated port on the first port and all the non-conjugated ports to refresh the The related entry of the dual uplink tangent ring where the first port is located.
  • the fault of the dual-uplink tangential ring triggers the refresh/clear of the other fault-free dual-uplink tangent ring entries, which causes the traffic to be interrupted. In this way, when a large amount of users is large, a large number of entries are updated, resulting in long-term service interruption.
  • An embodiment of the present invention provides a method for convergence of dual uplink tangent rings. Referring to FIG. 3, the method includes:
  • the first downstream network device detects that the primary link of the first downstream network device to the second upstream network device is faulty by using a broken ring protocol, and the first downstream network device is sent to the first upstream network device.
  • the alternate link is upgraded to the primary;
  • the ports of the first downstream network device to the standby link of the first upstream network device are the standby uplink port of the first downstream network device and the first port of the first upstream network device;
  • the standby link of the first downstream network device to the first upstream network device is upgraded, and the method includes: the first downstream network device changing the standby uplink port of the first downstream network device from a blocked state Forwarding state; correspondingly, the first port of the first upstream network device changes from a blocked state to a forwarded state.
  • the switch 1 passes the broken link protocol, and the SmartLink protocol is taken as an example to sense the primary link of the switch 1 to the router 1.
  • a failure occurs and the standby link from switch 1 to router 2 is promoted to the primary, that is, the port 102 of switch 1 is changed from the blocked state to the forwarded state; accordingly, the port 21 of the router 2 is changed from the blocked state to the forwarded state.
  • the first downstream network device sends an entry update to the first upstream network device.
  • the first downstream network device uses the backup of the first downstream network device from the first downstream network device.
  • the uplink port sends an entry update message to the first port of the first upstream network device.
  • the entry update message is an extended Flush message or an extended TC message; for details, refer to 201, and details are not described herein again.
  • the switch 1 sends an entry update message to the router 2, that is, the extended Flush message; specifically, the switch 1 slave switch Port 102 of 1 sends an extended Flush message to port 21 of router 2.
  • the first upstream network device receives the entry update message from a first port of the first upstream network device.
  • Router 2 receives the extended Flush message from port 21.
  • the first upstream network device determines whether the first port of the first upstream network device is a conjugate port
  • the first upstream network device queries whether the first port exists in the conjugate port set of the first upstream network device, and determines the first of the first upstream network device according to the query result. Determining whether the port is a conjugate port; determining the first port of the first upstream network device if the first port of the first upstream network device is found in the conjugate port set of the first upstream network device The first port is a conjugate port; if the first port of the first upstream network device is not found in the conjugate port set of the first upstream network device, determining the first upstream network device The first port is a non-conjugated port.
  • first port of the first upstream network device is a conjugate port
  • execute 305 if it is determined that the first port of the first upstream network device is a non-conjugate port, perform 306.
  • the router 2 determines whether the port 21 is a conjugate port. If the port 21 is a conjugate port, the process 305 is performed to refresh all entries related to the dual uplink tangential ring of the switch 1 on the router 2.
  • the entry update message includes a first control virtual local area network identifier; and correspondingly, before 304, the method further includes: 304a, the first And obtaining, by the upstream network device, the first control virtual local area network identifier in the entry update message, and determining, by the first control virtual local area network identifier, the control configured on the first port of the first upstream network device Whether the identifier of the virtual local area network is the same; if the identifier of the first control virtual local area network is the same as the identifier of the control virtual local area network configured on the first port, proceed to execute 304; if not, the update message of the entry is not performed.
  • the processing can be directly forwarded according to the SmartLink protocol, and the processing of the method ends.
  • the first upstream network device clears all the devices in the MAC table and the ARP table of the first upstream network device.
  • the first port of the first upstream network device is a MAC address entry and an ARP entry of the outbound interface, and all MAC entries that are outbound interfaces of the first upstream network device are removed.
  • an ARP request is made to the ARP entry that is the outbound interface of the non-conjugated port from the first port and all the non-conjugated ports of the first upstream network device, to refresh the first port.
  • the MAC address table of the first network device and the entry related to the dual uplink tangential ring of the first port in the ARP table are cleared/refreshed; and the MAC table of the first upstream network device is reserved.
  • the entry in the ARP table with all the other conjugate ports as the outgoing interface, that is, the other dual uplink tangential ring related entries are retained, so that in connection with the diagrams la and lb, the router 2 determines that the port 21 is determined to be When the conjugate port is used, Router 2 clears the port 21 as the port in the MAC table and ARP table of Router 2.
  • the MAC address entry and the ARP entry of the interface are cleared.
  • All the non-conjugated ports, that is, the port 25, are the MAC entries of the outbound interface.
  • the ARP request is performed to refresh the related entries of the dual uplink tangential ring of the switch 1. In this way, only the related entries of the dual uplink tangent ring of the switch 1 are cleared/refreshed, and the related items of the dual uplink tangential rings of the switch 2 and the switch 3 are not cleared/refreshed; thus, the switch 1 is avoided.
  • the failure of the dual uplink tangent ring affects the dual uplink tangential rings of Switch 2 and Switch 3.
  • the entry update message further includes the VLAN information, which may be one or more VLAN identifiers.
  • the extended Flush packet of the SmartLink protocol carries one or more refreshed packets through the VLAN bitmap field. Service VLAN ID.
  • the entry update message includes the VLAN information
  • the entry update message includes the VLAN information
  • all the VLANs identified by the VLAN information in the entry are updated in the entry. And performing a clear/refresh operation on the related entry of the dual uplink tangential ring of the first port of the first upstream network device.
  • the entry update message includes the VLAN information
  • the first port of the first upstream network device is determined to be a conjugate port
  • the first upstream network device is in the first upstream.
  • the MAC table and the ARP table of the network device all the MAC entries and the ARP entries that are the outbound interfaces of the first upstream network device are cleared, and all the non-first upstream network devices are cleared.
  • the conjugate port is the MAC address entry of the outbound interface, and is specifically: in the MAC table and the ARP table of the first upstream network device, the MAC address entry in all VLANs identified by the VLAN information in the entry update packet In the ARP entry, all the MAC entries and the ARP entries that are the outbound interfaces of the first upstream network device are cleared, and all the non-conjugate ports of the first upstream network device are cleared.
  • the first upstream from the first The first port and all the non-conjugated ports of the network device perform an ARP request for all the ARP entries that are the outbound interfaces of the non-conjugated port, specifically: the first port from the first upstream network device And all the non-conjugated ports perform ARP requests for all ARP entries with the non-conjugated port as the outbound interface in all the VLANs identified by the VLAN information in the update packet.
  • the first upstream network device may also replicate the entry update message and forward the packet to the second upstream network device.
  • the first upstream network device replicates the entry update message.
  • the duplicated entry update message is sent from the second port of the first upstream network device to the link connected to the second upstream network device by the first upstream network device, that is, dual uplink
  • the common tangent edge of the puncturing ring is forwarded to the second upstream network device.
  • the second upstream network device receives, from the second port of the second upstream network device, the entry update message that is copied and forwarded by the first upstream network device, because the second upstream network device
  • the second port is a common tangent edge port of the dual uplink tangential ring on the second upstream network device, and is a non-conjugated port, so the second upstream network device discards the entry update report. Text.
  • router 2 forwards the entry update message from port 25 to the port 15 of router 1 through the link between router 2 and router 1; router 1 receives the message from port 15 After the entry is updated, it is determined that the port 15 is a non-conjugate port, and the router 1 does not process the update packet of the entry, and discards the update packet of the entry.
  • the first upstream network device does not process the update packet of the entry; specifically, the first upstream The network device may discard the entry update message; the method ends. Further, after the first upstream network device senses that the primary link device to the second upstream network device recovers from the primary link by the broken protocol, the first downstream network device performs the a port of the first downstream network device to the standby link of the first upstream network device, that is, a state of the standby uplink port of the first downstream network device changes from a forwarding state to a blocking state, and correspondingly, the And changing a state of the first port of an upstream network device from a forwarding state to a blocked state; and transmitting a new entry update message to the second upstream network device; and then the second upstream network device receives and processes
  • the processing procedure and principle of the new item update text are similar to those of the foregoing 303 307, and are not described herein again.
  • the dual uplink tangential ring convergence method provided in this embodiment is configured to divide all ports related to multiple dual uplink tangential rings on the upstream network device into conjugate ports and non-conjugate ports, and the upstream network device receives from the first port.
  • the first port is a conjugate port, and when it is determined that the first port is a conjugate port, in the MAC table and the ARP table of the upstream network device, all are cleared.
  • the first port is the MAC address entry and the ARP entry of the outbound interface.
  • the MAC address entry and the ARP entry of all the conjugate ports are the outbound interfaces.
  • An embodiment of the present invention provides another method for convergence of dual uplink tangent rings. Referring to FIG. 4, The method includes:
  • the first upstream network device senses that a state of the first port of the first upstream network device changes.
  • the first downstream network device is dual-homed to the first upstream network device and the second upstream network device, where the first downstream network device is respectively connected to the first port of the first upstream network device and the first On the first port of the upstream network device.
  • a broken protocol such as MSTP or RSTP, is run on the first upstream network device and the second upstream network device. After the primary link of the first downstream network device to the second upstream network device fails, the standby link of the first downstream network device to the first upstream network device is elected as the primary, that is, the first upstream network.
  • the state of the first port of the device changes to a forwarding state, which is usually done by a broken protocol, such as MSTP or RSTP, which is not extended by the present invention.
  • the first upstream network device senses that the state of the first port of the first upstream network device changes, specifically, changes from a blocked state to a forwarded state.
  • Router 1 is the root node
  • ports 11, 12, and 15 of Router 1 are designated ports, and are in the forwarding state.
  • Ports 21 and 22 of Router 2 are The alternate port is in the blocking state
  • the port 25 is the root port, and is in the forwarding state
  • the port 101 of the switch 1 is the root port, in the forwarding state
  • the port 102 is the designated port, and is in the blocked state
  • the port 201 of the switch 2 is the root.
  • the port is in the forwarding state, and the port 202 is in the blocked state.
  • port 21 of router 2 changes from blocking state to forwarding state, and the standby link of switch 1 to router 2 rises to the primary; router 2 senses the state transmission change of port 21. .
  • the first upstream network device determines the manner of the first upstream network device. Whether the first port is a conjugate port;
  • the first upstream network device may query whether the first port of the first upstream network device exists in the conjugate port set of the first upstream network device, and determine, according to the query result, the first Whether the first port of an upstream network device is a conjugate port; if the first port of the first upstream network device is found in the conjugate port set of the first upstream network device, determining The first port of the first upstream network device is a conjugate port; if the first port of the first upstream network device is not found in the conjugate port set of the first upstream network device, Determining that the first port of the first upstream network device is a non-conjugated port.
  • first port of the first upstream network device is a conjugate port
  • execute 403 if it is determined that the first port of the first upstream network device is a non-conjugate port, perform 404.
  • the router 2 determines whether the port 21 is a conjugate port. If it is determined that the port 21 is a conjugate port, perform 403 to refresh all the routers associated with the dual uplink tangential ring of the switch 1. Entry.
  • the first upstream network device clears all the devices in the MAC table and the ARP table of the first upstream network device.
  • the first port of the first upstream network device is a MAC address entry and an ARP entry of the outbound interface, and all MAC entries that are outbound interfaces of the first upstream network device are removed.
  • an ARP request is made to the ARP entry that is the outbound interface of the non-conjugated port from the first port and all the non-conjugated ports of the first upstream network device, to refresh the first port.
  • All the non-conjugated ports that is, port 25, are deleted as the MAC address entries of the outbound interface. Further, all ports 15 and 25 are used as ports.
  • the ARP entry of the interface performs an ARP request to refresh the related entries of the dual uplink tangent ring. In this way, Router 2 does not clear all the entries, but only clears/refreshes the related entries of the dual uplink tangent ring of Switch 1. This prevents the double uplink tangential ring of Switch 1 from affecting the dual switch 2. Uplink tangent ring.
  • the first upstream network device does not change a state of the first port of the first upstream network device. deal with.
  • the first upstream network device may generate an entry update packet, such as an extended TC packet. And sent to the second upstream network device. Specifically, the first upstream network device generates the entry update message, and the generated entry update message is sent from the second port of the first upstream network device to the first upstream network.
  • the link connected to the second upstream network device that is, the common tangent edge of the dual uplink tangential ring, is sent to the second upstream network device.
  • the second upstream network device receives, from the second port of the second upstream network device, the entry update message generated and sent by the first upstream network device, because the second upstream network device Place The port is a non-conjugated port, so the second upstream network device discards the entry update message.
  • the first upstream network device may further send the entry update message to the first downstream network device from the first port of the first upstream network device, so that the first The downstream network device clears/refreshes related entries.
  • router 2 In conjunction with the example scenario of Figure 151, router 2 generates an extended TC message, and from port 25, through the link between router 2 and router 1, to port 15 of router 1; router 1 receives from port 15 After the TC is extended, it is determined that the port 15 is a non-conjugated port, and the router 1 does not process the extended TC packet, and discards the extended TC packet. Meanwhile, the router 2 can also be from the port 21 The switch sends an extended TC packet to switch 1, so that switch 1 clears the MAC entry.
  • the status of the first port of the second upstream network device changes to a forwarding state after the primary link of the first downstream network device recovers from the failure of the primary link device,
  • the first port of the first upstream network device becomes blocked, which is usually done by a broken protocol, such as MSTP or RSTP, which is not extended by the present invention.
  • the second upstream network device senses a state change of the first port of the second upstream network device, that is, changes from a blocked state to a forwarding state, and the second upstream network device determines the second upstream network device Whether the first port is a conjugate port.
  • the second upstream network device determines that the first port of the second upstream network device is a conjugate port, clearing/refreshing the dual uplink tangent of the first port of the second upstream network device
  • the process and the principle of the ring-related entry are similar to the above-mentioned 403, and are not described here.
  • the second upstream network device may generate an entry update message, for example, an extended TC message, and send it to the The first upstream network device; specifically, the second port sends the entry update message to the first downstream network device;
  • the first upstream network device receives the entry update message from the second port of the first upstream network device, because the second port of the first upstream network device is a port, The first upstream network device does not process the update packet of the entry, and discards the update packet of the entry.
  • the second upstream network device sends the entry update message to the first downstream network device to the first port of the second upstream network device, so that the first downstream network The device clears/refreshes related entries.
  • the first upstream network device sends the entry update message only to the first port of the first upstream network device that becomes the forwarding state, instead of going to the first upstream in the prior art. All the designated ports of the network device send the entry update message.
  • the primary link of switch 1 to router 1 is restored, the alternate port of router 2, that is, port 21 is again changed to the blocked state, and port 11 of router 1 is restored to the forwarding state, and router 1 senses The state of the port 11 is changed, and it is determined whether the port 11 is a conjugate port.
  • the MAC table and the ARP table of the router 1 the MAC entry and the ARP table with the outbound interface of port 11 are cleared.
  • all the non-conjugated ports that is, port 15, are the MAC entries of the outbound interface, and the ARP entries of all the ARP entries that are outbound on the interface are further forwarded from port 11 and port 15.
  • the router 1 can generate an extended TC message, and the slave port 15 transmits the link to the port 2 of the router 2 through the link between the router 2 and the router 1; after receiving the extended TC message from the port 25, the router 2 receives the extended TC message. It is determined that the port 25 is a non-conjugated port, and the router 2 does not process the extended TC packet, and discards the extended TC packet. At the same time, the router 1 can also send the extended TC packet from the port 11 to the switch 1, so that the switch 1 clears the MAC entry.
  • the dual uplink tangential ring convergence method provided in this embodiment is configured to divide all ports related to multiple dual uplink tangential rings on the upstream network device into conjugate ports and non-conjugate ports, and upstream.
  • the network device detects that the state of the first port is changed, and determines whether the first port is a conjugate port. When the first port is determined to be a conjugate port, the network device clears the MAC table and the ARP table of the network device. The MAC address entry and the ARP entry of the outbound interface are removed from the MAC address entry and the ARP entry. The MAC address entry and the ARP table with all the conjugate ports as the outgoing interfaces are deleted.
  • the network device includes:
  • the receiving module 501 is configured to receive an entry update message from the first port of the network device.
  • the first determining module 502 is configured to determine whether the first port of the network device is a conjugate port
  • An entry update module 503 configured to: when determining that the first port of the network device is a conjugate port, clear all the devices that are the first network device in a MAC table and an ARP table of the network device.
  • the first port is a MAC address entry and an ARP entry of the outbound interface, and all MAC entries that are outbound interfaces of the first network device are removed.
  • the network device may further include:
  • the ARP requesting module 504 is configured to: when the first port of the network device is a conjugate port, from the first port and all non-conjugate ports of the network device, to all the network devices The non-conjugated port performs an ARP request for the ARP entry of the outbound interface.
  • the network device further includes:
  • the discarding module 505 is configured to discard the entry update file when the first port of the network device is determined to be a non-conjugate port. In this way, if the first port is a non-conjugated port, the network device does not process the update packet of the entry.
  • the copy forwarding module 506 is configured to copy the entry update message and forward the message to the second network device.
  • the copy forwarding module 506 is specifically configured to copy the entry update message, and the copied entry is Updating the message from the second port of the network device to the second network device by using a link between the network device and the second network device, that is, a common tangent edge of the dual uplink tangential ring .
  • the second network device receives, from the second port of the second network device, the entry update message that is copied and forwarded by the network device, because the second port of the second network device
  • the port of the common tangent edge of the dual uplink tangent ring on the second network device is a non-conjugated port, so the second network device discards the entry update message.
  • the entry update message includes the first control virtual local area network identifier.
  • the network device further includes a second determining module 507. And determining, by the first control virtual local area network identifier in the entry update message, whether the identifier of the first control virtual local area network identifier and the control virtual local area network configured on the first port are the same;
  • the first determining module 502 is specifically configured to: when the determining result of the second determining module 508 is that the first control virtual local area network identifier is the same as the identifier of the control virtual local area network configured on the first port, Determining whether the first port of the network device is
  • the discarding module 505 is further configured to: when the judgment result of the second judging module 507 is that the first control virtual local area network identifier and the identifier of the control virtual local area network configured on the first port are not At the same time, the entry update is discarded.
  • the entry update message further includes the VLAN information, which may be one or more VLAN identifiers.
  • the extended Flush packet of the SmartLink protocol carries the one that needs to be refreshed through the VLAN bitmap field. Multiple service VLAN IDs.
  • the entry update module 503 is specifically configured to: when determining that the first port of the network device is a conjugate port, in the MAC table and the ARP table of the network device, the entry is updated.
  • the MAC address entry and the ARP entry in the outbound interface of the first port of the network device are cleared in the MAC address entry and the ARP entry in the VLAN.
  • the non-conjugate port of the network device is the MAC address entry of the outbound interface.
  • the ARP requesting module 504 is specifically configured to update the entry from the first port and all non-conjugated ports of the network device when the first port of the network device is determined to be a conjugate port. All ARP entries in the VLANs identified by the VLAN information are all ARP entries with the non-conjugated port as the outbound interface.
  • the network device provided in this embodiment divides all ports related to multiple dual uplink tangential rings on the network device into a conjugate port and a non-conjugate port; the network device receives the entry update message from the first port, and determines Whether the first port is a conjugate port, and when determining that the first port is a conjugate port, clearing, in the MAC table and the ARP table of the network device, the MAC table that uses the first port as an outgoing interface Item and ARP entry, clear all MAC entries with the non-conjugated port as the outbound interface, and retain all MAC entries and ARP entries that are outbound interfaces of all the conjugate ports, and the first port and all the non-
  • the conjugate port performs an ARP request on the ARP entry that is the outbound interface of the non-conjugated port to refresh the related entries of the dual uplink tangential ring of the first port.
  • the interface discards the update packet of the entry and does not process it. This can prevent the fault of a dual uplink tangential ring from causing the refresh/clear of other non-faulty dual-uplink tangential ring entries, causing traffic interruption, which can solve the problem of large number of entries being updated when the user volume is large. Time business interruption problem.
  • the embodiment of the present invention further provides a network device. As shown in FIG. 7, the network device includes: a processor 701, a memory 702, a communication interface 703, and a bus 704. The processor 701, the memory 702, and the communication interface 703 pass through the bus 704. Interconnected; the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 7, but it does not mean that there is only one bus or one type of bus.
  • the memory 702 is configured to store a program.
  • the program can include program code, the program code including computer operating instructions.
  • the communication interface 703 is used to connect to other devices and communicate with other devices; the communication interface 703 includes at least a first port.
  • the processor includes a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), and an application specific integrated circuit (application special integrated circuit).
  • CPU central processing unit
  • NP network processor
  • DSP digital signal processor
  • application special integrated circuit application specific integrated circuit
  • ASIC ready-made programmable gate array
  • FPGA microprocessor
  • memory includes high-speed random access memory (RAM) memory, non-volatile memory (non-volatile Memory (NVM for short), such as electrically erasable and programmable read only memory (EEPROM), flash (Flash), etc.
  • the processor 701 executes a program stored in the memory 702, and is configured to receive an entry update message from the first port, determine whether the first port is a conjugate port, and determine that the first port is a conjugate port.
  • MAC table and the ARP table of the network device All the MAC entries and the ARP entries that use the first port as the outbound interface are cleared, and all MAC entries that use the non-conjugated port of the network device as the outgoing interface are cleared.
  • the processor 701 is further configured to: when determining that the first port is a conjugate port, perform, on the first port and all non-conjugate ports, all ARP entries that are outbound interfaces that are non-conjugated ports.
  • the ARP request is used to refresh the related entries of the dual uplink tangential ring where the first port is located.
  • the processor 701 is further configured to discard the entry update message when the first port is a non-conjugate port. In this way, if the first port is a non-conjugated port, the network device does not process the update packet of the entry.
  • the entry update message includes a first control virtual local area network identifier.
  • the processor 701 is specifically configured to obtain the Determining, by the first control, a virtual local area network identifier, and determining whether the first control virtual local area network identifier is the same as the identifier of the control virtual local area network configured on the first port; and when the first control virtual local area network identifier is the first When the identifier of the control virtual local area network configured on the port is the same, it is determined whether the first port is a conjugate port, and when it is determined that the first port is a conjugate port, in the MAC table and the ARP table of the network device, All the MAC entries and the ARP entries that are outbound interfaces of the first port are cleared, and all the MAC entries that are the outbound interfaces of the non-conjugated port are cleared.
  • the yoke port performs an ARP request for all the ARP entries with the non-con
  • the entry update message further includes a VLAN information, which may be a VLAN identifier.
  • VLAN information may be a VLAN identifier.
  • the extended Flush packet of the SmartLink protocol carries one or more service VLANs to be refreshed through the VLAN bitmap field. logo. If the entry update message includes the VLAN information, when it is determined that the first port is a conjugate port, the first port is in all the VLANs identified by the VLAN information in the entry update message. The related entries of the dual uplink tangent ring are cleared/refreshed.
  • the processor 701 is specifically configured to: when determining that the first port of the first upstream network device is a conjugate port, in the first upstream network In the MAC address table and the ARP table of the device, in the MAC entry and the ARP entry in all the VLANs identified by the VLAN information in the entry update packet, all the first ports of the first upstream network device are cleared. All the MAC entries of the outbound interface of the first upstream network device are removed from the MAC address entry and the ARP entry of the outbound interface; All non-conjugated ports perform ARP requests for all ARP entries with the non-conjugated port as the outbound interface in all the VLANs identified by the VLAN information in the table.
  • the processor 701 may be further configured to: copy the entry update message to the second network device; specifically, the processor 701 copies the entry update message, and updates the copied entry. And forwarding from the second port of the network device to the second network device by using a link between the network device and the second network device, that is, a common tangent edge of the dual uplink tangential ring.
  • the communication interface 703 can also include a second port.
  • the second network device receives, from the second port of the second network device, the entry update message that is copied and forwarded by the network device, because the second port of the second network device
  • the port of the common tangent edge of the dual uplink tangent ring on the second network device is a non-conjugated port, so the second network device discards the entry update message.
  • the network device provided in this embodiment is configured to divide all ports related to multiple dual uplink tangential rings on the network device into conjugate ports and non-conjugate ports, and the processor receives the entry update message from the first port. Determining whether the first port is a conjugate port, and when determining that the first port is a conjugate port, clearing, by using the first port as an outgoing interface, in a MAC table and an ARP table of the network device.
  • the MAC address entry and the ARP entry clear all the MAC entries that are outbound interfaces with the non-conjugated port, and the MAC entries and ARP entries that are the outbound interfaces of all the conjugate ports.
  • All the non-conjugated ports perform an ARP request on the ARP entry that is the outbound interface of the non-conjugated port to refresh the related entries of the dual uplink tangential ring of the first port.
  • the conjugate port discards the entry update packet and does not process it. In this way, the fault of a double uplink tangential ring can be avoided, causing the refresh/clear of other non-faulty dual uplink tangential ring entries, which causes the traffic to be interrupted, which can solve the problem that a large number of entries are updated when the user volume is large. Long-term business interruption problems.
  • An embodiment of the present invention provides another network device. Referring to FIG. 8, the network device includes:
  • a state sensing module 801 configured to detect a state change of the first port of the network device, where the state of the first port of the network device is changed from a blocking state to a forwarding state;
  • a determining module 802 configured to determine whether the first port of the network device is a conjugate port
  • An entry update module 803, configured to: when determining that the first port of the network device is a conjugate port, clearing, in the MAC table and the ARP table of the network device, all the devices that are the first network device The first port is the MAC address entry and ARP table of the outbound interface. Item, clearing all MAC entries that use the non-conjugated port of the first network device as the outbound interface;
  • the network device may further include:
  • the ARP requesting module 804 is configured to, when determining that the first port of the network device is a conjugate port, from the first port and all non-conjugate ports of the network device, to all the network devices
  • the non-conjugated port performs an ARP request for the ARP entry of the outbound interface.
  • the network device further includes:
  • a message generation and sending module 805, configured to generate a first entry update message, for example, an extended TC message, and send the first entry update message to the second network device; 805.
  • the device is configured to generate the first entry update message, and send the generated first entry update message from the second port of the network device, by using the network device and the second network.
  • the link to which the device is connected that is, the common tangent edge of the dual uplink tangential ring, is sent to the second network device.
  • the second network device receives the first entry update message from the second port of the second network device, where the second port of the second network device is the second network
  • the port of the common tangent edge of the double uplink tangential ring on the device is a non-conjugated port, so the second network device discards the first entry update message.
  • the packet generation and sending module 805 is further configured to send the first entry update message to the downstream network device, so that the downstream network device clears/refrelates the related entry;
  • the sending module 805 is specifically configured to send the first entry update message from the first port of the network device to the downstream network device.
  • the network device may further include:
  • the receiving module 806 is configured to receive a second entry update message from the second port of the network device.
  • the second network device senses a state change of the first port of the second network device, and changes from a blocking state to a forwarding state, where the second network device generates the second entry update message, and generates
  • the second entry update message is sent from the second port of the second network device to the link connected to the second network device by the network device, that is, the common tangent edge of the dual uplink tangent ring Sending to the second network device; the network device receiving the second entry update message from the second port of the network device.
  • the determining module 802 is further configured to determine whether the second port of the network device is a conjugate port.
  • the discarding module 807 is configured to discard the second entry update message when the second port of the network device is determined to be a non-conjugated port.
  • the network device discards the second entry update message due to the first yoke port of the network device.
  • the network device provided in this embodiment divides all ports related to multiple dual uplink tangential rings on the network device into a conjugate port and a non-conjugate port, and the network device senses that the state of the first port changes to a forwarding state, and determines Whether the first port is a conjugate port, and when determining that the first port is a conjugate port, clearing, in the MAC table and the ARP table of the network device, the MAC table that uses the first port as an outgoing interface Item and ARP entry, clear all MAC entries with the non-conjugated port as the outbound interface, and retain all MAC entries and ARP entries that are outbound interfaces of all the conjugate ports, and the first port and all the non-
  • the conjugate port performs an ARP request on the ARP entry that is the outbound interface of the non-conjugated port to refresh the related entries of the dual uplink tangential ring of the first port; if the first port is determined to be non-conjugated
  • the port discard
  • the embodiment of the present invention further provides a network device.
  • the network device includes: a processor 1001, a memory 1002, a communication interface 1003, and a bus 1004.
  • the processor 1001, the memory 1002, and the communication interface 1003 pass through the bus 1004.
  • Interconnected; the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 8, but it does not mean that there is only one bus or one type of bus.
  • the memory 1002 is configured to store a program.
  • the program can include program code, the program code including computer operating instructions.
  • the communication interface 1003 is configured to connect to other devices and communicate with other devices; the communication interface 1003 includes at least a first port.
  • the processor includes a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), and an application specific integrated circuit (application special integrated circuit).
  • CPU central processing unit
  • NP network processor
  • DSP digital signal processor
  • application special integrated circuit application specific integrated circuit
  • ASIC ready-made programmable gate array
  • FPGA microprocessor
  • memory includes high-speed random access memory (RAM) memory, non-volatile memory (non-volatile Memory (NVM for short), such as electrically erasable and programmable read only memory (EEPROM), flash (Flash), etc.
  • the processor 1001 executes a program stored in the memory 1002 for sensing a state change of the first port, determining whether the first port is a conjugate port, and when determining that the first port is a conjugate port, In the MAC address table and the ARP table of the network device, all MAC entries and ARP entries that use the first port as the outgoing interface are cleared, and all MAC entries that are outbound interfaces of the network device are removed. .
  • the processor 1001 is further configured to: when determining that the first port is a conjugate port, The first port and all the non-conjugate ports are configured to perform an ARP request for the ARP entry that is the outbound interface of the first port to refresh the related entries of the dual uplink tangential ring.
  • the processor 1001 is further configured to: when the first port is a non-conjugate port, do not process the state change of the first port.
  • the processor 1001 may be further configured to generate a first entry update message, and send the first entry update message to the second network device. Specifically, the processor 1001 generates the An entry update message, such as an extended TC message, and the generated first entry update message is connected from the second port of the network device to the second network device by using the network device The link, that is, the common tangent edge of the dual uplink tangent ring, is sent to the second network device.
  • An entry update message such as an extended TC message
  • the second port may also be included in the communication interface 1003.
  • the processor 1001 is further configured to send the first entry update message to a downstream network device, so that the downstream network device clears/refreshes related entries; the processor 1001 is specifically configured to use the The first port sends the first entry update message to the downstream network device.
  • the processor 1001 is further configured to receive, by the second port, a second entry update message, to determine whether the second port is a conjugate port, and when determining that the second port is non-co- When the port is yoke, the second entry update message is discarded.
  • the network device provided in this embodiment is configured to divide all ports related to multiple dual uplink tangential rings on the network device into a conjugate port and a non-conjugate port, and the processor senses that the state of the first port changes to a forwarding state, and determines Whether the first port is a conjugate port, when determining the The first port is a conjugate port.
  • the MAC table and the ARP table of the network device the MAC address entry and the ARP entry that are the outbound interface of the first port are cleared, and all non-conjugated ports are cleared.
  • the item performs an ARP request to refresh the related entry of the dual uplink tangential ring of the first port; if the first port is determined to be a non-conjugated port, the entry update packet is discarded, and no processing is performed. .
  • the fault of a double uplink tangential ring can be avoided, causing the refresh/clear of other non-faulty dual uplink tangential ring entries, which causes the traffic to be interrupted, which can solve the problem that a large number of entries are updated when the user volume is large. Long-term business interruption problems.
  • the embodiment of the present invention further provides a dual uplink tangential ring system, where the system includes a first upstream network device, a second upstream network device, and one or more downstream network devices, as shown in FIG.
  • One or more downstream network devices are dual-homed to the first upstream network device and the second upstream network device; the link between the first upstream network device and the second upstream network device constitutes one or more dual uplink tangent rings
  • the first upstream network device and the second upstream network, the first upstream network device and the second upstream network device, the ports associated with the multiple dual uplink tangent rings are divided into a conjugate port and a non-conjugate port;
  • the first upstream network device and the second upstream network device may respectively maintain respective sets of conjugate ports.
  • the first upstream network device is configured to receive a first entry update message from the first port of the first upstream network device, or to detect that a state of the first port of the first upstream network device changes; Whether the first port of the first upstream network device is a conjugate port, and when determining that the first port of the first upstream network device is a conjugate port, the MAC of the first upstream network device In the table and ARP table, clear all Deleting, by the first port of the first upstream network device, a MAC entry and an ARP entry of the outbound interface, and deleting all the MAC entries that are the outbound interface of the first upstream network device; Performing an ARP request on the ARP entry of the first upstream network device that is the outbound interface of the first upstream network device, and the non-conjugated port of the first upstream network device, to refresh the If the first port of the first upstream network device is a non-conjugate port, the first entry update packet is not processed.
  • the first entry update message is
  • the first upstream network device may be further configured to: after receiving the first entry update message from the first port of the first upstream network device, copy the first entry update message and forward the packet to the second
  • the upstream network device is configured to: change the state of the first port of the first upstream network device, and generate a second entry update message, and send the message to the second upstream network device; An update message or the second entry update message, from the second port of the first upstream network device, the link that the first upstream network device connects to the second upstream network device, that is, the dual The common tangent edge of the uplink tangential ring is forwarded to the second upstream network device.
  • the second upstream network device is configured to receive the first entry update message or the second entry update message from a second port of the second upstream network device; determine the second upstream network Whether the second port of the device is a conjugate port, and when the second port of the second upstream network device is determined to be a non-conjugate port, discarding the first entry update message or the second entry update Message. Specifically, the second upstream network device discards the first entry update message or the second entry update, because the first port of the second upstream network device is a non-conjugate port. Message.
  • the first upstream network device is configured as a backup scenario.
  • the first downstream network device detects that the primary link of the first downstream network device to the second upstream network device fails through the broken protocol,
  • the first downstream network device is configured to upgrade the standby link of the first downstream network device to the first upstream network device, that is, change the standby uplink port of the first downstream network device from a blocking state. And transmitting, by the first uplink device of the first downstream network device, the first entry update message to the first port of the first upstream network device.
  • the method and the principle of the process of the first downstream network device to the second upstream network device may refer to the method embodiment of the present invention, and details are not described herein.
  • first upstream network device and the second upstream network device may be as shown in any embodiment of FIG. 5 to FIG. 10, and details are not described herein again.
  • the dual uplink tangential ring system provided in this embodiment divides all ports related to multiple dual uplink tangential rings on the first upstream network device and the second upstream network device into a conjugate port and a non-conjugate port.
  • the upstream network device receives the entry update message from the first port, or detects the state change of the first port, and the first upstream network device determines whether the first port is a conjugate port, and when determining that the first port is In the MAC address table and the ARP table of the first upstream network device, all the MAC entries and ARP entries that use the first port as the outgoing interface are cleared, and all non-conjugated ports are used as the outgoing interface.
  • the ARP entry that is the outbound interface of the non-conjugated port on the first port and all the non-conjugated ports.
  • the ARP request is performed to refresh the related entries of the dual uplink tangent ring where the first port is located. Therefore, the fault of one dual uplink tangential ring may be caused to cause the refresh/purge of the other non-faulty dual uplink tangential ring entries, which may cause the traffic to be interrupted, and the large number of entries may be updated when the user volume is large. Time business interruption problem.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Small-Scale Networks (AREA)

Abstract

本发明实施例公开了一种双上行相切环收敛的方法,设备和系统,通过将网络设备上的端口分为共轭端口和非共轭端口,当网络设备从第一端口接收到表项更新报文,或者感知到第一端口的状态变化时,若确定所述第一网络设备的所述第一端口为共轭端口,在自身的MAC表和ARP表中,清除以所述第一端口为出接口的MAC表项和ARP表项,清除所有以非共轭端口为出接口的MAC表项,从而可以避免一个双上行相切环的故障引发其他无故障的双上行相切环的表项的刷新/清除,导致流量中断的问题。

Description

双上行相切环收敛的方法、 设备和系统 技术领域
本发明实施例涉及计算机通信领域,尤其涉及双上行相切环收敛 的方法、 设备和系统。 背景技术
当下游设备连接到上游设备时,使用单上行方式容易出现单点故 障, 造成业务中断。 因此通常釆用双上行方式, 即, 一台下游设备同 时连接到两台上游设备, 以避免单点故障, 提高网络可靠性。 双上行 方式虽然能提高网络可靠性, 但是又引入了环路问题, 需要通过破环 协议, 例如生成树协议( spanning tree protocol, 简称 STP ) /多生成树 协议 (multiple spanning tree protocol , 简称 MSTP)/快速生成树协议 (rapid spanning tree protocol , 简称 MSTP) , 快速环网保护协议 ( rapid ring protection protocol, 简称 RRPP ) , 灵活链路 ( SmartLink )协议 等来消除环路。
但是,在多个下游网络设备双上行归属到一对相同的上游网络设 备, 与此对上游网络设备之间的链路构成多个双上行相切环的组网, 且虚拟局域网( virtual local area network , 简称为 VLAN )釆用每业务 每 VLAN ( er service per VLAN, 简称 PSPV )的方式部署时, 多个双 上行相切环的所有链路共享一个业务 VLAN, 虽然部署破环协议可以 防止流量在广播域内成环, 但是现有破环协议的操作, 是以 VLAN为 粒度进行相关表项的清除 /刷新操作的。 因此, 任何一个双上行相切 环的故障, 会引发其他无故障的双上行相切环的相关表项的清除 /刷 新, 从而导致流量中断; 当用户量大时, 会引发大量的表项更新, 从 而导致长时间的业务中断。 发明内容
有鉴于此, 本发明实施例提供了一种双上行相切环收敛的方法、 设备和系统,可以解决一个双上行相切环的故障引发其他无故障的双 上行相切环的表项的刷新 /清除, 导致流量中断的问题, 进而可以解 决用户量大时, 大量表项更新, 导致长时间业务中断的问题。
第一方面, 本发明提供了一种双上行相切环收敛的方法, 一个或 多个下游设备双上行归属到第一网络设备和第二网络设备,与所述第 一网络设备和所述第二网络设备之间的链路构成一个或多个双上行 相切环, 所述方法包括:
所述第一网络设备从所述第一网络设备的第一端口接收到表项 更新报文, 或者感知到所述第一网络设备的所述第一端口状态变化; 若确定所述第一网络设备的所述第一端口为共轭端口,则在所述 第一网络设备的介质访问控制 MAC表和地址解析协议 ARP表中, 清除以所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表项;
其中, 共轭端口为所述第一网络设备的接入侧端口。
结合第一方面, 在第一方面的第一种可能实现方式中, 若确定所 述第一网络设备的所述第一端口为共轭端口,在清除所述第一网络设 备的介质访问控制 MAC表和地址解析协议 ARP表中以所述第一网 络设备的所述第一端口为出接口的表项, 清除所述第一网络设备的 MAC表中所有以所述第一网络设备的非共轭端口为出接口的表项之 后, 还包括:
从所述第一网络设备的所述第一端口和所有非共轭端口,对所有 以所述第一网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
结合第一方面的第一种可能实现方式,在第一方面的第二种可能 实现方式中, 所述表项更新报文中包括虚拟局域网 VLAN信息; 相 应地, 所述在所述第一网络设备的介质访问控制 MAC表和地址解析 协议 ARP表中, 清除以所述第一网络设备的所述第一端口为出接口 的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭 端口为出接口的 MAC表项, 具体为:
在所述第一网络设备的介质访问控制 MAC表和地址解析协议 ARP表中,所述 VLAN信息标识的所有 VLAN内的 MAC表项和 ARP 表项中, 清除以所述第一网络设备的所述第一端口为出接口的 MAC 表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出 接口的 MAC表项;
所述从所述第一网络设备的所述第一端口和所有非共轭端口,对 所有以所述第一网络设备的非共轭端口为出接口的 ARP表项进行 ARP请求, 具体为:
从所述第一网络设备的所述第一端口和所有非共轭端口, 对 VLAN信息标识的所有 VLAN内的, 所有以非共轭端口为出接口的 ARP表项进行 ARP请求。
结合第一方面和第一方面的第一种和第二种可能实现方式中的 任一种, 在第一方面的第三种可能实现方式中, 所述表项更新报文为 灵活链路 SmartLink协议的扩展 Flush报文, 所述扩展 Flush报文中 包括第一控制虚拟局域网标识;在确定所述第一网络设备的所述第一 端口为共轭端口之前, 所述方法还包括:
所述第一网络设备获取所述扩展 Flush报文中的第一控制虚拟局 域网标识 ,判断确定所述第一控制虚拟局域网标识与所述第一网络设 备的所述第一端口上配置的控制虚拟局域网的标识相同。
结合第一方面和第一方面的第一种至第三种可能实现方式中的 任一种, 在第一方面的第四种可能实现方式中, 还包括: 所述第一网 络设备复制并转发所述表项更新报文至所述第二网络设备的第二端 口。
结合第一方面和第一方面的第一种至第四种可能实现方式中的 任一种, 在第一方面的第五种可能实现方式中, 还包括: 若确定所述 第一网络设备的所述第一端口为非共轭端口,则丟弃所述表项更新报 文。
第二方面, 本发明提供了一种网络设备, 包括:
接收模块, 用于从所述网络设备的第一端口接收表项更新报文; 判断模块,用于判断所述网络设备的所述第一端口是否为共轭端 口;
表项更新模块,用于当确定所述网络设备的所述第一端口为共轭 端口时, 在所述网络设备的介质访问控制 MAC 表和地址解析协议 ARP表中, 清除所有以所述第一网络设备的所述第一端口为出接口 的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭 端口为出接口的 MAC表项;
其中, 共轭端口为所述第一网络设备的接入侧端口。
结合第二方面, 在第二方面的第一种可能实现方式中, 还包括: ARP请求模块, 用于当确定所述网络设备的所述第一端口为共 轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对所 有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
结合第二方面的第一种可能实现方式,在第二方面的第二种可能 实现方式中, 所述表项更新报文中包括虚拟局域网 VLAN信息; 所述表项更新模块,具体用于当确定所述网络设备的所述第一端 口为共轭端口时,在所述网络设备的 MAC表和 ARP表中,所述 VLAN 信息标识的所有 VLAN内的 MAC表项和 ARP表项中, 清除所有以 所述网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述网络设备的非共轭端口为出接口的 MAC表项; 所述 ARP请求模块, 具体用于当确定所述网络设备的所述第一 端口为共轭端口时,从所述网络设备的所述第一端口和所有非共轭端 口, 对所述 VLAN信息标识的所有 VLAN内的, 所有以非共轭端口 为出接口的 ARP表项进行 ARP请求。
结合第二方面和第一方面的第一种至第二种可能实现方式中的 任一种, 在第二方面的第三种可能实现方式中, 所述表项更新报文为 灵活链路 SmartLink协议的扩展 Flush报文, 所述扩展 Flush报文中 包括第一控制虚拟局域网标识, 所述网络设备还包括:
第二判断模块,用于获取所述表项更新报文中的所述第一控制虚 拟局域网标识,并判断所述第一控制虚拟局域网标识与所述第一端口 上配置的控制虚拟局域网的标识是否相同;
相应地, 所述第一判断模块具体用于, 当所述第二判断模块的判 断结果为所述第一控制虚拟局域网标识与所述第一端口上配置的控 制虚拟局域网的标识相同时,判断所述网络设备的所述第一端口是否 为共轭端口。
结合第二方面和第一方面的第一种至第三种可能实现方式中的 任一种, 在第二方面的第四种可能实现方式中, 还包括:
复制转发模块,用于复制所述表项更新报文并转发至第二网络设 备; 丟弃模块 505 , 用于当确定所述网络设备的所述第一端口为非共 轭端口时, 丟弃所述表项更新报文。 第三方面, 本发明提供了一种网络设备, 包括:
状态感知模块, 用于感知所述网络设备的第一端口的状态变化; 判断模块,用于判断所述网络设备的所述第一端口是否为共轭端 口;
表项更新模块,用于当确定所述网络设备的所述第一端口为共轭 端口时, 在所述网络设备的 MAC表和 ARP表中, 清除所有以所述 第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表项; 其中, 共轭端口为所述第一网络设备的接入侧端口。
结合第三方面, 在第三方面的第一种可能实现方式中, 还包括: ARP请求模块, 用于当确定所述网络设备的所述第一端口为共 轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对所 有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
结合第三方面或第三方面的第一种可能实现方式,在第三方面的 第二种可能实现方式中, 还包括:
报文生成发送模块, 用于生成第一表项更新报文, 并发送所述第 一表项更新报文至所述第二网络设备;
结合第三方面和第三方面的第一种至第二种可能实现方式中的 任一种, 在第三方面的第三种可能实现方式中, 还包括接收模块, 用 于从所述网络设备的第二端口接收第二表项更新报文;
相应地, 所述判断模块, 还用于判断所述网络设备的所述第二端 口是否为共轭端口。 第四方面, 本发明提供了一种双上行相切环系统, 包括第一上游 网络设备, 第二上游网络设备, 以及一个或多个下游网络设备; 其中,一个或多个下游网络设备双上行归属到第一上游网络设备 和第二上游网络设备,与第一上游网络设备和第二上游网络设之间的 链路构成一个或多个双上行相切环;
所述第一上游网络设备和所述第二上游网络设备上所述一个或 多个双上行相切环相关的端口中的接入侧端口为共轭端口;
所述第一上游网络设备,用于从所述第一上游网络设备的第一端 口接收第一表项更新报文,或者感知所述第一上游网络设备的第一端 口的状态发生变化;判断所述第一上游网络设备的所述第一端口是否 为共轭端口, 当确定所述第一上游网络设备的所述第一端口为共轭端 口时, 在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有 以所述第一上游网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一上游网络设备的非共轭端口为出接 口的 MAC表项; 并在所述第一上游网络设备的所述第一端口和所有 非共轭端口,对所有以所述第一上游网络设备的非共轭端口为出接口 的 ARP表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切 环的相关表项。
结合第四方面, 在第四方面的第一种可能实现方式中, 所述第一上游网络设备,还用于在从所述第一上游网络设备的第 一端口接收到第一表项更新报文之后,复制所述第一表项更新报文并 转发至第二上游网络设备; 或者, 在感知所述第一上游网络设备的第 一端口的状态发生变化, 生成第二表项更新报文, 并发送至第二上游 网络设备;
所述第二上游网络设备,用于从所述第二上游网络设备的第二端 口接收所述第一表项更新报文或所述第二表项更新报文;判断所述第 二上游网络设备的第二端口是否为共轭端口。
结合第四方面或第四方面的第一种可能实现方式,在第四方面的 第二种可能实现方式中,
所述第一下游网络设备,用于向所述第一上游网络设备的所述第 一端口发送所述第一表项更新报文。
釆用本发明实施例提供的技术方案,将网络设备上与多个双上行 相切环相关的所有端口分为共轭端口和非共轭端口,网络设备从第一 端口接收到表项更新报文, 或者感知到第一端口的状态变化, 第一网 络设备判断所述第一端口是否为共轭端口, 当确定所述第一端口为共 轭端口时, 在网络设备的 MAC表和 ARP表中, 清除所有以所述第 一端口为出接口的 MAC表项和 ARP表项, 清除所有以非共轭端口 为出接口的 MAC表项, 保留以其他所有共轭端口为出接口的 MAC 表项和 ARP表项; 并在所述第一端口和所有非共轭端口对所有以非 共轭端口为出接口的 ARP表项进行 ARP请求 , 以刷新所述第一端口 所在的双上行相切环的相关表项。从而可以避免一个双上行相切环的 故障引发其他无故障的双上行相切环的表项的刷新 /清除, 导致流量 中断的问题, 进而可以解决用户量大时, 大量表项更新, 导致长时间 业务中断的问题。
附图说明
图 l a, lb是本发明实施例提供的多个双上行相切环的组网示意; 图 2是本发明实施例提供的一种双上行相切环收敛的方法流程 图;
图 3是本发明实施例提供的另一种双上行相切环收敛的方法流程 图;
图 4是本发明实施例提供的又一种双上行相切环收敛的方法流程 图;
图 5 , 图 6是本发明实施例提供的一种网络设备结构示意图; 图 7是本发明实施例提供的一种网络设备硬件结构示意图; 图 8, 图 9是本发明实施例提供的另一种网络设备结构示意图; 图 10是本发明实施例提供的另一种网络设备硬件结构示意图; 图 11是本发明实施例提供的一种双上行相切环系统示意图。 具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术 方案进行清楚、 完整地描述, 显然, 所描述的实施例仅仅是本发 明的一部分实施例, 而不是全部的实施例。 基于本发明中的实施 例, 本领域普通技术人员在没有做出创造性劳动前提下所获得的 所有其他实施例, 都属于本发明保护的范围。 多个双上行相切环相关的端口, 分为共轭端口和非共轭端口; 其中, 侧端口;非共轭端口表示构成双上行相切环的公共相切边的网络设备 间的端口, 即双上行相切环的公共相切边的端口。
网络设备可以维护共轭端口集合。
若网络设备上有 N个共轭端口, 其中 N为正整数, 则表明有 N 个双上行环接入该网络设备。
参见图 la, 是本发明实施例提供的一种多个双上行相切环的组 网示意图;如图 la所示, 2个下游网络设备——交换机 1和交换机 2, 双上行归属到一对上游网络设备一一路由器 1和路由器 2, 与路由器 1和路由器 2之间的链路构成 2个双上行相切环, 其中路由器 1和路 络设备一一路由器 1和路由器 2作为业务网关,釆用 PSPV的部署方 式, 2个双上行相切环共享一个业务 VLAN, 并且部署了破环协议, 例如 SmartLink、 RRPP、 STP、 RSTP或 MSTP等, 以防止流量在广 播域内成环。
如图 la所示, 路由器 1中端口 11、 12和 15是多个双上行相切 环上的端口, 其中端口 11和 12是接入侧端口, 均为共轭端口, 而端 口 15是双上行相切环的公共相切边的端口, 为非共轭端口; 同样, 路由器 2的端口 21、 22和 25是多个双上行相切环上的端口, 其中端 口 21和 22是接入侧端口 , 均为共轭端口, 而端口 25是双上行相切 环的公共相切边的端口, 为非共轭端口。
参见图 lb, 是本发明实施例提供的另一种多个双上行相切环的 组网示意图; 如图 2所示, 3个下游网络设备——交换机 1、 交换机 2和交换机 3 , 其中交换机 1、 交换机 2双上行归属到一对上游网络 设备一一路由器 1和路由器 2, 与路由器 1和路由器 2之间的链路构 成 2个双上行相切环,其中路由器 1和路由器 2之间的链路构成 2个 双上行相切环的公共相切边;而交换机 3双上行归属到一对上游网络 设备一一路由器 1和路由器 3 , 与路由器 1和路由器 3之间的链路构 成 1个双上行相切环,其中路由器 1和路由器 3之间的链路构成该双 上行相切环的公共相切边; 路由器 1、 路由器 2和路由器 3作为业务 网关,釆用 PSPV的部署方式, 3个双上行相切环共享一个业务 VLAN, 并且部署了破环协议, 以防止流量在广播域内成环。 如图 lb所示, 路由器 1中端口 11、 12、 13、 15和 16是多个双上行相切环上的端口, 其中端口 11、 12和 13是接入侧端口, 均为共轭端口, 而端口 15和 16 是双上行相切环的公共相切边的端口, 为非共轭端口; 同样, 路 由器 2的端口 21、 22和 25是多个双上行相切环上的端口, 其中端口 21和 22是接入侧端口 , 均为共轭端口 , 而端口 25是双上行相切环 的公共相切边的端口, 为非共轭端口; 同样, 路由器 3的端口 33和 36是双上行相切环上的端口, 其中端口 33是接入侧端口, 为共轭端 口, 而端口 36是双上行相切环的公共相切边的端口, 为非共轭端口。
优选地,本发明技术方案用于多个下游网络设备双上行归属到一 对上游网络设备, 部署相同业务 VLAN, 组成同一业务 VLAN内的 多个双上行相切环的场景。 当然, 本领域技术人员可以理解的是, 当 只有一个下游网络设备双上行归属到一对上游网络设备 ,即一个双上 行相切环时, 本发明技术方案同样适用。
本领域技术人员可以理解的是,共轭端口和非共轭端口是为了方 便区分接入侧端口和非接入侧端口, 给出的自定义命名; 当然也可以 作出其他命名, 本发明对此不作限定。 本发明实施例提供了一种双上行相切环收敛的方法, 如图 2 所 示, 所述方法包括:
201、 第一网络设备从所述第一网络设备的第一端口接收到表项 更新报文,或者感知所述第一网络设备的所述第一端口的状态发生变 化;
其中, 所述表项更新报文可以为扩展 Flush报文或扩展拓朴变化 ( topology change, 简称 TC )报文。 具体地, 若部署的破环协议是 SmartLink或 RRPP协议, 则所述表项更新报文为扩展 Flush报文, 例如, 可以扩展 SmartLink协议中 Flush报文的控制类型 (Control Type )或控制版本( Control Version ) 字段得到所述扩展 Flush报文, 本发明对此不作限定; 若部署的破环协议是 RSTP或 MSTP等, 则所 述表项更新报文为扩展 TC报文, 例如, 可以对 TC报文的协议版本 标识 ( Protocol Version identifier )字段进行扩展, 例如用当前协议中 未使用的标识值, 得到所述扩展 TC报文, 本发明对此不作限定。 当 然所述表项更新报文也可以相应破环协议进行自定义。
所述第一网络设备感知所述第一网络设备的所述第一端口的状 态发送变化,具体为所述第一网络设备感知到所述第一网络设备的所 述第一端口的状态由阻塞状态变为转发(forwarding )状态; 本领域 中, 端口的状态通常用阻塞或转发表示, 有时候也可用 down或 up 表示。
202、 所述第一网络设备判断所述第一网络设备的所述第一端口 是否为共轭端口;
203、 若确定所述第一网络设备的所述第一端口为共轭端口, 所 述第一网络设备在所述第一网络设备的介质访问控制 (media access control, 简称 MAC )表和 ( address resolution protocol, 简称 ARP ) 表中, 清除所有以所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项,清除所有以所述第一网络设备的非共轭端口 为出接口的 MAC表项;
进一步地 ,在所述第一网络设备的所述第一端口和所有非共轭端 口, 对所有以所述第一网络设备的非共轭端口为出接口的 ARP表项 进行 ARP请求 , 以刷新所述第一端口所在的双上行相切环的相关表 项; 通过主动发起 ARP请求, 可以加快相关表项的更新, 使得业务 快速恢复。
这样, 所述第一网络设备仅对所述第一网络设备的 MAC表和 ARP表中所述第一端口所在双上行相切环相关的表项进行清除 /刷 新, 而保留所述第一网络设备的 MAC表和 ARP表中以其他所有共 轭端口为出接口的表项, 即, 保留其他双上行相切环相关的表项, 从 清除 /刷新。
若确定所述第一网络设备的所述第一端口为非共轭端口,所述第 一网络设备对所述表项更新报文不做处理,具体地可以丟弃所述表项 更新报文; 或者对所述第一端口的状态变化不做处理。
可选地,所述第一网络设备复制所述表项更新报文并转发至第二 网络设备; 具体地, 所述第一网络设备复制所述表项更新报文, 并将 复制的所述表项更新报文从所述第一网络设备的第二端口 ,通过所述 第一网络设备与所述第二网络设备连接的链路,即双上行相切环的公 共相切边, 转发至所述第二网络设备。 相应地, 所述第二网络设备从 所述第二网络设备的第二端口接收所述第一网络设备复制并转发的 所述表项更新报文,由于所述第二网络设备的所述第二端口是所述第 二网络设备上双上行相切环的公共相切边的端口, 为非共轭端口, 因 此所述第二网络设备丟弃所述表项更新报文。
本实施例提供的双上行相切环收敛的方法,将第一网络设备上与 多个双上行相切环相关的所有端口分为共轭端口和非共轭端口,第一 网络设备从第一端口接收到表项更新报文,或者感知到第一端口的状 态变化, 第一网络设备判断所述第一端口是否为共轭端口, 当确定所 述第一端口为共轭端口时, 在第一网络设备的 MAC表和 ARP表中, 清除所有以所述第一端口为出接口的 MAC表项和 ARP表项, 清除 所有以非共轭端口为出接口的 MAC表项, 保留以其他所有共轭端口 为出接口的 MAC表项和 ARP表项; 并在所述第一端口和所有非共 轭端口对所有以非共轭端口为出接口的 ARP表项进行 ARP请求 , 以 刷新所述第一端口所在的双上行相切环的相关表项。从而可以避免一 个双上行相切环的故障引发其他无故障的双上行相切环的表项的刷 新 /清除, 导致流量中断的问题, 进而可以解决用户量大时, 大量表 项更新, 导致长时间业务中断的问题。 本发明实施例提供了一种双上行相切环收敛的方法, 参见图 3 , 所述方法包括:
301、 第一下游网络设备通过破环协议, 感知到所述第一下游网 络设备到第二上游网络设备的主用链路发生故障,并将所述第一下游 网络设备到第一上游网络设备的备用链路升为主;
所述第一下游网络设备到所述第一上游网络设备的备用链路两 端的端口为所述第一下游网络设备的备用上行端口和所述第一上游 网络设备的第一端口;所述将所述第一下游网络设备到所述第一上游 网络设备的备用链路升为主, 具体包括: 所述第一下游网络设备将所 述第一下游网络设备的备用上行端口从阻塞状态变为转发状态;相应 地, 所述第一上游网络设备的第一端口从被阻塞状态变为转发状态。
具体结合图 la, lb中交换机 1到路由器 1的主用链路发生故障 的场景来说, 交换机 1通过破环协议, 具体以 SmartLink协议为例, 感知到交换机 1到路由器 1的主用链路发生故障,并将交换机 1到路 由器 2的备用链路升为主,即将交换机 1的端口 102从阻塞状态变为 转发状态; 相应地, 路由器 2的端口 21从被阻塞状态变为转发状态。
302、 所述第一下游网络设备向所述第一上游网络设备发送表项 更新 4艮文;
具体地,所述第一下游网络设备在所述第一下游网络设备的备用 上行端口的状态从阻塞状态变为转发状态之后,所述第一下游网络设 备从所述第一下游网络设备的备用上行端口向所述第一上游网络设 备的所述第一端口发送表项更新报文。 所述表项更新报文为扩展 Flush报文或扩展 TC报文; 具体参见 201 , 在此不再赘述。
结合图 la, lb来说, 在交换机 1的端口 102的状态从阻塞状态 变为转发状态之后, 交换机 1向路由器 2发送表项更新报文, 即扩展 Flush报文; 具体地, 交换机 1从交换机 1的端口 102向路由器 2的 端口 21发送扩展 Flush报文。
303、 所述第一上游网络设备从所述第一上游网络设备的第一端 口接收所述表项更新报文;
结合图 la, lb来说, 路由器 2从端口 21接收扩展 Flush报文。
304、 所述第一上游网络设备判断所述第一上游网络设备的所述 第一端口是否为共轭端口;
具体地,所述第一上游网络设备在所述第一上游网络设备的共轭 端口集合内查询是否存在所述第一端口,并根据查询结果判断所述第 一上游网络设备的所述第一端口是否为共轭端口;若在所述第一上游 网络设备的共轭端口集合内查找到所述第一上游网络设备的所述第 一端口, 则确定所述第一上游网络设备的所述第一端口为共轭端口; 若在所述第一上游网络设备的共轭端口集合内没有查找到所述第一 上游网络设备的所述第一端口,则确定所述第一上游网络设备的所述 第一端口为非共轭端口。
若确定所述第一上游网络设备的所述第一端口为共轭端口,则执 行 305;若确定所述第一上游网络设备的所述第一端口为非共轭端口, 则执行 306。
结合图 la, lb来说, 路由器 2判断端口 21是否为共轭端口, 若 端口 21为共轭端口, 则执行 305 , 刷新路由器 2上与交换机 1的双 上行相切环相关的所有表项。 可选地, 若部署的破环协议是 SmartLink协议, 则所述表项更新 报文中包括第一控制虚拟局域网标识; 相应地, 在 304之前, 所述方 法还包括: 304a、 所述第一上游网络设备获取所述表项更新报文中的 所述第一控制虚拟局域网标识,并判断所述第一控制虚拟局域网标识 与所述第一上游网络设备的所述第一端口上配置的控制虚拟局域网 的标识是否相同;若所述第一控制虚拟局域网标识与所述第一端口上 配置的控制虚拟局域网的标识相同, 则继续执行 304; 若不同, 则不 对所述表项更新报文做处理, 可以按照 SmartLink协议直接转发所述 表项更新报文, 本方法处理过程结束。
305、 当确定所述第一上游网络设备的所述第一端口为共轭端口 时, 所述第一上游网络设备在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有以所述第一上游网络设备的所述第一端口为出 接口的 MAC表项和 ARP表项, 清除所有以所述第一上游网络设备 的非共轭端口为出接口的 MAC表项;
进一步地,从所述第一上游网络设备的所述第一端口和所有非共 轭端口 , 对所有以非共轭端口为出接口的 ARP表项进行 ARP请求 , 以刷新所述第一端口所在双上行相切环的相关表项; 通过主动发起 ARP请求, 可以加快相关表项的刷新, 使得业务可以快速恢复。
这样, 仅对所述第一网络设备的 MAC表和 ARP表中所述第一 端口所在双上行相切环相关的表项进行清除 /刷新; 而保留了所述第 一上游网络设备的 MAC表和 ARP表中以其他所有共轭端口为出接 口的表项, 即, 保留了其他双上行相切环相关的表项, 从而使得一个 结合图 la、 lb来说, 路由器 2判断确定端口 21为共轭端口时, 路由器 2在路由器 2的 MAC表和 ARP表中, 清除以端口 21为出接 口的 MAC表项和 ARP表项, 清除所有以非共轭端口, 即端口 25 , 为出接口的 MAC表项, 进一步地从端口 21和端口 25对所有以端口 25为出接口的 ARP表项进行 ARP请求,以刷新交换机 1所在双上行 相切环的相关表项。 这样, 仅会对交换机 1的双上行相切环的相关表 项进行清除 /刷新, 而不会引发交换机 2和交换机 3的双上行相切环 的相关表项的清除 /刷新; 从而避免交换机 1 的双上行相切环的故障 影响到交换机 2和交换机 3的双上行相切环。
可选地, 所述表项更新报文中还包括 VLAN信息, 具体可以是 一个或多个 VLAN标识; 例如, SmartLink协议的扩展 Flush报文中 通过 VLAN位图表 bitmap 字段携带需要刷新的一个或多个业务 VLAN标识。
若所述表项更新报文中包括 VLAN信息, 当确定所述第一上游 网络设备的所述第一端口为共轭端口时, 在所述表项更新 4艮文中 VLAN信息所标识的所有 VLAN内, 对所述第一上游网络设备的所 述第一端口所在的双上行相切环的相关表项进行清除 /刷新操作。
具体地, 若所述表项更新报文中包括 VLAN信息, 当确定所述 第一上游网络设备的所述第一端口为共轭端口时 ,所述第一上游网络 设备在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有以 所述第一上游网络设备的所述第一端口为出接口的 MAC表项和 ARP 表项, 清除所有以所述第一上游网络设备的非共轭端口为出接口的 MAC表项, 具体为: 在所述第一上游网络设备的 MAC表和 ARP表 中, 所述表项更新报文中 VLAN信息标识的所有 VLAN内的 MAC 表项和 ARP表项中, 清除所有以所述第一上游网络设备的所述第一 端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一上游 网络设备的非共轭端口为出接口的 MAC表项; 所述从所述第一上游 网络设备的所述第一端口和所有非共轭端口,对所有以非共轭端口为 出接口的 ARP表项进行 ARP请求, 具体为: 从所述第一上游网络设 备的所述第一端口和所有非共轭端口,对所述表项更新报文中 VLAN 信息标识的所有 VLAN内的, 所有以非共轭端口为出接口的 ARP表 项进行 ARP请求。
进一步可选地,所述第一上游网络设备还可以复制所述表项更新 报文, 并转发至第二上游网络设备; 具体地, 所述第一上游网络设备 复制所述表项更新报文,并将复制的所述表项更新报文从所述第一上 游网络设备的第二端口,通过所述第一上游网络设备与所述第二上游 网络设备连接的链路, 即双上行相切环的公共相切边, 转发至所述第 二上游网络设备。
相应地,所述第二上游网络设备从所述第二上游网络设备的第二 端口接收所述第一上游网络设备复制并转发的所述表项更新报文,由 于所述第二上游网络设备的所述第二端口是所述第二上游网络设备 上双上行相切环的公共相切边的端口, 为非共轭端口, 因此所述第二 上游网络设备丟弃所述表项更新报文。
结合图 la, lb来说, 路由器 2从端口 25, 通过路由器 2与路由 器 1之间的链路, 转发所述表项更新报文至路由器 1 的端口 15; 路 由器 1从端口 15收到所述表项更新报文后,判断确定端口 15为非共 轭端口, 路由器 1对所述表项更新报文不做处理, 丟弃所述表项更新 报文。
306、 若确定所述第一上游网络设备的所述第一端口为非共轭端 口, 所述第一上游网络设备对所述表项更新报文不做处理; 具体地, 所述第一上游网络设备可以丟弃所述表项更新报文;本方法处理过程 结束。 进一步可选地,所述第一上游网络设备通过破环协议感知到所述 第一下游网络设备到第二上游网络设备的主用链路故障恢复后,所述 第一下游网络设备将所述第一下游网络设备到所述第一上游网络设 备的备用链路的端口,即所述第一下游网络设备的所述备用上行端口 的状态从转发状态变为阻塞状态, 相应地, 所述第一上游网络设备的 所述第一端口的状态从转发状态变为被阻塞状态;并向所述第二上游 网络设备发送新的表项更新报文;然后所述第二上游网络设备接收并 处理所述新的项更新 文的处理过程和原理与上述 303 307相似,在 此不再赘述。
本实施例提供的双上行相切环收敛的方法 ,将上游网络设备上与 多个双上行相切环相关的所有端口分为共轭端口和非共轭端口,上游 网络设备从第一端口接收到表项更新报文时,判断所述第一端口是否 为共轭端口, 当确定所述第一端口为共轭端口时, 在上游网络设备的 MAC表和 ARP表中, 清除所有以所述第一端口为出接口的 MAC表 项和 ARP表项, 清除所有以非共轭端口为出接口的 MAC表项, 保 留以其他所有共轭端口为出接口的 MAC表项和 ARP表项, 并在所 述第一端口和所有非共轭端口对所有以非共轭端口为出接口的 ARP 表项进行 ARP请求, 以刷新所述第一端口所在双上行相切环的相关 表项;若确定所述第一端口为非共轭端口,则丟弃所述表项更新报文, 不做处理。 这样, 一个双上行相切拓朴结构发生变化, 所产生的相关 表项的清除 /刷新, 不会引发其他双上行相切环的表项清除 /刷新; 从 环的表项的刷新 /清除, 导致流量中断的问题, 进而可以解决用户量 大时, 大量表项更新, 导致长时间业务中断的问题。 本发明实施例提供了另一种双上行相切环收敛的方法,参见图 4, 所述方法包括:
401、 第一上游网络设备感知所述第一上游网络设备的第一端口 的状态发生变化;
第一下游网络设备双上行归属到所述第一上游网络设备和第二 上游网络设备上,所述第一下游网络设备分别连接到所述第一上游网 络设备的第一端口上和所述第二上游网络设备的第一端口上。所述第 一上游网络设备和所述第二上游网络设备上运行破环协议, 例如 MSTP或 RSTP。 当第一下游网络设备到所述第二上游网络设备的主 用链路发生故障后,第一下游网络设备到所述第一上游网络设备的备 用链路升为主,也即第一上游网络设备的所述第一端口的状态变为转 发状态, 这通常由破环协议, 例如 MSTP或 RSTP完成, 本发明对此 不作扩展。
然后,所述第一上游网络设备感知到所述第一上游网络设备的所 述第一端口的状态发生变化, 具体的, 从阻塞状态变为转发状态。
具体结合图 la中部署 RSTP/MSTP协议的场景为例, 初始配置 如下:路由器 1为根节点,路由器 1的端口 11 , 12和 15为指定端口、 处于转发状态; 路由器 2的端口 21和 22为备用 ( alternate )端口、 处于阻塞状态, 端口 25为根端口、 处于转发状态; 交换机 1的端口 101为根端口、处于转发状态, 端口 102指定端口、处于被阻塞状态; 交换机 2的端口 201为根端口、 处于转发状态, 端口 202指定端口、 处于被阻塞状态。
当交换机 1到路由器 1 的主用链路发生故障, 路由器 2的端口 21从阻塞状态变为转发状态, 交换机 1到路由器 2的备用链路升为 主; 路由器 2感知到端口 21的状态发送变化。
402、 所述第一上游网络设备判断所述第一上游网络设备的所述 第一端口是否为共轭端口;
具体地,所述第一上游网络设备可以在所述第一上游网络设备的 共轭端口集合内查询是否存在所述第一上游网络设备的所述第一端 口,并根据查询结果判断所述第一上游网络设备的所述第一端口是否 为共轭端口;若在所述第一上游网络设备的共轭端口集合内查找到所 述第一上游网络设备的所述第一端口,则确定所述第一上游网络设备 的所述第一端口为共轭端口;若在所述第一上游网络设备的共轭端口 集合内没有查找到所述第一上游网络设备的所述第一端口,则确定所 述第一上游网络设备的所述第一端口为非共轭端口。
若确定所述第一上游网络设备的所述第一端口为共轭端口,则执 行 403;若确定所述第一上游网络设备的所述第一端口为非共轭端口, 则执行 404。
结合 401中图 la场景例子来说,路由器 2判断端口 21是否为共 轭端口, 若确定端口 21为共轭端口, 则执行 403 , 刷新路由器 2上 与交换机 1的双上行相切环相关的所有表项。
403、 当确定所述第一上游网络设备的所述第一端口为共轭端口 时, 所述第一上游网络设备在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有以所述第一上游网络设备的所述第一端口为出 接口的 MAC表项和 ARP表项, 清除所有以所述第一上游网络设备 的非共轭端口为出接口的 MAC表项;
进一步地,从所述第一上游网络设备的所述第一端口和所有非共 轭端口 , 对所有以非共轭端口为出接口的 ARP表项进行 ARP请求 , 以刷新所述第一端口所在双上行相切环的相关表项; 通过主动发起 ARP请求, 可以加快相关表项的刷新, 使得业务可以快速恢复。
这样, 仅对所述第一网络设备的 MAC表和 ARP表中所述第一 端口所在双上行相切环相关的表项进行清除 /刷新; 而保留了所述第 一上游网络设备的 MAC表和 ARP表中以其他所有共轭端口为出接 口的表项, 即, 保留了其他双上行相切环相关的表项, 从而使得一个 结合 401中图 la场景例子来说,路由器 2判断确定端口 21为共 轭端口时, 路由器 2在路由器 2的 MAC表和 ARP表中, 清除以端 口 21为出接口的 MAC表项和 ARP表项, 清除所有以非共轭端口, 即端口 25 , 为出接口的 MAC表项, 进一步地从端口 21和端口 25对 所有以端口 25为出接口的 ARP表项进行 ARP请求,以刷新交换机 1 所在双上行相切环的相关表项。 这样, 路由器 2不会清除所有表项, 而仅对交换机 1 的双上行相切环的相关表项进行清除 /刷新, 从而可 以避免交换机 1的双上行相切环的故障影响到交换机 2的双上行相切 环。
404、 当确定所述第一上游网络设备的所述第一端口为非共轭端 口时,所述第一上游网络设备对所述第一上游网络设备的所述第一端 口的状态变化不做处理。
可选地, 当所述第一上游网络设备感知到所述第一上游网络设备 的第一端口的状态变化后,所述第一上游网络设备可以生成表项更新 报文, 例如扩展 TC报文, 并发送至所述第二上游网络设备。具体地, 所述第一上游网络设备生成所述表项更新报文,并将生成的所述表项 更新报文从所述第一上游网络设备的第二端口,通过所述第一上游网 络设备与所述第二上游网络设备连接的链路,即双上行相切环的公共 相切边, 发送至所述第二上游网络设备。 相应地, 所述第二上游网络 设备从所述第二上游网络设备的第二端口接收所述第一上游网络设 备生成并发送的所述表项更新报文,由于所述第二上游网络设备的所 的端口, 为非共轭端口, 因此所述第二上游网络设备丟弃所述表项更 新报文。 可选地, 所述第一上游网络设备还可以从所述第一上游网络 设备的所述第一端口发送所述表项更新报文至所述第一下游网络设 备, 以使所述第一下游网络设备清除 /刷新相关表项。
结合 401中图 la场景例子来说, 路由器 2生成扩展 TC报文, 并从端口 25 , 通过路由器 2与路由器 1之间的链路, 发送至路由器 1 的端口 15; 路由器 1从端口 15收到所述扩展 TC "^文后, 判断确定 端口 15为非共轭端口, 路由器 1对所述扩展 TC报文不做处理, 丟 弃所述扩展 TC报文。 同时, 路由器 2还可以从端口 21向交换机 1 发送扩展 TC报文, 使交换机 1清除 MAC表项。
进一步可选地 , 当所述第一下游网络设备到所述第二上游网络设 备的主用链路故障恢复后,所述第二上游网络设备的第一端口的状态 变为转发状态, 所述第一上游网络设备的第一端口变为阻塞状态, 这 通常由破环协议, 例如 MSTP或 RSTP完成, 本发明对此不作扩展。 然后 ,所述第二上游网络设备感知到所述第二上游网络设备的第一端 口的状态变化, 即从阻塞状态变为转发状态, 所述第二上游网络设备 判断所述第二上游网络设备的所述第一端口是否为共轭端口。若所述 第二上游网络设备确定所述第二上游网络设备的所述第一端口为共 轭端口, 则清除 /刷新所述第二上游网络设备的所述第一端口所在的 双上行相切环相关的表项, 其处理过程和原理与上述 403相似, 在此 不再赘述; 可选地, 所述第二上游网络设备可以生成表项更新报文, 例如扩展 TC报文, 发送至所述第一上游网络设备; 具体地, 所述第 的第二端口发送所述表项更新报文至所述第一下游网络设备; 相应 地 ,所述第一上游网络设备从所述第一上游网络设备的第二端口接收 所述表项更新报文,由于所述第一上游网络设备的所述第二端口是所 端口, 因此所述第一上游网络设备对所述表项更新报文不做处理, 丟 弃所述表项更新报文。 可选地, 所述第二上游网络设备向所述第二上 游网络设备的所述第一端口发送所述表项更新报文至所述第一下游 网络设备, 以使所述第一下游网络设备清除 /刷新相关表项。 所述第 一上游网络设备仅向变为转发状态的所述第一上游网络设备的所述 第一端口发送所述表项更新报文 ,而不会像现有技术中向所述第一上 游网络设备的所有指定端口发送所述表项更新报文。
结合 401中图 la场景例子来说, 交换机 1到路由器 1的主用链 路恢复, 路由器 2的 alternate端口, 即端口 21重新变为阻塞状态, 路由器 1的端口 11恢复转发状态,路由器 1感知到端口 11的状态变 化, 判断端口 11是否为共轭端口, 当确定端口 11为共轭端口时, 在 路由器 1的 MAC表和 ARP表中, 清除以端口 11为出接口的 MAC 表项和 ARP表项, 清除所有以非共轭端口, 即端口 15 , 为出接口的 MAC表项,进一步地从端口 11和端口 15对所有以端口 15为出接口 的 ARP表项进行 ARP请求。 进一步的, 路由器 1可以生成扩展 TC 报文, 从端口 15 , 通过路由器 2与路由器 1之间的链路, 发送至路 由器 2的端口 25; 路由器 2从端口 25收到所述扩展 TC报文后, 判 断确定端口 25为非共轭端口, 路由器 2对所述扩展 TC报文不做处 理, 丟弃所述扩展 TC报文。 同时, 路由器 1还可以从端口 11向交 换机 1发送扩展 TC报文, 使交换机 1清除 MAC表项。
本实施例提供的双上行相切环收敛的方法 ,将上游网络设备上与 多个双上行相切环相关的所有端口分为共轭端口和非共轭端口,上游 网络设备感知第一端口的状态发生变化,判断所述第一端口是否为共 轭端口,当确定所述第一端口为共轭端口时,在所述网络设备的 MAC 表和 ARP表中, 清除以所述第一端口为出接口的 MAC表项和 ARP 表项, 清除所有以非共轭端口为出接口的 MAC表项, 保留以其他所 有共轭端口为出接口的 MAC表项和 ARP表项, 并在所述第一端口 和所有非共轭端口对所有以非共轭端口为出接口的 ARP表项进行 ARP请求, 以刷新所述第一端口所在双上行相切环的相关表项; 若 确定所述第一端口为非共轭端口, 则丟弃所述表项更新报文, 不做处 理。 这样, 一个双上行相切拓朴结构发生变化, 所产生的相关表项的 清除 /刷新, 不会引发其他双上行相切环的表项清除 /刷新; 从而可以 避免一个双上行相切环的故障引发其他无故障的双上行相切环的表 项的刷新 /清除, 导致流量中断的问题, 进而可以解决用户量大时, 大量表项更新, 导致长时间业务中断的问题。 本发明实施例提供了一种网络设备, 参见图 5, 所述网络设备包 括:
接收模块 501 , 用于从所述网络设备的第一端口接收表项更新报 文;
第一判断模块 502, 用于判断所述网络设备的所述第一端口是否 为共轭端口;
表项更新模块 503 , 用于当确定所述网络设备的所述第一端口为 共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除所有以 所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表 项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表 项。
所述网络设备还可以包括: ARP请求模块 504 ,用于当确定所述网络设备的所述第一端口为 共轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对 所有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
可选地, 如图 6所示, 所述网络设备还包括:
丟弃模块 505 , 用于当确定所述网络设备的所述第一端口为非共 轭端口时, 丟弃所述表项更新 文。 这样, 若所述第一端口为非共轭 端口, 所述网络设备对所述表项更新报文不做处理。
复制转发模块 506, 用于复制所述表项更新报文并转发至第二网 络设备; 所述复制转发模块 506, 具体用于复制所述表项更新报文, 并将复制的所述表项更新报文从所述网络设备的第二端口,通过所述 网络设备与所述第二网络设备连接的链路,即双上行相切环的公共相 切边, 转发至所述第二网络设备。 相应地, 所述第二网络设备从所述 第二网络设备的第二端口接收所述网络设备复制并转发的所述表项 更新报文,由于所述第二网络设备的所述第二端口是所述第二网络设 备上双上行相切环的公共相切边的端口, 为非共轭端口, 因此所述第 二网络设备丟弃所述表项更新报文。
进一步可选地, 若部署的破环协议是 SmartLink协议, 则所述表 项更新报文中包括第一控制虚拟局域网标识; 相应地, 参见图 7, 所 述网络设备还包括第二判断模块 507, 用于获取所述表项更新报文中 的所述第一控制虚拟局域网标识,并判断所述第一控制虚拟局域网标 识与所述第一端口上配置的控制虚拟局域网的标识是否相同; 相应 地, 所述第一判断模块 502具体用于, 当所述第二判断模块 508的判 断结果为所述第一控制虚拟局域网标识与所述第一端口上配置的控 制虚拟局域网的标识相同时,判断所述网络设备的所述第一端口是否 为共轭端口; 所述丟弃模块 505还用于, 当所述第二判断模块 507的 判断结果为所述第一控制虚拟局域网标识与所述第一端口上配置的 控制虚拟局域网的标识不同时, 丟弃所述表项更新 文。
进一步可选地, 所述表项更新报文中还包括 VLAN信息, 具体 可以是一个或多个 VLAN标识; 例如, SmartLink协议的扩展 Flush 报文中通过 VLAN位图表 bitmap字段携带需要刷新的一个或多个业 务 VLAN标识。 相应地, 所述表项更新模块 503 , 具体用于当确定所 述网络设备的所述第一端口为共轭端口时, 在所述网络设备的 MAC 表和 ARP表中, 所述表项更新报文中 VLAN信息标识的所有 VLAN 内的 MAC表项和 ARP表项中, 清除所有以所述网络设备的所述第 一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述网络设 备的非共轭端口为出接口的 MAC表项;
ARP请求模块 504 ,具体用于当确定所述网络设备的所述第一端 口为共轭端口时, 从所述网络设备的所述第一端口和所有非共轭端 口 , 对所述表项更新报文中 VLAN信息标识的所有 VLAN内的, 所 有以非共轭端口为出接口的 ARP表项进行 ARP请求。
本实施例提供的网络设备,将网络设备上与多个双上行相切环相 关的所有端口分为共轭端口和非共轭端口; 网络设备从第一端口接收 到表项更新报文, 判断所述第一端口是否为共轭端口, 当确定所述第 一端口为共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除 以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以非共 轭端口为出接口的 MAC表项, 保留以其他所有共轭端口为出接口的 MAC表项和 ARP表项, 并在所述第一端口和所有非共轭端口对所有 以非共轭端口为出接口的 ARP表项进行 ARP请求, 以刷新所述第一端 口所在双上行相切环的相关表项; 若确定所述第一端口为非共轭端 口, 则丟弃所述表项更新报文, 不做处理。 这样可以避免一个双上行 相切环的故障引发其他无故障的双上行相切环的表项的刷新 /清除, 导致流量中断的问题, 进而可以解决用户量大时, 大量表项更新, 导 致长时间业务中断的问题。 本发明实施例还提供一种网络设备, 所述网络设备如图 7所示, 包括: 处理器 701 , 存储器 702, 通信接口 703和总线 704; 处理器 701、 存储器 702和通信接口 703通过总线 704相互连接; 所述总线 可以分为地址总线、 数据总线、 控制总线等。 为便于表示, 图 7中仅 用一条粗线表示, 但并不表示仅有一根总线或一种类型的总线。
存储器 702, 用于存放程序。 具体地, 程序可以包括程序代码, 所述程序代码包括计算机操作指令。
所述通信接口 703用于连接其他设备, 并与其他设备进行通信; 所述通信接口 703中至少包括第一端口。
其中, 处理器包括中央处理器 ( central processing unit , 简称 CPU ), 网络处理器 (network processor, 简称 NP ), 数字信号处理 器 ( digital Signal processor, 简称 DSP ) , 专用集成电路 ( application special integrated circuit, 简称 ASIC ) , 现成可编程门阵列 (field programmable gate array, 简称 FPGA ) , 微处理器等; 存储器包括高 速随机存取存储器 ( random access memory, 简称 RAM )存储器, 非 易失性存储器 ( non-volatile memory , 简称 NVM ) , 例如电可擦可 编程只读存储器 ( electrically erasable and programmable read only memory, 简称 EEPROM ), 闪存 ( Flash ) 等。
处理器 701执行存储器 702所存放的程序,用于从所述第一端口 接收到表项更新报文, 判断所述第一端口是否为共轭端口, 当确定所 述第一端口为共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除所有以所述第一端口为出接口的 MAC表项和 ARP表项, 清除 所有以所述网络设备的非共轭端口为出接口的 MAC表项。
所述处理器 701 , 还用于当确定所述第一端口为共轭端口时, 在 所述第一端口和所有非共轭端口, 对所有以非共轭端口为出接口的 ARP表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切环 的相关表项。
所述处理器 701还用于当所述第一端口为非共轭端口时,丟弃所 述表项更新报文。 这样, 若所述第一端口为非共轭端口, 所述网络设 备对所述表项更新报文不做处理。
可选地, 若部署的破环协议是 SmartLink协议, 则所述表项更新 报文中包括第一控制虚拟局域网标识; 相应地, 处理器 701具体用于 获取所述表项更新报文中的所述第一控制虚拟局域网标识,并判断所 述第一控制虚拟局域网标识与所述第一端口上配置的控制虚拟局域 网的标识是否相同; 当所述第一控制虚拟局域网标识与所述第一端口 上配置的控制虚拟局域网的标识相同时,判断所述第一端口是否为共 轭端口,当确定所述第一端口为共轭端口时,在所述网络设备的 MAC 表和 ARP表中, 清除所有以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以非共轭端口为出接口的 MAC表项; 在所述网 络设备的所述第一端口和所有非共轭端口,对所有以非共轭端口为出 接口的 ARP表项进行 ARP请求 , 以刷新所述第一端口所在的双上行 相切环的相关表项。
进一步可选地, 所述表项更新报文中还包括 VLAN信息, 具体 可以是 VLAN标识; 例如 , SmartLink协议的扩展 Flush报文中通过 VLAN位图表 bitmap字段携带需要刷新的一个或多个业务 VLAN标 识。 若所述表项更新报文中包括 VLAN信息, 当确定所述第一端口 为共轭端口时, 在所述表项更新报文中 VLAN信息所标识的所有 VLAN内,对所述第一端口所在的双上行相切环的相关表项进行清除 /刷新操作。若所述表项更新报文中包括 VLAN信息,所述处理器 701 具体用于, 当确定所述第一上游网络设备的所述第一端口为共轭端口 时, 在所述第一上游网络设备的 MAC表和 ARP表中, 所述表项更 新报文中 VLAN信息标识的所有 VLAN内的 MAC表项和 ARP表项 中, 清除所有以所述第一上游网络设备的所述第一端口为出接口的 MAC表项和 ARP表项,清除所有以所述第一上游网络设备的非共轭 端口为出接口的 MAC表项; 从所述第一上游网络设备的所述第一端 口和所有非共轭端口, 对所述表项更新 ^艮文中 VLAN信息标识的所 有 VLAN 内的, 所有以非共轭端口为出接口的 ARP表项进行 ARP 请求。
所述处理器 701 还可以用于复制所述表项更新报文并转发至第 二网络设备; 具体地, 处理器 701复制所述表项更新报文, 并将复制 的所述表项更新报文从所述网络设备的第二端口,通过所述网络设备 与所述第二网络设备连接的链路, 即双上行相切环的公共相切边, 转 发至所述第二网络设备。
所述通信接口 703中还可以包括第二端口。
相应地,所述第二网络设备从所述第二网络设备的第二端口接收 所述网络设备复制并转发的所述表项更新报文,由于所述第二网络设 备的所述第二端口是所述第二网络设备上双上行相切环的公共相切 边的端口, 为非共轭端口, 因此所述第二网络设备丟弃所述表项更新 报文。
本领域技术人员可以理解的是,本发明实施例中仅示出了网络设 备中与本发明相关的部分结构, 还可以包括比图示更多的部件, 或者 不同的部件布置。
本实施例提供的网络设备,将所述网络设备上与多个双上行相切 环相关的所有端口分为共轭端口和非共轭端口,处理器从第一端口接 收到表项更新报文, 判断所述第一端口是否为共轭端口, 当确定所述 第一端口为共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有 以非共轭端口为出接口的 MAC表项, 保留以其他所有共轭端口为出 接口的 MAC表项和 ARP表项, 并在所述第一端口和所有非共轭端 口对所有以非共轭端口为出接口的 ARP表项进行 ARP请求 , 以刷新 所述第一端口所在双上行相切环的相关表项;若确定所述第一端口为 非共轭端口, 则丟弃所述表项更新报文, 不做处理。 这样, 可以避免 一个双上行相切环的故障引发其他无故障的双上行相切环的表项的 刷新 /清除, 导致流量中断的问题, 进而可以解决用户量大时, 大量 表项更新, 导致长时间业务中断的问题。 本发明实施例提供了另一种网络设备, 参见图 8, 所述网络设备 包括:
状态感知模块 801 , 用于感知所述网络设备的第一端口的状态变 化;具体用于感知所述网络设备的所述第一端口的状态从阻塞状态变 为转发状态;
判断模块 802, 用于判断所述网络设备的所述第一端口是否为共 轭端口;
表项更新模块 803 , 用于当确定所述网络设备的所述第一端口为 共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除所有以 所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表 项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表 项;
所述网络设备还可以包括:
ARP请求模块 804 ,用于当确定所述网络设备的所述第一端口为 共轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对 所有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
可选地, 如图 9所示, 所述网络设备还包括:
报文生成发送模块 805 , 用于生成第一表项更新报文, 例如扩展 TC报文, 并发送所述第一表项更新报文至所述第二网络设备; 所述 报文生成发送模块 805 , 具体用于生成所述第一表项更新报文, 并将 生成的所述第一表项更新报文从所述网络设备的第二端口,通过所述 网络设备与所述第二网络设备连接的链路,即双上行相切环的公共相 切边, 发送至所述第二网络设备。 相应地, 所述第二网络设备从所述 第二网络设备的第二端口接收所述第一表项更新报文,由于所述第二 网络设备的所述第二端口是所述第二网络设备上双上行相切环的公 共相切边的端口, 为非共轭端口, 因此所述第二网络设备丟弃所述第 一表项更新 4艮文。
可选地, 所述报文生成发送模块 805 , 还用于发送所述第一表项 更新报文至下游网络设备, 以使所述下游网络设备清除 /刷新相关表 项; 所述报文生成发送模块 805 , 具体用于从所述网络设备的所述第 一端口发送所述第一表项更新报文至所述下游网络设备。
进一步地, 所述网络设备还可以包括:
接收模块 806 , 用于从所述网络设备的第二端口接收第二表项更 新才艮文。 具体地,第二网络设备感知所述第二网络设备的第一端口的状态 变化, 从阻塞状态变为转发状态, 所述第二网络设备生成所述第二表 项更新报文,并将生成的所述第二表项更新报文从所述第二网络设备 的第二端口, 通过所述网络设备与所述第二网络设备连接的链路, 即 双上行相切环的公共相切边, 发送至所述第二网络设备; 所述网络设 备从所述网络设备的第二端口接收所述第二表项更新报文。
相应地, 所述判断模块 802, 还用于判断所述网络设备的所述第 二端口是否为共轭端口。
丟弃模块 807 , 用于当确定所述网络设备的所述第二端口为非共 轭端口时, 丟弃所述第二表项更新报文。 由于所述网络设备的所述第 轭端口, 因此所述网络设备丟弃所述第二表项更新报文。
本实施例提供的网络设备,将网络设备上与多个双上行相切环相 关的所有端口分为共轭端口和非共轭端口,网络设备感知到第一端口 的状态变为转发状态, 判断所述第一端口是否为共轭端口, 当确定所 述第一端口为共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清除以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以 非共轭端口为出接口的 MAC表项, 保留以其他所有共轭端口为出接 口的 MAC表项和 ARP表项, 并在所述第一端口和所有非共轭端口对 所有以非共轭端口为出接口的 ARP表项进行 ARP请求, 以刷新所述第 一端口所在双上行相切环的相关表项;若确定所述第一端口为非共轭 端口, 则丟弃所述表项更新报文, 不做处理。 这样可以避免一个双上 行相切环的故障引发其他无故障的双上行相切环的表项的刷新 /清 除,导致流量中断的问题,进而可以解决用户量大时, 大量表项更新, 导致长时间业务中断的问题。 本发明实施例还提供一种网络设备,所述网络设备如图 10所示, 包括: 处理器 1001 , 存储器 1002, 通信接口 1003和总线 1004; 处 理器 1001、 存储器 1002和通信接口 1003通过总线 1004相互连接; 所述总线可以分为地址总线、 数据总线、 控制总线等。 为便于表示, 图 8中仅用一条粗线表示,但并不表示仅有一根总线或一种类型的总 线。
存储器 1002, 用于存放程序。 具体地, 程序可以包括程序代码, 所述程序代码包括计算机操作指令。
所述通信接口 1003用于连接其他设备,并与其他设备进行通信; 所述通信接口 1003中至少包括第一端口。
其中, 处理器包括中央处理器 ( central processing unit , 简称 CPU ), 网络处理器 (network processor, 简称 NP ), 数字信号处理 器 ( digital Signal processor, 简称 DSP ) , 专用集成电路 ( application special integrated circuit, 简称 ASIC ) , 现成可编程门阵列 (field programmable gate array, 简称 FPGA ) , 微处理器等; 存储器包括高 速随机存取存储器 ( random access memory, 简称 RAM )存储器, 非 易失性存储器 ( non-volatile memory , 简称 NVM ) , 例如电可擦可 编程只读存储器 ( electrically erasable and programmable read only memory, 简称 EEPROM ), 闪存 ( Flash ) 等。
处理器 1001执行存储器 1002所存放的程序,用于感知所述第一 端口的状态变化; 判断所述第一端口是否为共轭端口, 当确定所述第 一端口为共轭端口时, 在所述网络设备的 MAC表和 ARP表中, 清 除所有以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所 有以所述网络设备的非共轭端口为出接口的 MAC表项。
处理器 1001 , 还用于当确定所述第一端口为共轭端口时, 在所 述第一端口和所有非共轭端口,对所有以非共轭端口为出接口的 ARP 表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切环的相 关表项。
所述处理器 1001还用于当所述第一端口为非共轭端口时, 不对 所述第一端口的状态变化做处理。
所述处理器 1001 , 还可以用于生成第一表项更新报文, 并发送 所述第一表项更新报文至所述第二网络设备; 具体地, 所述处理器 1001生成所述第一表项更新报文, 例如扩展 TC报文, 并将生成的所 述第一表项更新报文从所述网络设备的第二端口,通过所述网络设备 与所述第二网络设备连接的链路, 即双上行相切环的公共相切边, 发 送至所述第二网络设备。
所述通信接口 1003中还可以包括所述第二端口。
所述处理器 1001 , 还可用于发送所述第一表项更新报文至下游 网络设备, 以使所述下游网络设备清除 /刷新相关表项; 所述处理器 1001 ,具体用于从所述第一端口发送所述第一表项更新报文至所述下 游网络设备。
进一步可选地, 所述处理器 1001还用于从所述第二端口接收第 二表项更新报文, 判断所述第二端口是否为共轭端口, 当确定所述第 二端口为非共轭端口时, 丟弃所述第二表项更新报文。
本领域技术人员可以理解的是,本发明实施例中仅示出了网络设 备中与本发明相关的部分结构, 还可以包括比图示更多的部件, 或者 不同的部件布置。
本实施例提供的网络设备,将网络设备上与多个双上行相切环相 关的所有端口分为共轭端口和非共轭端口,通过处理器感知第一端口 的状态变为转发状态, 判断所述第一端口是否共轭端口, 当确定所述 第一端口为共轭端口, 在所述网络设备的 MAC表和 ARP表中, 清 除以所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以 非共轭端口为出接口的 MAC表项, 保留以其他所有共轭端口为出接 口的 MAC表项和 ARP表项, 并在所述第一端口和所有非共轭端口 对所有以非共轭端口为出接口的 ARP表项进行 ARP请求 , 以刷新所 述第一端口所在双上行相切环的相关表项;若确定所述第一端口为非 共轭端口, 则丟弃所述表项更新报文, 不做处理。 这样, 可以避免一 个双上行相切环的故障引发其他无故障的双上行相切环的表项的刷 新 /清除, 导致流量中断的问题, 进而可以解决用户量大时, 大量表 项更新, 导致长时间业务中断的问题。 本发明实施例还提供了一种双上行相切环的系统,所述系统如图 11所示, 包括第一上游网络设备, 第二上游网络设备, 以及一个或多 个下游网络设备; 其中,
一个或多个下游网络设备双上行归属到第一上游网络设备和第 二上游网络设备;与第一上游网络设备和第二上游网络设之间的链路 构成一个或多个双上行相切环,其中第一上游网络设备和第二上游网 所述第一上游网络设备和所述第二上游网络设备上多个双上行 相切环相关的端口分为共轭端口和非共轭端口;所述第一上游网络设 备和所述第二上游网络设备可以分别维护各自的共轭端口集合。
所述第一上游网络设备,用于从所述第一上游网络设备的第一端 口接收第一表项更新报文,或者感知所述第一上游网络设备的第一端 口的状态发生变化;判断所述第一上游网络设备的所述第一端口是否 为共轭端口, 当确定所述第一上游网络设备的所述第一端口为共轭端 口时, 在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有 以所述第一上游网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一上游网络设备的非共轭端口为出接 口的 MAC表项; 并在所述第一上游网络设备的所述第一端口和所有 非共轭端口,对所有以所述第一上游网络设备的非共轭端口为出接口 的 ARP表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切 环的相关表项;若确定所述第一上游网络设备的所述第一端口为非共 轭端口, 对所述第一表项更新报文不做处理, 例如, 丟弃所述第一表 项更新报文, 或者对所述第一端口的状态变化不做处理。
所述第一上游网络设备,还可用于在从所述第一上游网络设备的 第一端口接收到第一表项更新报文之后,复制所述第一表项更新报文 并转发至第二上游网络设备; 或者, 在感知所述第一上游网络设备的 第一端口的状态发生变化, 生成第二表项更新报文, 并发送至第二上 游网络设备; 具体将所述第一表项更新报文或所述第二表项更新报 文, 从所述第一上游网络设备的第二端口, 通过所述第一上游网络设 备与所述第二上游网络设备连接的链路 ,即双上行相切环的公共相切 边, 转发至所述第二上游网络设备。
所述第二上游网络设备,用于从所述第二上游网络设备的第二端 口接收所述第一表项更新报文或所述第二表项更新报文;判断所述第 二上游网络设备的第二端口是否为共轭端口, 当确定所述第二上游网 络设备的第二端口为非共轭端口时,丟弃所述第一表项更新报文或所 述第二表项更新报文。 具体的, 由于所述第二上游网络设备的所述第 口, 为非共轭端口, 因此所述第二上游网络设备丟弃所述第一表项更 新报文或所述第二表项更新报文。
以所述第一下游网络设备以所述第二上游网络设备为主,以所述 第一上游网络设备为备的场景为例, 当所述第一下游网络设备通过破 环协议感知到所述第一下游网络设备到所述第二上游网络设备的主 用链路发生故障时, 所述第一下游网络设备, 用于将所述第一下游网 络设备到所述第一上游网络设备的备用链路升为主,即将所述第一下 游网络设备的备用上行端口从阻塞状态变为转发状态;并从所述第一 下游网络设备的备用上行端口向所述第一上游网络设备的所述第一 端口发送所述第一表项更新报文。
而当所述第一下游网络设备到所述第二上游网络设备的主用链 路故障恢复后, 其处理过程和原理可以参考本发明方法实施例, 在此 不再赘述。
具体地,所述第一上游网络设备和所述第二上游网络设备可以如 本发明图 5〜图 10任一实施例所示, 在此不再赘述。
本实施例提供的双上行相切环系统 ,将第一上游网络设备和第二 上游网络设备上与多个双上行相切环相关的所有端口分为共轭端口 和非共轭端口, 第一上游网络设备从第一端口接收到表项更新报文, 或者感知到第一端口的状态变化,第一上游网络设备判断所述第一端 口是否为共轭端口, 当确定所述第一端口为共轭端口时, 在第一上游 网络设备的 MAC表和 ARP表中, 清除所有以所述第一端口为出接 口的 MAC表项和 ARP表项,清除所有以非共轭端口为出接口的 MAC 表项, 保留以其他所有共轭端口为出接口的 MAC表项和 ARP表项; 并在所述第一端口和所有非共轭端口对所有以非共轭端口为出接口 的 ARP表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切 环的相关表项。从而可以避免一个双上行相切环的故障引发其他无故 障的双上行相切环的表项的刷新 /清除, 导致流量中断的问题, 进而 可以解决用户量大时, 大量表项更新, 导致长时间业务中断的问题。 以上所述, 仅为本发明较佳的具体实施方式, 但本发明的保护范 围并不局限于此,任何熟悉本技术领域的技术人员在本发明揭露的技 术范围内, 可轻易想到的变化或替换, 都应涵盖在本发明的保护范围 之内。

Claims

权 利 要 求 书
1、 一种双上行相切环收敛的方法, 其特征在于, 一个或多个下 游设备双上行归属到第一网络设备和第二网络设备,与所述第一网络 设备和所述第二网络设备之间的链路构成一个或多个双上行相切环, 所述方法包括:
所述第一网络设备从所述第一网络设备的第一端口接收到表项 更新报文, 或者感知到所述第一网络设备的所述第一端口状态变化; 若确定所述第一网络设备的所述第一端口为共轭端口,则在所述 第一网络设备的介质访问控制 MAC表和地址解析协议 ARP表中, 清除以所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表项;
其中, 共轭端口为所述第一网络设备的接入侧端口。
2、 根据权利要求 1所述的方法, 其特征在于, 若确定所述第一 网络设备的所述第一端口为共轭端口 ,在清除所述第一网络设备的介 质访问控制 MAC表和地址解析协议 ARP表中以所述第一网络设备 的所述第一端口为出接口的表项, 清除所述第一网络设备的 MAC表 中所有以所述第一网络设备的非共轭端口为出接口的表项之后,还包 括:
从所述第一网络设备的所述第一端口和所有非共轭端口,对所有 以所述第一网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
3、 根据权利要求 2所述的方法, 其特征在于, 所述表项更新报 文中包括虚拟局域网 VLAN信息; 相应地,
所述在所述第一网络设备的介质访问控制 MAC表和地址解析协 议 ARP表中, 清除以所述第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项,清除所有以所述第一网络设备的非共轭端口 为出接口的 MAC表项, 具体为:
在所述第一网络设备的介质访问控制 MAC表和地址解析协议 ARP表中,所述 VLAN信息标识的所有 VLAN内的 MAC表项和 ARP 表项中, 清除以所述第一网络设备的所述第一端口为出接口的 MAC 表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出 接口的 MAC表项;
所述从所述第一网络设备的所述第一端口和所有非共轭端口,对 所有以所述第一网络设备的非共轭端口为出接口的 ARP表项进行 ARP请求, 具体为:
从所述第一网络设备的所述第一端口和所有非共轭端口, 对 VLAN信息标识的所有 VLAN内的, 所有以非共轭端口为出接口的 ARP表项进行 ARP请求。
4、 根据权利要求 1至 3任一项所述的方法, 其特征在于, 所述 表项更新报文为灵活链路 SmartLink协议的扩展 Flush报文, 所述扩 展 Flush报文中包括第一控制虚拟局域网标识;
在确定所述第一网络设备的所述第一端口为共轭端口之前 ,所述 方法还包括:
所述第一网络设备获取所述扩展 Flush报文中的第一控制虚拟局 域网标识 ,判断确定所述第一控制虚拟局域网标识与所述第一网络设 备的所述第一端口上配置的控制虚拟局域网的标识相同。
5、 根据权利要求 1至 4任一项所述的方法, 其特征在于, 还包 括:
所述第一网络设备复制并转发所述表项更新报文至所述第二网 络设备的第二端口。
6、 根据权利要求 1至 5任一项所述的方法, 其特征在于, 还包 括:
若确定所述第一网络设备的所述第一端口为非共轭端口,则丟弃 所述表项更新报文。
7、 一种网络设备, 其特征在于, 包括:
接收模块, 用于从所述网络设备的第一端口接收表项更新报文; 判断模块,用于判断所述网络设备的所述第一端口是否为共轭端 口;
表项更新模块,用于当确定所述网络设备的所述第一端口为共轭 端口时, 在所述网络设备的介质访问控制 MAC 表和地址解析协议 ARP表中, 清除所有以所述第一网络设备的所述第一端口为出接口 的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭 端口为出接口的 MAC表项;
其中, 共轭端口为所述第一网络设备的接入侧端口。
8、 根据权利要求 7所述的网络设备, 其特征在于, 还包括: ARP请求模块, 用于当确定所述网络设备的所述第一端口为共 轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对所 有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
9、 根据权利要求 8所述的网络设备, 其特征在于, 所述表项更 新报文中包括虚拟局域网 VLAN信息;
所述表项更新模块,具体用于当确定所述网络设备的所述第一端 口为共轭端口时,在所述网络设备的 MAC表和 ARP表中,所述 VLAN 信息标识的所有 VLAN内的 MAC表项和 ARP表项中, 清除所有以 所述网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述网络设备的非共轭端口为出接口的 MAC表项;
所述 ARP请求模块, 具体用于当确定所述网络设备的所述第一 端口为共轭端口时,从所述网络设备的所述第一端口和所有非共轭端 口, 对所述 VLAN信息标识的所有 VLAN内的, 所有以非共轭端口 为出接口的 ARP表项进行 ARP请求。
10、 根据权利要求 7至 9任一项所述的网络设备, 其特征在于, 所述表项更新报文为灵活链路 SmartLink协议的扩展 Flush报文, 所 述扩展 Flush报文中包括第一控制虚拟局域网标识, 所述网络设备还 包括:
第二判断模块,用于获取所述表项更新报文中的所述第一控制虚 拟局域网标识,并判断所述第一控制虚拟局域网标识与所述第一端口 上配置的控制虚拟局域网的标识是否相同;
相应地, 所述第一判断模块具体用于, 当所述第二判断模块的判 断结果为所述第一控制虚拟局域网标识与所述第一端口上配置的控 制虚拟局域网的标识相同时,判断所述网络设备的所述第一端口是否 为共轭端口。
11、根据权利要求 7至 10任一项所述的网络设备, 其特征在于, 还包括:
复制转发模块,用于复制所述表项更新报文并转发至第二网络设 备;
丟弃模块 505 , 用于当确定所述网络设备的所述第一端口为非共 轭端口时, 丟弃所述表项更新报文。
12、 一种网络设备, 其特征在于, 包括:
状态感知模块, 用于感知所述网络设备的第一端口的状态变化; 判断模块,用于判断所述网络设备的所述第一端口是否为共轭端 口;
表项更新模块,用于当确定所述网络设备的所述第一端口为共轭 端口时, 在所述网络设备的 MAC表和 ARP表中, 清除所有以所述 第一网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一网络设备的非共轭端口为出接口的 MAC表项; 其中, 共轭端口为所述第一网络设备的接入侧端口。
13、 根据权利要求 12所述的网络设备, 其特征在于, 还包括: ARP请求模块, 用于当确定所述网络设备的所述第一端口为共 轭端口时, 从所述网络设备的所述第一端口和所有非共轭端口, 对所 有以所述网络设备的非共轭端口为出接口的 ARP表项进行 ARP请 求。
14、 根据权利要求 12或 13所述的网络设备, 其特征在于, 还包 括:
报文生成发送模块, 用于生成第一表项更新报文, 并发送所述第 一表项更新报文至所述第二网络设备;
15、根据权利要求 12至 14任一项所述的网络设备,其特征在于, 还包括接收模块,用于从所述网络设备的第二端口接收第二表项更新 报文;
相应地, 所述判断模块, 还用于判断所述网络设备的所述第二端 口是否为共轭端口。
16、 一种双上行相切环系统, 其特征在于, 包括第一上游网络设 备, 第二上游网络设备, 以及一个或多个下游网络设备;
其中,一个或多个下游网络设备双上行归属到第一上游网络设备 和第二上游网络设备,与第一上游网络设备和第二上游网络设之间的 链路构成一个或多个双上行相切环;
所述第一上游网络设备和所述第二上游网络设备上所述一个或 多个双上行相切环相关的端口中的接入侧端口为共轭端口;
所述第一上游网络设备,用于从所述第一上游网络设备的第一端 口接收第一表项更新报文,或者感知所述第一上游网络设备的第一端 口的状态发生变化;判断所述第一上游网络设备的所述第一端口是否 为共轭端口, 当确定所述第一上游网络设备的所述第一端口为共轭端 口时, 在所述第一上游网络设备的 MAC表和 ARP表中, 清除所有 以所述第一上游网络设备的所述第一端口为出接口的 MAC表项和 ARP表项, 清除所有以所述第一上游网络设备的非共轭端口为出接 口的 MAC表项; 并在所述第一上游网络设备的所述第一端口和所有 非共轭端口,对所有以所述第一上游网络设备的非共轭端口为出接口 的 ARP表项进行 ARP请求, 以刷新所述第一端口所在的双上行相切 环的相关表项。
17、 根据权利要求 16所述的系统, 其特征在于,
所述第一上游网络设备,还用于在从所述第一上游网络设备的第 一端口接收到第一表项更新报文之后,复制所述第一表项更新报文并 转发至第二上游网络设备; 或者, 在感知所述第一上游网络设备的第 一端口的状态发生变化, 生成第二表项更新报文, 并发送至第二上游 网络设备;
所述第二上游网络设备,用于从所述第二上游网络设备的第二端 口接收所述第一表项更新报文或所述第二表项更新报文;判断所述第 二上游网络设备的第二端口是否为共轭端口。
18、 根据权利要求 16或 17所述的系统, 其特征在于, 所述第一下游网络设备,用于向所述第一上游网络设备的所述第 一端口发送所述第一表项更新报文。
PCT/CN2013/072759 2013-03-15 2013-03-15 双上行相切环收敛的方法、设备和系统 WO2014139175A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
ES13861532.3T ES2586941T3 (es) 2013-03-15 2013-03-15 Método, dispositivo y sistema para una convergencia de anillo tangente a doble enlace ascendente
EP13861532.3A EP2802105B1 (en) 2013-03-15 2013-03-15 Method, device, and system for convergence of dual-uplink tangent ring
PCT/CN2013/072759 WO2014139175A1 (zh) 2013-03-15 2013-03-15 双上行相切环收敛的方法、设备和系统
CN201380000223.1A CN104160667B (zh) 2013-03-15 2013-03-15 双上行相切环收敛的方法、设备和系统
US14/315,541 US9553760B2 (en) 2013-03-15 2014-06-26 Method, device, and system for dual-uplink tangent ring convergence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2013/072759 WO2014139175A1 (zh) 2013-03-15 2013-03-15 双上行相切环收敛的方法、设备和系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/315,541 Continuation US9553760B2 (en) 2013-03-15 2014-06-26 Method, device, and system for dual-uplink tangent ring convergence

Publications (1)

Publication Number Publication Date
WO2014139175A1 true WO2014139175A1 (zh) 2014-09-18

Family

ID=51535852

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/072759 WO2014139175A1 (zh) 2013-03-15 2013-03-15 双上行相切环收敛的方法、设备和系统

Country Status (5)

Country Link
US (1) US9553760B2 (zh)
EP (1) EP2802105B1 (zh)
CN (1) CN104160667B (zh)
ES (1) ES2586941T3 (zh)
WO (1) WO2014139175A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104780138A (zh) * 2014-01-09 2015-07-15 北京东土科技股份有限公司 私有冗余协议网络中stp/rstp报文的透传方法及装置
WO2019061269A1 (en) * 2017-09-29 2019-04-04 Interdigital Ce Patent Holdings IMPLEMENTING PSPUPV FOR DOCSIS ACCESS NETWORK

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104780104B (zh) * 2015-04-20 2018-05-11 新华三技术有限公司 在Smart Link组网中的报文转发方法及装置
US9923731B1 (en) * 2016-09-12 2018-03-20 Fujitsu Limited Seamless migration from multiple spanning tree protocol to ethernet ring protection switching protocol
CN107968753B (zh) * 2016-10-20 2021-08-13 华为技术有限公司 修改媒体接入控制地址转发表的方法和装置
CN110022236B (zh) * 2019-05-30 2022-03-22 新华三技术有限公司 一种报文转发方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101035047A (zh) * 2007-04-03 2007-09-12 中兴通讯股份有限公司 环网保护的处理方法
CN101252503A (zh) * 2008-02-05 2008-08-27 华为技术有限公司 环网的业务倒换方法和装置
US20130003530A1 (en) * 2011-06-28 2013-01-03 Shahram Davari Ethernet Ring Protection without MAC Table Flushing
CN102932183A (zh) * 2012-11-06 2013-02-13 杭州华三通信技术有限公司 双上行链路故障处理方法及设备

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6766482B1 (en) * 2001-10-31 2004-07-20 Extreme Networks Ethernet automatic protection switching
DE602006002961D1 (de) * 2006-05-04 2008-11-13 Nokia Siemens Networks Gmbh Automatische Sicherung der Weiterleitung der Pakete in das MPLS Netzwerk in einer doppeltangebundenen Ethernet Brücke
GB2449178B (en) * 2007-05-11 2009-09-23 Boeing Co Lightweight node based network redundancy solution leveraging rapid spanning tree protocol (RSTP)
US8054840B2 (en) * 2007-06-12 2011-11-08 International Business Machines Corporation Data center virtual local area network system and method
CN101834772B (zh) * 2009-03-13 2013-11-06 中兴通讯股份有限公司 一种基于以太环网的介质访问控制地址分类方法和装置
US8730963B1 (en) * 2010-11-19 2014-05-20 Extreme Networks, Inc. Methods, systems, and computer readable media for improved multi-switch link aggregation group (MLAG) convergence
US8565069B2 (en) * 2010-11-23 2013-10-22 Force10 Networks, Inc. Method of shrinking a data loss window in a packet network device
US20140301185A1 (en) * 2011-04-15 2014-10-09 Hangzhou H3C Technologies Co., Ltd Handling a fault in an ethernet ring network
CN102611628B (zh) * 2012-04-05 2015-08-05 杭州华三通信技术有限公司 用于实现传输路径切换的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101035047A (zh) * 2007-04-03 2007-09-12 中兴通讯股份有限公司 环网保护的处理方法
CN101252503A (zh) * 2008-02-05 2008-08-27 华为技术有限公司 环网的业务倒换方法和装置
US20130003530A1 (en) * 2011-06-28 2013-01-03 Shahram Davari Ethernet Ring Protection without MAC Table Flushing
CN102932183A (zh) * 2012-11-06 2013-02-13 杭州华三通信技术有限公司 双上行链路故障处理方法及设备

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104780138A (zh) * 2014-01-09 2015-07-15 北京东土科技股份有限公司 私有冗余协议网络中stp/rstp报文的透传方法及装置
CN104780138B (zh) * 2014-01-09 2017-12-19 北京东土科技股份有限公司 私有冗余协议网络中stp/rstp报文的透传方法及装置
WO2019061269A1 (en) * 2017-09-29 2019-04-04 Interdigital Ce Patent Holdings IMPLEMENTING PSPUPV FOR DOCSIS ACCESS NETWORK

Also Published As

Publication number Publication date
US20140307539A1 (en) 2014-10-16
EP2802105A4 (en) 2015-01-14
ES2586941T3 (es) 2016-10-19
EP2802105A1 (en) 2014-11-12
CN104160667B (zh) 2017-08-25
US9553760B2 (en) 2017-01-24
EP2802105B1 (en) 2016-05-18
CN104160667A (zh) 2014-11-19

Similar Documents

Publication Publication Date Title
US10862783B2 (en) OAM mechanisms for EVPN active-active services
US10686749B2 (en) Packet sending method and network device
US8204061B1 (en) Virtual port channel switches with distributed control planes
KR101808890B1 (ko) 네트워크 장애로부터 복구되기 위한 고속 플러딩 기반 고속 수렴
JP5899305B2 (ja) ネットワークノードを動作させる技術
US10110482B2 (en) Technique for network service availability
WO2014139175A1 (zh) 双上行相切环收敛的方法、设备和系统
WO2011120438A1 (zh) 一种通信网络组播保护方法、系统及设备
WO2007115493A1 (fr) Procédé, dispositif et système pour réaliser la commutation dans le réseau à double anneau de réseau vpls
WO2021031648A1 (zh) Evpn和vpls共存双活的方法、设备及系统
WO2012028029A1 (zh) 一种切换方法及系统
WO2019076295A1 (zh) 一种处理报文的方法、设备及系统
WO2007140683A1 (fr) Procédé, système et dispositif de protection de service, basés sur l'absence de connexion
JP2008236212A (ja) Vpn装置
CN111935013B (zh) 流量转发控制方法及装置、流量转发方法及芯片、交换机
WO2017054770A1 (zh) 集群通信
WO2012075831A1 (zh) 一种实现组播保护的方法及系统
US20220255847A1 (en) Packet Sending Method and First Network Device
JP7430224B2 (ja) パケット処理方法およびゲートウェイ・デバイス
WO2012171378A1 (zh) 解决vpls接入l3故障切换导致断流的方法及路由器
WO2012146097A1 (zh) Vpls网络和以太环网的倒换方法及装置
US8976659B2 (en) Intelligent layer-2 forwarding
WO2012152134A1 (zh) 以太网二层组播快速收敛的方法及以太网系统
WO2014169856A1 (zh) 一种组播通信方法和汇聚交换机
WO2023035983A1 (zh) 指定转发者df选举的方法及设备

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 2013861532

Country of ref document: EP

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13861532

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE