WO2015184759A1 - Apparatus and method for state detection and fault tolerance of service network port - Google Patents

Apparatus and method for state detection and fault tolerance of service network port Download PDF

Info

Publication number
WO2015184759A1
WO2015184759A1 PCT/CN2014/093489 CN2014093489W WO2015184759A1 WO 2015184759 A1 WO2015184759 A1 WO 2015184759A1 CN 2014093489 W CN2014093489 W CN 2014093489W WO 2015184759 A1 WO2015184759 A1 WO 2015184759A1
Authority
WO
WIPO (PCT)
Prior art keywords
network port
service
network
load
fault
Prior art date
Application number
PCT/CN2014/093489
Other languages
French (fr)
Chinese (zh)
Inventor
陈君
樊皓
李明哲
吴京洪
叶晓舟
郑艳伟
Original Assignee
中国科学院声学研究所
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 中国科学院声学研究所 filed Critical 中国科学院声学研究所
Publication of WO2015184759A1 publication Critical patent/WO2015184759A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks

Definitions

  • the present invention relates to the field of network communications, and in particular, to an apparatus and method for detecting and fault tolerance of a service network port.
  • the web service system serves the user by providing various contents according to the user's request.
  • the reliability of the system equipment is very high, and the system equipment can run stably for a long time. Once an abnormal situation occurs, it needs to be restored immediately, which requires a series of detection and prevention mechanisms.
  • Network port fault tolerance is a fault-tolerant mechanism commonly used in the industry.
  • the common fault of the network port is to prepare a backup network port.
  • the service of the abnormal network port is migrated to the standby network port.
  • the service migrated back but this caused the backup network port to be idle and not fully utilized the network port resources.
  • the object of the present invention is to overcome the shortcomings of the network port fault-tolerant method in the prior art, which need to be equipped with a dedicated backup network port, and can not fully utilize the defects of the network port resources, thereby providing a way to timely migrate the data served by the network port to the rich network. Devices and methods on the mouth.
  • the present invention provides a device for detecting and fault-tolerant service network port, which is used for managing various network ports on a server, including: a timing packet sending module, a timing detecting module, and a resource management module; ,
  • the timing packet sending module periodically sends the ping packet of the ICMP to each network port in the specified gateway or the server, so that the inbound traffic of the network port increases periodically; the timing detecting module periodically queries each network of the traffic register of the network port in the server.
  • the inbound traffic of the port reports to the resource management module that the growth of the inbound traffic is abnormal.
  • the resource management module queries all the load information of the abnormal network port and transfers the load information to other rich devices. A network port whose bandwidth can meet the load requirements.
  • the timing detection module compares the amount of increase of the inbound traffic of the network port with the specific amount of data sent by the network port, and if the amount of the inbound traffic of the network port increases by less than the specific amount of data sent by the network, the network The growth of the inbound and outbound traffic did not meet the standard.
  • the timing period of the timing detection module is greater than the timing period of the timing packet issuing module.
  • the resource management module records the status information of each network port and the related service load information.
  • the status information of the network port includes: the network port number, whether the network port runs normally, the maximum outgoing bandwidth of the network port, and the used Outbound bandwidth;
  • the related service load information of the network port includes each load information on the network port.
  • Each load information includes: bandwidth used by the load, destination MAC address, destination IP address, source IP address, and source MAC address.
  • the resource management module transfers the load information of the abnormal port to another network port that can meet the load requirement of the rich bandwidth, including: selecting a fault-tolerant network port with a rich bandwidth to meet the load requirement, and an abnormal network to be generated
  • the load data of the interface is migrated to the output of the fault-tolerant network port.
  • the source IP address and source MAC address information of the original load data packet are changed to the IP address and MAC address of the fault-tolerant network port, and the used bandwidth of the abnormal network port and the fault-tolerant network port is changed.
  • the source IP and source MAC information in the load information if there is no rich network port, notify the user in time that the service cannot be performed, and let the user restart the service.
  • the present invention also provides a method implemented by the device for detecting and fault-tolerant of a service network port, including:
  • Step 1) periodically send an ICMP ping packet to the designated gateway or each network port;
  • Step 2 Each network port periodically checks its own inbound traffic, and compares whether the growth of the inbound traffic is not less than the specific amount of data sent by itself. If the condition is met, the network port is working normally; if the condition is not met, the network port is in the normal state. Abnormal state
  • Step 3 Record the status information of each network port and related service load information.
  • the load data served by the network port is migrated to other network ports with rich bandwidth for output.
  • There is no rich network port and the user is notified in time that the service cannot be performed to restart the service.
  • the device and the method of the present invention can detect the abnormality of the network port in time, and migrate the service of the network port to other network ports on the server, thereby ensuring uninterrupted service of the user;
  • the device and method of the present invention can fully utilize the network port resources of the server and improve the number of concurrent services.
  • FIG. 1 is a functional block diagram of a service network port state detection and fault tolerance device of the present invention.
  • the service network port state detection and fault tolerance device of the present invention is located on the server, and can perform state detection and fault tolerance processing on each network port on the server; as shown in FIG. 1 , the device includes: a timing packet sending module, a timing detecting module, and resource management.
  • the timing packet sending module periodically sends the ICMP ping packet to the designated gateway or a network port, so that the incoming traffic of the network port can be periodically increased; the timing detection module periodically queries the traffic registers of each network port in the server.
  • the inbound traffic of the network port reports the network port abnormality to the resource management module if the traffic of the inbound traffic of the network port is not up to standard.
  • the resource management module queries all the load information of the abnormal network port for each load. Query whether the remaining network ports have rich bandwidth. If yes, migrate all the load data of the abnormal network port to another rich network port according to the load balancing policy, and modify the corresponding network port status information and related service load information. If not, notify the user that the service cannot be performed in time
  • the ICMP (Internet Control Message Protocol) used by the timing packet sending module is an Internet Control Message Protocol. It is a sub-protocol of the TCP/IP protocol suite for passing control messages between IP hosts and routers.
  • the control message refers to the network itself, such as the network is unreachable, the host is reachable, and the route is available. Although these control messages do not transmit user data, they play an important role in the transmission of user data.
  • the "ping" package can check if the network is connected, which can help us analyze and determine network failures. Therefore, the ICMP "ping" packet has a specific format.
  • the "ping" packet using ICMP is versatile. You can not only increase the inbound traffic of the network port by "pinging" its own network port, but also specify the "ping" by "ping". The address or gateway increases the incoming traffic of the network port by replying to the "ping" packet.
  • the timing detection module periodically queries the traffic registers of the network ports to check whether the increase of the traffic of the corresponding network port is not less than the specific data volume sent by itself (refers to the data stream containing the "ping" packet specifically used for network port state detection. Traffic, if yes, the outgoing link of the network port is connected, and the network port works normally; if not, the network port is in an abnormal state and cannot provide services to the external resource management module. Let the resource management module migrate the load data on the network port to the output of other network ports to ensure that the service continues. It should be noted that the timing period of the timing detection module needs to be greater than the timing period of the timing packet sending module. Preferably, in one timing detection process, two to three "ping" packets are sent out.
  • the resource management module needs to record the status information of each network port and the related service load information.
  • the network port status information includes: the network port number, whether the network port runs normally, the maximum outgoing bandwidth of the network port, the used bandwidth, etc.
  • the load information includes each load information on the network port, and each load information includes: a bandwidth used by the load, a destination MAC, a destination IP, a source IP, a source MAC, and the like.
  • the timing detection module reports that a network port is abnormal, it will happen.
  • the abnormal network port is used as the faulty network port.
  • the resource management module queries all the load information of the faulty network port. For each load, check whether the remaining bandwidth of the remaining network ports can meet the bandwidth of the load. If yes, the load balancing policy is used.
  • the source IP address and source MAC address of the original load data packet are changed to the IP address and MAC address of the fault-tolerant network port, and the error is changed.
  • the service network port state detection and fault tolerance method implemented by the present invention includes the following steps:
  • Step 1) Each network port periodically sends data of a specific rule to itself, including "pinging" a specific address.
  • Step 2 Each network port periodically checks its own inbound traffic, and compares whether the growth of the inbound traffic is not less than the specific amount of data sent by itself. If the condition is met, the network port is working normally; if the condition is not met, the network port is in the normal state. Abnormal state.
  • Step 3 Record the status information of each network port and related service load information.
  • the load data served by the network port is migrated to other available network ports for output to ensure uninterrupted service.

Abstract

The present invention relates to an apparatus for state detection and fault tolerance of a service network port. The apparatus is used for managing various network ports on a server, and comprises a periodic packet sending module, a periodic detection module, and a resource management module. The periodic packet sending module periodically sends an ICMP ping packet to each network port in a designated gateway or server to periodically increase the ingress flow of the network port. The periodic detection module periodically queries a flow register of each network port in the server for the ingress flow of the network port, and reports to the resource management module that an anomaly occurs on network ports whose ingress flow increase does not meet the criterion. The resource management module queries for all load information of the network ports on which the anomaly occurs, and transfers the load information to other network ports whose surplus bandwidth can meet load requirements.

Description

一种服务网口状态检测和容错的装置及其方法Device and method for detecting status and fault tolerance of service network port 技术领域Technical field
本发明涉及网络通信领域，特别涉及一种服务网口状态检测和容错的装置及其方法。The present invention relates to the field of network communications, and in particular, to an apparatus and method for detecting and fault tolerance of a service network port.
背景技术Background technique
随着互联网技术和广播电视技术的不断发展，基于网络的服务业务突飞猛进，网络服务将成为未来互联网的核心应用之一。With the continuous development of Internet technology and broadcast TV technology, network-based service business is advancing by leaps and bounds, and network services will become one of the core applications of the Internet in the future.
网络服务系统根据用户的请求，通过提供各种内容来服务用户。对于服务系统来说，系统设备的可靠性需要很高，系统设备能够长时间稳定运行，一旦出现异常情况，需要立刻能够恢复，这就需要一系列的检测和预防机制。The web service system serves the user by providing various contents according to the user's request. For the service system, the reliability of the system equipment is very high, and the system equipment can run stably for a long time. Once an abnormal situation occurs, it needs to be restored immediately, which requires a series of detection and prevention mechanisms.
通常情况下，如果服务系统正在提供服务的节点传输链路因拥塞而失效、节点的服务能力突然下降、接收的数据不完整等，这些情况都会严重影响用户的体验。为了保证用户节点接受服务的连续性，必须采取一些容错机制使网络的服务能力不受影响或尽快恢复。Usually, if the service transmission node of the service system is failing due to congestion, the service capability of the node suddenly drops, and the received data is incomplete, these conditions will seriously affect the user experience. In order to ensure the continuity of the service received by the user node, some fault tolerance mechanism must be adopted to make the service capability of the network unaffected or recover as soon as possible.
网口容错是业界通常采用的一种容错机制。所述网口容错的普遍做法是：预先准备一个备用网口，当检测到某个网口发生异常时，将发生异常网口的服务迁移到备用网口上，当异常网口恢复时，又将服务迁移回来，但这样就导致备用网口经常处于空闲状态，未能充分利用网口资源。Network port fault tolerance is a fault-tolerant mechanism commonly used in the industry. The common fault of the network port is to prepare a backup network port. When an abnormality occurs on a network port, the service of the abnormal network port is migrated to the standby network port. When the abnormal network port is restored, The service migrated back, but this caused the backup network port to be idle and not fully utilized the network port resources.
发明内容Summary of the invention
本发明的目的在于克服现有技术中的网口容错方法需要配备专用的备用网口，不能充分利用网口资源的缺陷，从而提供一种能够及时将该网口所服务的数据迁移到富裕网口上的装置与方法。The object of the present invention is to overcome the shortcomings of the network port fault-tolerant method in the prior art, which need to be equipped with a dedicated backup network port, and can not fully utilize the defects of the network port resources, thereby providing a way to timely migrate the data served by the network port to the rich network. Devices and methods on the mouth.
为了实现上述目的，本发明提供了一种服务网口状态检测和容错的装置，该装置用于对服务器上的各个网口进行管理，包括：定时发包模块、定时检测模块以及资源管理模块；其中，In order to achieve the above object, the present invention provides a device for detecting and fault-tolerant service network port, which is used for managing various network ports on a server, including: a timing packet sending module, a timing detecting module, and a resource management module; ,
所述定时发包模块定时发送ICMP的ping包到指定网关或者服务器中的各个网口，使网口的入端流量定时增长；所述定时检测模块定时向服务器中各网口的流量寄存器查询各网口的入端流量，向资源管理模块报告入端流量的增长未达标的网口出现异常；所述资源管理模块查询发生异常的网口的所有负载信息，将这些负载信息转移到其他富裕 带宽能够满足负载需求的网口。The timing packet sending module periodically sends the ping packet of the ICMP to each network port in the specified gateway or the server, so that the inbound traffic of the network port increases periodically; the timing detecting module periodically queries each network of the traffic register of the network port in the server. The inbound traffic of the port reports to the resource management module that the growth of the inbound traffic is abnormal. The resource management module queries all the load information of the abnormal network port and transfers the load information to other rich devices. A network port whose bandwidth can meet the load requirements.
上述技术方案中，所述定时检测模块将网口的入端流量的增长量与自身发送的特定数据量进行比较，若网口的入端流量的增长量小于自身发送的特定数据量，则网口入端流量的增长未达标。In the above technical solution, the timing detection module compares the amount of increase of the inbound traffic of the network port with the specific amount of data sent by the network port, and if the amount of the inbound traffic of the network port increases by less than the specific amount of data sent by the network, the network The growth of the inbound and outbound traffic did not meet the standard.
上述技术方案中，所述定时检测模块的定时周期大于所述定时发包模块的定时周期。In the above technical solution, the timing period of the timing detection module is greater than the timing period of the timing packet issuing module.
上述技术方案中，所述资源管理模块记录每个网口的状态信息和相关服务负载信息；其中，网口的状态信息包括：网口号、网口是否正常运行、网口最大出带宽、已使用出带宽；网口的相关服务负载信息包括该网口上的每个负载信息，每个负载信息包括：负载使用的带宽、目的MAC、目的IP、源IP、源MAC。In the foregoing technical solution, the resource management module records the status information of each network port and the related service load information. The status information of the network port includes: the network port number, whether the network port runs normally, the maximum outgoing bandwidth of the network port, and the used Outbound bandwidth; the related service load information of the network port includes each load information on the network port. Each load information includes: bandwidth used by the load, destination MAC address, destination IP address, source IP address, and source MAC address.
上述技术方案中，所述资源管理模块将出现异常的端口的负载信息转移到其他富裕带宽能够满足负载需求的网口包括：选取一个富裕带宽能够满足负载需求的容错网口，将发生异常的网口的负载数据迁移到该容错网口输出，原始负载数据报文中的源IP和源MAC信息变更为容错网口的IP和MAC，并更改发生异常网口和容错网口的已使用出带宽以及负载信息中的源IP和源MAC信息；如果没有富裕网口，及时向用户通知该服务不能进行，让用户重新开启服务。In the foregoing technical solution, the resource management module transfers the load information of the abnormal port to another network port that can meet the load requirement of the rich bandwidth, including: selecting a fault-tolerant network port with a rich bandwidth to meet the load requirement, and an abnormal network to be generated The load data of the interface is migrated to the output of the fault-tolerant network port. The source IP address and source MAC address information of the original load data packet are changed to the IP address and MAC address of the fault-tolerant network port, and the used bandwidth of the abnormal network port and the fault-tolerant network port is changed. And the source IP and source MAC information in the load information; if there is no rich network port, notify the user in time that the service cannot be performed, and let the user restart the service.
本发明还提供了基于所述的服务网口状态检测和容错的装置所实现的方法，包括：The present invention also provides a method implemented by the device for detecting and fault-tolerant of a service network port, including:
步骤1)、向指定网关或各网口定时发送ICMP的ping包；Step 1), periodically send an ICMP ping packet to the designated gateway or each network port;
步骤2)、各网口定期检测自身的入端流量，比较入端流量的增长量是否不小于自身发送的特定数据量，达到条件则表明网口在正常工作；未达到条件则表明网口处于异常状态；Step 2) Each network port periodically checks its own inbound traffic, and compares whether the growth of the inbound traffic is not less than the specific amount of data sent by itself. If the condition is met, the network port is working normally; if the condition is not met, the network port is in the normal state. Abnormal state
步骤3)、记录每个网口的状态信息和相关服务负载信息，当检测到某个网口异常时，将该网口所服务的负载数据迁移到其他具有富裕带宽的网口进行输出；若没有富裕网口，及时向用户通知该服务不能进行，以重新开启服务。Step 3) Record the status information of each network port and related service load information. When a network port is abnormal, the load data served by the network port is migrated to other network ports with rich bandwidth for output. There is no rich network port, and the user is notified in time that the service cannot be performed to restart the service.
本发明的优点在于：The advantages of the invention are:
1、本发明的装置与方法能够及时检测出发生异常的网口，将该网口的服务迁移到服务器上的其他网口，从而保证用户服务的不间断；The device and the method of the present invention can detect the abnormality of the network port in time, and migrate the service of the network port to other network ports on the server, thereby ensuring uninterrupted service of the user;
2、本发明的装置与方法能够充分利用服务器的网口资源，提高服务并发数。2. The device and method of the present invention can fully utilize the network port resources of the server and improve the number of concurrent services.
附图说明DRAWINGS
图1是本发明的服务网口状态检测和容错装置的功能模块图。 1 is a functional block diagram of a service network port state detection and fault tolerance device of the present invention.
具体实施方式detailed description
现结合附图对本发明作进一步的描述。The invention will now be further described with reference to the drawings.
本发明的服务网口状态检测和容错装置位于服务器上，能够对服务器上的各个网口做状态检测与容错处理；如图1所示，该装置包括：定时发包模块、定时检测模块以及资源管理模块；其中的定时发包模块定时发送ICMP的ping包到指定网关或者是某一网口，使该网口的入端流量能够定时增长；定时检测模块定时向服务器中各网口的流量寄存器查询各网口的入端流量，若某一个网口的入端流量的增长未达标，则向资源管理模块报告该网口异常；资源管理模块查询发生异常的网口的所有负载信息，针对每个负载，查询其余网口是否有富裕带宽，如果有，则根据负载均衡策略，将该异常网口的所有负载数据迁移到另一个富裕网口输出，并修改相应的网口状态信息和相关服务负载信息，如果没有，则及时通知用户服务不能进行。The service network port state detection and fault tolerance device of the present invention is located on the server, and can perform state detection and fault tolerance processing on each network port on the server; as shown in FIG. 1 , the device includes: a timing packet sending module, a timing detecting module, and resource management. The timing packet sending module periodically sends the ICMP ping packet to the designated gateway or a network port, so that the incoming traffic of the network port can be periodically increased; the timing detection module periodically queries the traffic registers of each network port in the server. The inbound traffic of the network port reports the network port abnormality to the resource management module if the traffic of the inbound traffic of the network port is not up to standard. The resource management module queries all the load information of the abnormal network port for each load. Query whether the remaining network ports have rich bandwidth. If yes, migrate all the load data of the abnormal network port to another rich network port according to the load balancing policy, and modify the corresponding network port status information and related service load information. If not, notify the user that the service cannot be performed in time.
下面对本发明的装置中的各个模块做进一步说明。Further description of each module in the apparatus of the present invention will be given below.
定时发包模块所采用的ICMP(Internet Control Message Protocol)是Internet控制报文协议。它是TCP/IP协议族的一个子协议，用于在IP主机、路由器之间传递控制消息。控制消息是指网络通不通、主机是否可达、路由是否可用等网络本身的消息。这些控制消息虽然并不传输用户数据，但是对于用户数据的传递起着重要的作用。“ping”包可以检查网络是否连通，可以很好地帮助我们分析和判定网络故障。因此，ICMP的“ping”包有特定的格式，采用ICMP的“ping”包具有通用性，不仅可以通过“ping”自身网口来增加网口的入端流量，也可以通过“ping”指定的地址或者网关，通过回复的“ping”包来增加网口的入端流量。The ICMP (Internet Control Message Protocol) used by the timing packet sending module is an Internet Control Message Protocol. It is a sub-protocol of the TCP/IP protocol suite for passing control messages between IP hosts and routers. The control message refers to the network itself, such as the network is unreachable, the host is reachable, and the route is available. Although these control messages do not transmit user data, they play an important role in the transmission of user data. The "ping" package can check if the network is connected, which can help us analyze and determine network failures. Therefore, the ICMP "ping" packet has a specific format. The "ping" packet using ICMP is versatile. You can not only increase the inbound traffic of the network port by "pinging" its own network port, but also specify the "ping" by "ping". The address or gateway increases the incoming traffic of the network port by replying to the "ping" packet.
定时检测模块定时向各网口的流量寄存器查询对应网口入端流量的增长量是否不小于自身发送的特定数据量(指包含有专门用于网口状态检测的“ping”包的数据流的流量)，如果是，则说明该网口向外的链路是联通的，网口正常工作；如果不是，则说明该网口处于异常状态，不能向外提供服务，需要向资源管理模块报告，让资源管理模块将该网口上的负载数据迁移到其他网口上输出，保证服务继续进行。需要注意定时检测模块的定时周期需要大于定时发包模块的定时周期，最好在一次定时检测过程中，有两到三个“ping”包发送出去。The timing detection module periodically queries the traffic registers of the network ports to check whether the increase of the traffic of the corresponding network port is not less than the specific data volume sent by itself (refers to the data stream containing the "ping" packet specifically used for network port state detection. Traffic, if yes, the outgoing link of the network port is connected, and the network port works normally; if not, the network port is in an abnormal state and cannot provide services to the external resource management module. Let the resource management module migrate the load data on the network port to the output of other network ports to ensure that the service continues. It should be noted that the timing period of the timing detection module needs to be greater than the timing period of the timing packet sending module. Preferably, in one timing detection process, two to three "ping" packets are sent out.
资源管理模块需要记录每个网口的状态信息和相关服务负载信息，网口状态信息包括：网口号、网口是否正常运行、网口最大出带宽、已使用出带宽等；网口相关的服务负载信息包括该网口上的每个负载信息，每个负载信息包括：负载使用的带宽、目的MAC、目的IP、源IP、源MAC等。当定时检测模块报告某个网口异常时，将该发生 异常的网口作为出错网口，资源管理模块查询该出错网口的所有负载信息，针对每个负载，查询其余网口的剩余带宽是否能满足该负载的带宽，如果满足，则根据负载均衡策略，选取一个容错网口，将该负载在出错网口的负载数据迁移到容错网口输出，原始负载数据报文中的源IP和源MAC信息变更为容错网口的IP和MAC，并更改出错网口和容错网口的已使用出带宽以及负载信息中的源IP和源MAC信息；如果没有富裕网口，则及时向用户通知该服务不能进行，让用户重新开启服务。The resource management module needs to record the status information of each network port and the related service load information. The network port status information includes: the network port number, whether the network port runs normally, the maximum outgoing bandwidth of the network port, the used bandwidth, etc. The load information includes each load information on the network port, and each load information includes: a bandwidth used by the load, a destination MAC, a destination IP, a source IP, a source MAC, and the like. When the timing detection module reports that a network port is abnormal, it will happen. The abnormal network port is used as the faulty network port. The resource management module queries all the load information of the faulty network port. For each load, check whether the remaining bandwidth of the remaining network ports can meet the bandwidth of the load. If yes, the load balancing policy is used. Select a fault-tolerant network port to migrate the load data of the faulty network port to the fault-tolerant network port. The source IP address and source MAC address of the original load data packet are changed to the IP address and MAC address of the fault-tolerant network port, and the error is changed. The source IP and source MAC information in the used bandwidth and load information of the network port and the fault-tolerant network port; if there is no rich network port, notify the user in time that the service cannot be performed, and let the user restart the service.
基于本发明的装置，本发明所实现的服务网口状态检测和容错方法包括以下步骤：Based on the device of the present invention, the service network port state detection and fault tolerance method implemented by the present invention includes the following steps:
步骤1)、各网口定期向自身发送特定规则的数据，包括“ping”一个特定地址。Step 1) Each network port periodically sends data of a specific rule to itself, including "pinging" a specific address.
步骤2)、各网口定期检测自身的入端流量，比较入端流量的增长量是否不小于自身发送的特定数据量，达到条件则表明网口在正常工作；未达到条件则表明网口处于异常状态。Step 2) Each network port periodically checks its own inbound traffic, and compares whether the growth of the inbound traffic is not less than the specific amount of data sent by itself. If the condition is met, the network port is working normally; if the condition is not met, the network port is in the normal state. Abnormal state.
步骤3)、记录每个网口的状态信息和相关服务负载信息。当检测到某个网口异常时，将该网口所服务的负载数据迁移到其他可用的网口进行输出，保证用户服务的不间断。Step 3) Record the status information of each network port and related service load information. When a certain network port is abnormal, the load data served by the network port is migrated to other available network ports for output to ensure uninterrupted service.
最后所应说明的是，以上实施例仅用以说明本发明的技术方案而非限制。尽管参照实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，对本发明的技术方案进行修改或者等同替换，都不脱离本发明技术方案的精神和范围，其均应涵盖在本发明的权利要求范围当中。 Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention and not limiting. While the invention has been described in detail herein with reference to the embodiments of the embodiments of the invention Within the scope of the claims.

Claims (6)

  1. 一种服务网口状态检测和容错的装置，该装置用于对服务器上的各个网口进行管理，其特征在于，包括：定时发包模块、定时检测模块以及资源管理模块；其中，A device for detecting and fault-tolerating a service network port, wherein the device is configured to manage each network port on the server, and is characterized by: a timing sending module, a timing detecting module, and a resource management module; wherein
    所述定时发包模块定时发送ICMP的ping包到指定网关或者服务器中的各个网口，使网口的入端流量定时增长；所述定时检测模块定时向服务器中各网口的流量寄存器查询各网口的入端流量，向资源管理模块报告入端流量的增长未达标的网口出现异常；所述资源管理模块查询发生异常的网口的所有负载信息，将这些负载信息转移到其他富裕带宽能够满足负载需求的网口。The timing packet sending module periodically sends the ping packet of the ICMP to each network port in the specified gateway or the server, so that the inbound traffic of the network port increases periodically; the timing detecting module periodically queries each network of the traffic register of the network port in the server. The inbound traffic of the port reports to the resource management module that the growth of the inbound traffic is abnormal. The resource management module queries all the load information of the abnormal network port and transfers the load information to other rich bandwidth. A network port that meets the load requirements.
  2. 根据权利要求1所述的服务网口状态检测和容错的装置，其特征在于，所述定时检测模块将网口的入端流量的增长量与自身发送的特定数据量进行比较，若网口的入端流量的增长量小于自身发送的特定数据量，则网口入端流量的增长未达标。The apparatus for detecting and fault-tolerant service network port status according to claim 1, wherein the timing detection module compares the amount of increase of the incoming traffic of the network port with the specific amount of data sent by itself, if the network port If the increase in the incoming traffic is less than the specific amount of data sent by itself, the increase in the incoming traffic of the network port is not up to standard.
  3. 根据权利要求1所述的服务网口状态检测和容错的装置，其特征在于，所述定时检测模块的定时周期大于所述定时发包模块的定时周期。The apparatus for detecting and fault-tolerant service network port according to claim 1, wherein the timing period of the timing detection module is greater than the timing period of the timing packet issuing module.
  4. 根据权利要求1所述的服务网口状态检测和容错的装置，其特征在于，所述资源管理模块记录每个网口的状态信息和相关服务负载信息；其中，网口的状态信息包括：网口号、网口是否正常运行、网口最大出带宽、已使用出带宽；网口的相关服务负载信息包括该网口上的每个负载信息，每个负载信息包括：负载使用的带宽、目的MAC、目的IP、源IP、源MAC。The apparatus for detecting and fault-tolerant service network port status according to claim 1, wherein the resource management module records status information of each network port and related service load information; wherein the status information of the network port includes: Whether the slogan, the network port is working properly, the maximum outgoing bandwidth of the network port, and the used bandwidth; the service load information of the network port includes each load information on the network port. Each load information includes: bandwidth used by the load, destination MAC address, Destination IP, source IP, and source MAC.
  5. 根据权利要求4所述的服务网口状态检测和容错的装置，其特征在于，所述资源管理模块将出现异常的端口的负载信息转移到其他富裕带宽能够满足负载需求的网口包括：选取一个富裕带宽能够满足负载需求的容错网口，将发生异常的网口的负载数据迁移到该容错网口输出，原始负载数据报文中的源IP和源MAC信息变更为容错网口的IP和MAC，并更改发生异常网口和容错网口的已使用出带宽以及负载信息中的源IP和源MAC信息；如果没有富裕网口，及时向用户通知该服务不能进行，让用户重新开启服务。The apparatus for detecting and fault-tolerant service network port status according to claim 4, wherein the resource management module transfers the load information of the abnormally occurring port to the network port of the other rich bandwidth that can meet the load requirement, including: selecting one A fault-tolerant network port that meets the load demand of the rich bandwidth. The load data of the abnormal network port is migrated to the output of the fault-tolerant network port. The source IP address and source MAC address information of the original load data packet are changed to the IP address and MAC address of the fault-tolerant network port. And change the used outgoing bandwidth of the abnormal network port and the fault-tolerant network port and the source IP and source MAC information in the load information; if there is no rich network port, notify the user in time that the service cannot be performed, and let the user restart the service.
  6. 基于权利要求1-5之一所述的服务网口状态检测和容错的装置所实现的方法，包括： The method implemented by the device for detecting and fault-tolerant service network port according to any one of claims 1 to 5, comprising:
    步骤1)、向指定网关或各网口定时发送ICMP的ping包；Step 1), periodically send an ICMP ping packet to the designated gateway or each network port;
    步骤2)、各网口定期检测自身的入端流量，比较入端流量的增长量是否不小于自身发送的特定数据量，达到条件则表明网口在正常工作；未达到条件则表明网口处于异常状态；Step 2) Each network port periodically checks its own inbound traffic, and compares whether the growth of the inbound traffic is not less than the specific amount of data sent by itself. If the condition is met, the network port is working normally; if the condition is not met, the network port is in the normal state. Abnormal state
    步骤3)、记录每个网口的状态信息和相关服务负载信息，当检测到某个网口异常时，将该网口所服务的负载数据迁移到其他具有富裕带宽的网口进行输出；若没有富裕网口，及时向用户通知该服务不能进行，以重新开启服务。 Step 3) Record the status information of each network port and related service load information. When a network port is abnormal, the load data served by the network port is migrated to other network ports with rich bandwidth for output. There is no rich network port, and the user is notified in time that the service cannot be performed to restart the service.
PCT/CN2014/093489 2014-06-04 2014-12-10 Apparatus and method for state detection and fault tolerance of service network port WO2015184759A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410245842.3 2014-06-04
CN201410245842.3A CN105281929B (en) 2014-06-04 2014-06-04 A kind of service network interface state-detection and fault-tolerant devices and methods therefor

Publications (1)

Publication Number Publication Date
WO2015184759A1 true WO2015184759A1 (en) 2015-12-10

Family

ID=54766020

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/093489 WO2015184759A1 (en) 2014-06-04 2014-12-10 Apparatus and method for state detection and fault tolerance of service network port

Country Status (2)

Country Link
CN (1) CN105281929B (en)
WO (1) WO2015184759A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112445662A (en) * 2019-08-30 2021-03-05 上海哔哩哔哩科技有限公司 Internet data broadcast socket testing method, server and storage medium
CN112565746A (en) * 2020-12-30 2021-03-26 杭州视洞科技有限公司 Automatic pressure test method and process for detecting IP address of wired network port of camera
CN112672203A (en) * 2020-12-16 2021-04-16 努比亚技术有限公司 File transfer control method, mobile terminal and computer readable storage medium
CN114244723A (en) * 2021-09-29 2022-03-25 浙江国利网安科技有限公司 Service flow simulation method and device and service flow simulator

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102307122A (en) * 2011-09-06 2012-01-04 北京傲天动联技术有限公司 Ethernet over Coax (EoC) link failure detection system and method
CN202649363U (en) * 2011-12-19 2013-01-02 光一科技股份有限公司 Network port detection device

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201114126Y (en) * 2007-10-26 2008-09-10 中兴通讯股份有限公司 Multi- net opening test device
JP5170000B2 (en) * 2009-06-04 2013-03-27 富士通株式会社 Redundant pair detection method, communication device, redundant pair detection program, recording medium
CN102447639B (en) * 2012-01-17 2016-03-09 华为技术有限公司 A kind of policy routing method and device
CN102833591B (en) * 2012-08-09 2015-08-12 中兴通讯股份有限公司 The unbroken method of order program service and device in interactive Web TV system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102307122A (en) * 2011-09-06 2012-01-04 北京傲天动联技术有限公司 Ethernet over Coax (EoC) link failure detection system and method
CN202649363U (en) * 2011-12-19 2013-01-02 光一科技股份有限公司 Network port detection device

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112445662A (en) * 2019-08-30 2021-03-05 上海哔哩哔哩科技有限公司 Internet data broadcast socket testing method, server and storage medium
CN112445662B (en) * 2019-08-30 2022-12-02 上海哔哩哔哩科技有限公司 Internet data broadcast socket testing method, server and storage medium
CN112672203A (en) * 2020-12-16 2021-04-16 努比亚技术有限公司 File transfer control method, mobile terminal and computer readable storage medium
CN112672203B (en) * 2020-12-16 2023-05-23 努比亚技术有限公司 File transfer control method, mobile terminal and computer readable storage medium
CN112565746A (en) * 2020-12-30 2021-03-26 杭州视洞科技有限公司 Automatic pressure test method and process for detecting IP address of wired network port of camera
CN114244723A (en) * 2021-09-29 2022-03-25 浙江国利网安科技有限公司 Service flow simulation method and device and service flow simulator

Also Published As

Publication number Publication date
CN105281929B (en) 2018-10-02
CN105281929A (en) 2016-01-27

Similar Documents

Publication Publication Date Title
US20230308421A1 (en) Method and system of establishing a virtual private network in a cloud service for branch networking
US9705735B2 (en) System and method using RSVP hello suppression for graceful restart capable neighbors
EP2242325B1 (en) Method, system and equipment for access of a network device to a packet exchange network
US9059902B2 (en) Procedures, apparatuses, systems, and computer-readable media for operating primary and backup network elements
US8868998B2 (en) Packet communication apparatus and packet communication method
US9077617B1 (en) Kernel-based TCP-layer assist for fast recovery by backup control unit of a device
US20140119176A1 (en) Methods and Apparatus for Improving Network Communication Using Ethernet Switching Protection
WO2021018309A1 (en) Method, device and system for determination of message transmission path, and computer storage medium
WO2018113425A1 (en) Method, apparatus and system for detecting time delay
US11902130B2 (en) Data packet loss detection
JP7313480B2 (en) Congestion Avoidance in Slice-Based Networks
US20150016245A1 (en) Method and apparatus for protection switching in packet transport system
WO2015184759A1 (en) Apparatus and method for state detection and fault tolerance of service network port
EP3576347A1 (en) Network device snapshots
EP4142239A1 (en) Network performance monitoring and fault management based on wide area network link health assessments
WO2015149353A1 (en) Oam packet processing method, network device and network system
US20120320737A1 (en) Method and apparatus for lossless link recovery between two devices interconnected via multi link trunk/link aggregation group (mlt/lag)
JP5352502B2 (en) Packet communication system and packet communication apparatus control method
KR20200072941A (en) Method and apparatus for handling VRRP(Virtual Router Redundancy Protocol)-based network failure using real-time fault detection
US11916770B2 (en) Pinpointing sources of jitter in network flows
US11290319B2 (en) Dynamic distribution of bidirectional forwarding detection echo sessions across a multi-processor system
CN113037622B (en) System and method for preventing BFD from vibrating
US10924391B2 (en) Systems and methods for automatic traffic recovery after VRRP VMAC installation failures in a LAG fabric
Zhang et al. A service protection mechanism impelemented on P4 by packet replication
CN107241206A (en) The method and device that a kind of business service state judges

Legal Events

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

Ref document number: 14893944

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14893944

Country of ref document: EP

Kind code of ref document: A1