TWI232659B - Method for managing a stack of switches - Google Patents

Method for managing a stack of switches Download PDF

Info

Publication number
TWI232659B
TWI232659B TW93106156A TW93106156A TWI232659B TW I232659 B TWI232659 B TW I232659B TW 93106156 A TW93106156 A TW 93106156A TW 93106156 A TW93106156 A TW 93106156A TW I232659 B TWI232659 B TW I232659B
Authority
TW
Taiwan
Prior art keywords
switch
module
master
slave
command
Prior art date
Application number
TW93106156A
Other languages
Chinese (zh)
Other versions
TW200531478A (en
Inventor
Chuan-Cheng Hsu
Yi-Chiung Weng
Original Assignee
Hon Hai Prec Ind Co Ltd
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 Hon Hai Prec Ind Co Ltd filed Critical Hon Hai Prec Ind Co Ltd
Priority to TW93106156A priority Critical patent/TWI232659B/en
Application granted granted Critical
Publication of TWI232659B publication Critical patent/TWI232659B/en
Publication of TW200531478A publication Critical patent/TW200531478A/en

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

A method for managing a stack of switches is provided. The method comprising the steps of: building a stack of switches; sending out advertisement packets which contains topology information to both stacking links to accomplish topology discovery; electing a switch as master switch, and the other switches as slave switches; configuring the slave switches via the master switch; obtaining the ports statistics of slave switches via the master switch; obtaining the status of slave switches via the master switch; Via the circuit method of the present invention, users can manage all the slave switches conveniently via the master switch.

Description

I232659_ 五、發明說明(1) 【發明所屬之技術領域】 本發明係關於一種Φ ;β挪士、土 堆疊式交換機之管理方J換機s理方法’特別係關於-種 【先前技術】 乙太網交換機的堆Α, θ 户A 一起,組成可以槎#夕二疋將夕σ乙太網交換機連接在 用戶提供太網交換機,為更多的 太網交換機由於JL愈=;=罔而要支持堆疊功能的乙 企業網及寬帶组網^用:^ 性’統—的對外網管,在 疊式交換機中需要;“j越來越受到重視。但是在堆 曰公開心!稱γλ國家知識產權局於_年4月; 了 明專利申冑,盆=Λ現以太網交換機堆疊的方法”的發 將乙太網交換機輸豐的方法,該方法具體為: 千兆串行訊號,再將:千::仃讯號通過轉換電路轉換為 ?乙太網交換機堆疊經線纜傳輸至需要與 父換璣接收盥該乙太網f施二乙太網交換機;而當乙太網 =來的千兆串行訊;= —台乙太網交換機 、換j路轉換為千兆並行:千過轉換電路 ,由該發明所提供的技術,可以J!:太網父換機。 〜未提出如何有效管理堆疊中=:的:;;離 _ J貫現乙太網交換機機堆疊 ”=申請公開 太網交換機輪出的工、+ 2 > 法该方法具體為: 第5頁 T232659 五、發明說明(2) 種可以有效管理堆疊中交換機的管理方。 【發明内容】 為了解決上述之技術問題,本發明提供 父換機管理方法。該堆疊式交換機管理方法—種堆疊式 驟:(a)藉由堆疊線將複數可堆疊之交換匕括^下步 拓扑結構連接起來,建立一堆疊;(b )夂_按肊特定的 的交換機發送有包括有自身拓=結 交換機向鄰近 弋換機為主交換機,則其餘之交換機一 :T其 二拖T交換機對從交換機進行遠程配置 獲取=交換機的狀態訊息。十況心,(ί)精由主交換機 藉由本發明所提供 通過主交換機對;:二:法,者可以报方便地 【實施方式】 堆$中的所有從交換機進行管理。 i i為i J :月所主要涉及的-些技術術語: -組交換機的‘:2應連接器和線纜物理連接在一起的 主交換機:土奋田 代表,其餘交·換::口㉟,為對堆疊進行管理的 ; 馎杓可通過邊交換機的I p位址來進行管理 從交換機:埯w丄 。 隹豐中的交換機之一,其不可直接被管理 本說明書中 所述之堆疊是指堆疊式交換機,其一般由I232659_ V. Description of the invention (1) [Technical field to which the invention belongs] The present invention relates to a Φ; β Norse, earth-stacked switch manager J replacement method, especially related to-a kind of [prior art] B The stack A of the Ethernet switch, θ and the user A together, can be combined to provide a #Ethernet switch to the user to provide the Ethernet switch. For more Ethernet switches, the JL becomes more important; Enterprise network B and broadband networking that support stacking functions ^ Use: ^ "External" network management is required in stacked switches; "j is getting more and more attention. But openly in the heap! Called γλ national intellectual property Bureau in April _; a clear patent application, basin = Λ method of stacking existing Ethernet switches "issued a method to send Ethernet switches to rich, the method is specifically: Gigabit serial signal, and then: Thousands :: The signal is converted into the Ethernet switch stack through the conversion circuit and transmitted via the cable to the Ethernet switch that needs to be exchanged with the parent to receive the ethernet Ethernet switch; and when Ethernet = come thousand Terabit serial communication; = —Ethernet switch, change j Gigabit parallel conversion: conversion circuit over one thousand, by the technique provided by the invention may be a parent J !: Ethernet switch. ~ Did not propose how to effectively manage the stack == :: ;; __J to show the stack of the Ethernet switch machine "= apply for the public switch of the Ethernet switch, + 2 > method This method is specifically: Page 5 T232659 V. Description of the invention (2) A management party that can effectively manage switches in a stack. [Summary] In order to solve the above technical problems, the present invention provides a parent switch management method. The stacking switch management method-a stacking step : (A) Connect multiple stackable switches by stacking lines ^ The next step is to connect the topological structure to establish a stack; (b) 夂 _ Press the specific switch to send it to the neighbor switch弋 When the switch is the master switch, the remaining switches are: one of the other switches: the T switch is used to remotely configure the slave switch to obtain the status information of the switch. (10) Hearts, (ί) The master switch provides Switch pair ;: Two: method, you can report conveniently [Implementation] All slave switches in the stack $ are managed. Ii is i J: the main terms involved in the month-some technical terms:-group The main switch of the ': 2 should be physically connected to the connector and the cable: Representative of Dofentian, the rest is exchanged :: 口 ㉟, which is used to manage the stack; 馎 杓 I p Address to manage the slave switch: 埯 w 丄. One of the switches in HSBC, which cannot be managed directly. The stacking described in this manual refers to the stacking switch, which is generally composed of

12326591232659

五、發明說明(3) 複數獨立之交換機按一定的拓 構一般可分為菊花 厂構釔、…而成,該拓扑結 )拓扑結構,其中严 aiSy Chained )或環型(Ring ;堆疊中所包2的:ϊ : 構提供有一冗餘之堆疊鏈路 第一圖即示出了 ^ ^ 飯攸一到十幾個不等。 每一交換# = ! 一 交換機所組成的堆叠I,盆中 :上含ΓΓ疊端π,其-為上行端二用 其中通過;ΐ;:=:Γ“,用於下行鏈路連接。 如卜卢持j、 °卩刀的連接構成了菊花鏈型拓扑結構,另外 :二。_卩分的冗餘線則構成了 一個環型拓扑結構。 Μ模:弟ί圖所示,係本發明主交換機與從交換機内部邏 f&lif]思圖。主交換機10與從交換機20均包含有用戶介 、/ 1 0 ;服務模組1 2 0、2 2 0 ;硬體概念層模組i 3 〇、 2 3 0 ;驅動模組14〇、24〇 ;資料庫管理模組丨“、25〇及交 換,間通訊模組160、26〇。其中,用戶介面11〇、21〇係用 於父換機與使用者間進行通訊,在該發明中,從交換機的 使用者不會和從交換機的用戶介面2 1 0進行通訊,而當該 從父換機變成主交換機時,使用者即可利用該用戶介面進 行管理;服務模組120、2 2 0係包含有複數應用程式,其可 為交換機提供相應的服務;而硬體概念層模組(Hardware Abstract Layer,HAL ) 130、230是交換機中的硬體的抽象 ’是用來支持交換機中的各種程式及服務的;驅動模組( Driver) 140、240是用來驅動交換機中的各種硬體;而資 料庫管理模組(Database Maintaince Protocol,DMP) 1 5 0、2 5 0是用來管理交換機中的資料庫、本地快取存儲器V. Description of the invention (3) According to a certain topology, a plurality of independent switches can generally be divided into chrysanthemum factory structures, such as yttrium, yttrium, topological structure), among which strict aisy chained) or ring (ring; Package 2: ϊ: The architecture provides a redundant stacking link. The first figure shows that ^^^ is ranging from one to a dozen. Each exchange # =! A stack consisting of a switch I, in the basin : It contains ΓΓ stacked end π, where-is the upstream end for two to pass through; ΐ;: =: Γ ", used for downlink connection. For example, the connection of bluji j, ° 卩 knife constitutes a daisy chain topology Structure, in addition: two. Redundant lines of _ 卩 form a ring topology. M mode: as shown in the figure, it is the internal logic of the master switch and the slave switch of the present invention. The master switch Both 10 and slave switch 20 include user interface, / 1 0; service modules 1 2 0, 2 2 0; hardware concept layer modules i 3 0, 2 3 0; drive modules 14 0, 24 0; information Library management module 丨 ", 250 and exchange, and inter-communication modules 160 and 26. Among them, the user interface 11 and 21 are used for parent machine replacement and use. In this invention, the user of the slave switch does not communicate with the user interface 2 0 of the slave switch, and when the slave parent switches to the master switch, the user can use the user interface for management ; Service modules 120 and 220 include multiple applications that can provide corresponding services for the switch; and hardware abstract layer (HAL) 130 and 230 are abstractions of the hardware in the switch 'Is used to support various programs and services in the switch; driver modules (Drivers) 140, 240 are used to drive various hardware in the switch; and the database management module (Database Maintaince Protocol, DMP) 1 5 0 2, 50 is used to manage the database in the switch, and the local cache memory

第7頁 1232659 五、發明說明(4) 的内容及運作;交換機間通訊模組(Inter_Swi tchPage 7 1232659 V. Content and operation of invention description (4); inter-switch communication module (Inter_Swi tch

Comnuimcation,ISC ) 160、2 6 0是用於在主交換機與^六 換機間進行通訊,主交換機的交換機問福 一&又 状城间逋讯板組1 6 0盥從 交換機的交換機間通訊模組2 6 0電遠桩尤 , ” 包逆接在一起,主交換播 可耩之將命令訊息傳送給從交換機,而從交換機將回應 息、事件訊息、統計訊息等藉交換機間通訊模二= 交換機。 、、、、1寻k、,,a主 本發明所提供之堆疊式交換機管理方法, 下部分之步驟·· 王要包括如 (一)建立堆疊 在建\堆疊 ’交換機的帶電 鄰近的交 之前,需確 連接可以會 換機,用堆 行端口連接 線鍵路。g 。之後即可 檢測 交換機中, 發送拓扑檢 最後可獲取 比較所接收 構資訊D如 。如果自身 拓扑檢測包 換機的下 一條堆疊 構的堆疊 拓扑結構 堆疊的各 ’其可以 應資訊、 檢測裝置 身拉扑結 的檢測包 訊附於該 在 測裝置 得其相 寻4石扑 訊與自 接收到 結構資 保所有的父換機處於關機狀態 引起堆疊的失敗;然後藉堆疊 疊線將一交換機的上行端口與 起來,並且確保任何兩交換機 樣即可建立起菊花鏈型或環型 上電啟動該堆疊。 其CPU中都包含有進行拓扑檢 測包給鄰近的交換機,從而獲 整個堆參的拓扑結構資訊。該 的拓扑檢測包中的拓扑結構資 果自身的優先級高,則舍棄所 的優先级你’則將自身的拓扑 中,向鄰述父換機傳送’並回 第8頁 1232659 五、發明說明(5) C:原交換機。並且通過該等交換機所具有的學習 匕敫取、、、彳隻得整個堆疊的拓扑結構資訊。 整個:疊的拓扑結構資訊包括如下部分"隹疊的連接 姑^ ,如莉花鏈型或環型;連接的交換機的數目,·每一交 換機的詳細資訊,包括:交換機的類型(第二層戍;:; ):e^^o^(12G.24G^48G) 咖功率、系統記憶體大小等),交換機的軟體 (三)主交換機推舉 主六i = r輯上,主交換機是整個堆疊的代表,使用者藉由 $ 、機的IP位址來管理堆疊中的其他交換機。 =場過手動!置’或通過-排序機制來自動推舉 位址:Μ制d疋基Γ 些交換機的固有屬性,如交換機的MAC 。(Media Access Contr〇1 Adress )及其中之 奴處於排序清單第一位的交換機被認為是主交'。 發明的另一實施例中,該排序機制可以整合進r U包中,所以…檢測階段即可選擇-交換機 機。二餘之交換機作為從交換 進行石ί 通過一用戶介面和一遠程網絡管理口。 八士互動,接收使用者的需求命令;且可以發送兩 - :η給堆疊中的從交換機;從交換機接收該等;:/命 二讯α後進行相應操作後,將響應訊息回送給主六抬/命 亚且在從交換機上有事件發生時,則其亦機; 爭件訊息 第9頁 1232659 五 發明說明 (6) 傳送=主交換機。 機,·:;;:::機時’則,用主交換機成為新的主交換 從交換機變成:5:換機;=排序清單中最靠前的 進行拓扑檢測操作。 有的父換機,重新 (四)堆疊之管理 使用者對堆疊之管理可以藉由四 控制台、遠程TELNET訪問、遠程網站訪n = f仃· RS232 。其中RS232控制台管理方式,係指利用問1运程SNMP管理 交換機通過-KS232 口相連,對該堆疊制台與該主 種方式均▲遠程管理方^,均需要利 s 1,而後- 址。其中需要注意的是,為了安全考】到=機, 機的管理需要通過主交換機來進行,而對攸父換 棧進仃官理。當主父換機的CPU接收到又換 配置從交換機時,其會發送一包括有 、而,而要 交換機的CPU ’具體的工作過程:制訊心的封包給從 在本發明巾,為了響應使用/的7進行敘述^ 可以從從交換機中獲取資料並# Λ二理需求,主父換機 有效率,主交換機可保持有二=^換機的資料。為更 換機利用該等·資料來直接響is用去2料庫的複製,主交 避免了交換機間的通訊产L二 有的督理需求。這樣就 *主交換機的中時間。一 機的訊息:配置訊息,狀能j 匕 有二種有關攸父換 訊息記錄有使用者對從六=Λ息和統計訊息。其中,配置 的配置情形。狀態訊息是指Comnuimcation (ISC) 160, 2 60 is used for communication between the main switch and the switch, and the switch of the main switch asks the switch between the switchboard and the intercity communication board group 160. The communication module 260 is remotely connected, and the packets are connected together in reverse. The master switch can send the command message to the slave switch, and the slave switch uses the response information, event information, and statistical information to communicate between the switches. = Switch. ,,,, 1k ,,,, a Mainly provide the stacking switch management method provided by the present invention. The steps in the next section include: (1) Establishing a stack of a switch under construction \ stacking. Before the transfer, you need to confirm that the connection can be changed, and use the stack port to connect the line key. G. After that, you can detect the switch, send the topology check, and finally you can get and compare the received structure information D. If your own topology detection is replaced The next stacking structure of the machine's stacking topology is stacked. It can respond to information, and the detection package of the detection device pull knot is attached to the device under test. Received that all parent switches of the structure insurance were turned off and caused the stack to fail; then the stack ports were used to connect the uplink ports of a switch, and to ensure that any two switches could be daisy-chained or ring-shaped. Start the stack. The CPU contains a topology detection packet to the neighboring switch, so as to obtain the topology information of the entire stack parameter. The topology information in the topology detection packet has a high priority, so it is discarded. You ’ll then “transmit it to your neighbor ’s parent machine in its own topology” and return to page 8 1232659 V. Description of the invention (5) C: The original switch. The topological information of the entire stack can only be obtained. The whole: The topological information of the stack includes the following parts " stacked connections, such as Lihua chain type or ring type; the number of connected switches, each switch Detailed information, including: the type of switch (second layer 戍;:;): e ^^ o ^ (12G.24G ^ 48G) coffee power, system memory size, etc.), software of the switch (3) The main switch recommends the main six i = r series. The main switch is the representative of the entire stack. The user manages the other switches in the stack by $, the IP address of the machine. -Sorting mechanism to automatically recommend addresses: M system based on the inherent properties of some switches, such as the MAC of the switch (Media Access Contr0 Adress) and the switch in which the slave is at the top of the list is considered the master In another embodiment of the invention, the sorting mechanism can be integrated into the r U package, so ... the detection phase can be selected-the switch machine. Two switches can be used as a slave switch through a user interface and a remote Network management port. Baishi interaction, receiving user demand commands; and can send two-: η to the slave switches in the stack; slaves receive these;: / After the second message α performs the corresponding operation, the response message is returned to the master six If there is an event on the slave switch, it will also be a chance; the content message, page 9 1232659 Five invention description (6) Transmission = master switch. Machine, ·: ;; :::: machine time ', then use the master switch to become the new master switch. The slave switch becomes: 5: change machine; = the first in the sort list to perform the topology detection operation. Some parents change the machine, and re- (4) stack management Users can manage the stack through four consoles, remote TELNET access, remote website access n = f 仃 RS232. Among them, the RS232 console management mode refers to the use of a 1-way SNMP management switch connected through the -KS232 port. The stacking console and this main method are both remote management parties, and both need to use s 1 and then-address. It should be noted that, for safety reasons, the management of the machine needs to be carried out through the main switch, and it is official to change the stack of the father. When the master-swap CPU changes and receives the configuration slave switch, it will send an included and required CPU of the switch. The specific working process: the packet of the message center is sent to the slave in this invention. / 7 for description ^ You can obtain information from the switch and # Λ second reason requirements, the master and the father are efficient in switching, and the master switch can maintain the data of the two = ^ switching. In order to change the machine, the data is used to directly copy the data in the 2 warehouses, and the main communication avoids the supervision requirements of the communication between the switches. This way * the middle time of the master switch. One machine's message: configuration message, the state can have two kinds of information about the parent exchange, and the message records the user's pair of information and statistical information. Among them, the configuration of the configuration. Status messages are

第]0頁 1232659 五、發明說明(7) 系統的操作 換機會定期 含有一緩衝 取從交換機 衝區中的狀 的計數值, 狀態訊息緩 作情形是: 其獲取相關 獲取統計訊 應訊息,觔 一遠程苟問 存中的統計 當時間超出 之。 下面則 行遠程配置 詳細說明。 如第三 置之流程圖 求對從交換 途集成晶片 狀態訊息, 報告其狀態 區以存儲該 的狀悲’其 態訊息資料 其無須週期 存來減少交 當主交換機 統計訊息並 息的需求時 直接從緩存 以獲取相應 訊息均會設 後,其不再 例如端口的 訊息,主交 等報告,當 會呼叫D Μ P 。而統計訊 更新,但本 換機間通訊 接收第一次 將其存於本 ,則首先檢 中獲取;如 之統計訊息 置一預定時 有效,則會 連接狀態 換機的資 主交換機 •則DMP返 息是每一 發明 的系 統計 地的 索该 沒有 ,另 間表 產生 之實 統開 訊息 緩存 本地 相應 外— 來驗 〆遠 訊息等,從交 料庫管理模組 的HAL需要獲 回存儲於其緩 從交換機提供 施例提供了 一 銷,其大致工 的需求後,則 中,下次如有 緩存,如有相 訊息,則產生 對每次存入緩 證其有效性, 程訪問來更新 結合流程圖分別介紹本發明中,對從交換機進 ,獲取從交換機遠程資料報告的具體過程進行 圖所示,係為本發明之對從交換機進行遠程配 。、中遠程配置是指主交換機按照使用者的需 機進行設置,其一般是改變從交換機中特定用 (AS I C )的設置來實現,例如可以遠程設置從 交換機的端口速率等。當主交換機藉由用戶介面接收到使 用者的一遠程配置命令後,其硬體概念層模組會判斷其為Page] 01232659 V. Description of the invention (7) The operation of the system periodically contains a buffer to take the count value of the state from the switch's punching area. The status information is temporarily delayed: it obtains relevant acquisition statistics, response information, and information. The statistics in a long-range questioning time exceeded. The remote configuration is detailed below. For example, the flow chart of the third set is to integrate the chip status information on the slave exchange, and report its status area to store the status. Its status information data does not need to be stored periodically to reduce the need for statistical information to be exchanged with the master switch. After the corresponding information is obtained from the cache, it is no longer such as the port's message, the master report, etc., and it will call DMP. The statistics are updated, but if the communication between the exchanges is stored for the first time, it will be checked first. If the statistics information is valid when scheduled, it will be connected to the master switch of the exchange state. • DMP The rebate is calculated by the system of each invention. The actual message generated by the other table is cached locally and correspondingly — to verify the remote information, etc., the HAL from the delivery management module needs to be retrieved and stored in It provides a sale from the example provided by the switch. After its approximate work requirements, the next time, if there is a cache, if there is a message, the validity of each deposit is verified, and the program access is updated. In the present invention, the specific process of obtaining a remote data report from a slave switch is illustrated in the present invention with reference to a flowchart, which is a remote configuration of the slave switch according to the present invention. Medium and remote configuration refers to the setting of the master switch according to the user's needs. It is generally implemented by changing the ASIC settings in the slave switch. For example, the port rate of the slave switch can be set remotely. When the master switch receives a remote configuration command from the user through the user interface, its hardware concept layer module will judge it as

第11頁 1232659, 五、發明說明(8) 一遠程操作命令,並呼叫相應的資 式介面(步驟S310) ; %資料庫管=輪組的應用程 芩數的設置命令(步驟S32〇) 六果、'且建立一帶有相應 模組則封包該設置命令,並 ^奐機的父換機間通訊 娜μ ;從交換機的交換機間通;;=給從交換機(步驟 進行解封包的動作(步驟S34〇n=到該封包後並 杈組取出其t的命令和參數並呼=、機的貝料庫管理 用程式介面(步驟S35G); HAL呼應動的硬體^層的應 介面來配置該從交換機的特定 '動曰=二用程式 實現對從交換機的配置,並將:尸曰片(Asi〇 ’以 交換機的Smp (步驟S360 ) 後的狀態訊息返回給從 返回代衝應命令 該響應命令並發送給主交換機(步驟二換機=封包 I s C接收該封包並進行解封包 ,乂 、機 換機的DMP取出1中的動作(步驟S39〇),·主交 s395 ),則t/V二H #代碼,並將其傳送至HAL (步驟 :/對攸父換機進行遠程配置的流程結束。 流程圖°弟1圖ί第五圖所示,係為本發明遠程資料報告的 r"二I 4 2 t貝料報告包括從交換機向主交換機報告其狀 怨訊息和統計訊息。苴中筐同 報止里β 口访4 /、中弟四圖係為從交換機向主交換機 Ϊ二二? 息的流程圖。在本發明的設計中,從交 換機的ML會週期性地收集其端σ的統計訊息(步驟s4i〇 ),仉父換機的DMP建立一包含有統計訊息的報告命令( 步驟S4 2 0 );從交換機的I sc封包該報告命令並將其傳送 給主交換機(步驟S4 3 0 );主交換機的ISC接收該命令封 第〗2頁 1232659 五、發明說明(9) 包後’解開該封包(步驟S4 4 0 );主交換機的DMP將該報 告命令中的統計訊息存儲於H a L的緩衝區中(步驟g 4 $ 〇 ) ,當使用者需要獲取從交換機的統計訊息時,則主交換機 會呼叫HAL的API以獲取統計訊息(步驟S47〇 );則主交換 機的HAL直接從其缓衝區返回所要求的統計訊息(步驟 S480 ) 〇 為本發明中從交換機遠程報告其 如第五圖所示 口狀態訊息的示意圖。首先,從交換機的HA]L會週期性地 收木其纟而口的狀#%说息(步驟S 5 1 0 );從交換機的D Μ P建 立一包含r有相應狀態訊息的報告命令(步驟S 5 2 0 );從交 換機的I S £封包該報告命令並將其傳送給主交換機(步驟 S5 3 0 );主交換機的iSC接收該命令封包後,,解開該封包 (步驟S540 );主交換機的DMP將該報告命令中的統計值 存儲於本地緩衝區中(步驟S55〇);主交換機會週期性地 乎叫H A L的A P I以獲取相應狀態訊息(步驟μ 7 〇 );主交換 機的HAL判斷其為一遠程操作,其呼叫相應DMp的Αρι (步 驟S58G) ;DMP返回存儲於其本地緩衝 步驟S5 9 0 ) 。 τ j狀心几〜 遠程訪問被1sc戶斤封包和同步,立 HAL在ISC返回相應荃數前不會進行處理 :置=所確二機制’ 二未接!/:!重試和時間表機制,當主交換機的 ISC未接到相應回應時’其會重發遠程訪問命+,並真在 1232659_ 五、發明說明(ίο) 幾次重試之後,返回失敗提示給呼叫功能模組。 本發明雖以較佳實施例揭露如上,然其並非用以限定 本發明。任何熟悉此項技藝者,在不脫離本發明之精神和 範圍内,當可做更動與潤飾,因此本發明之保護範圍當視 後附申請專利範圍所界定者為準。Page 111232659, V. Description of the invention (8) A remote operation command, and call the corresponding data interface (step S310);% database management = setting command of the application number of the wheel set (step S32〇) 6 If you set up a command with the corresponding module, set up the command, and communicate between the parent and the switch of the machine; communicate between the switches of the slave switch; = give the slave switch (step to unpack the action (step S34〇n = After the packet is received, the command and parameters of t are taken out by the group and called =, the shell database management program interface (step S35G); HAL echoes the interface of the hardware ^ layer to configure the The specific switch of the slave switch is used to implement the configuration of the slave switch, and the status message after the corpse (Asi0) is sent to the switch's Smp (step S360) is returned to the slave to respond to the response command. Command and send it to the main switch (step 2 exchange = packet I s C receives the packet and decapsulates the packet, 乂, the DMP of the switch exchange takes out the action in step 1 (step S39〇), the main transaction s395), then t / V 二 H #code, and send it to HAL (step: / on the parent The process of changing the device for remote configuration ends. The flow chart is shown in Figure 1. Figure 5 shows the r " II I 2 2 t report of the remote data report of the present invention, which includes reporting the complaint from the switch to the main switch. Information and statistical information. 苴 In the same report, the β interview 4 /, the younger brother 4 is a flow chart from the switch to the master switch. In the design of the present invention, the ML of the slave switch will Periodically collect the statistical information of its terminal σ (step s4i〇), the DMP of the father changing machine creates a report command containing the statistical information (step S4 2 0); the report command is packetized from the switch's I sc and it Transfer to the main switch (step S4 30); the ISC of the main switch receives the command packet; page 2 1232659 5. Description of the invention (9) after the packet 'unpacks the packet (step S4 40); the DMP of the main switch will The statistical information in the report command is stored in the buffer of H a L (step g 4 $ 〇). When the user needs to obtain the statistical information of the slave switch, the master switch will call the HAL API to obtain the statistical information (step S47〇); then the HAL of the master switch The required statistical information is returned from its buffer (step S480). 0 is a schematic diagram of remotely reporting the status information of the port as shown in the fifth figure from the switch in the present invention. First, the HA of the slave switch will receive it periodically. The status of the wooden mouth is #% said (step S 5 1 0); a report command containing the corresponding status information is established from the DP of the switch (step S 5 2 0); the IS £ packet from the switch The report command is transmitted to the main switch (step S5 30); after the iSC of the main switch receives the command packet, it unpacks the packet (step S540); the DMP of the main switch stores the statistical value in the report command In the local buffer (step S55); the main switch periodically calls the HAL API to obtain the corresponding status information (step μ 7 〇); the HAL of the main switch determines that it is a remote operation, and it calls the corresponding DMp Αρι (step S58G); DMP returns to its local buffering step S590). τ j Shaped heart ~ Remote access is packetized and synchronized by 1sc households, and the HAL will not process it until the ISC returns the corresponding number: Set = confirmed second mechanism 'Second missed! / :! Retry and schedule mechanism When the ISC of the main switch does not receive a corresponding response, it will resend the remote access command +, and it will really be 1232659_ V. Invention Description (ίο) After several retries, it returns a failure prompt to the calling function module. Although the present invention is disclosed as above with the preferred embodiments, it is not intended to limit the present invention. Anyone who is familiar with this technology can make changes and retouching without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the scope of the attached patent application.

第14頁 1232659_ 圖式簡單說明 【圖式簡單說明】 第一圖係為一堆疊式交換機示意圖。 第二圖係為本發明主交換機與從交換機之内部邏輯架構示 意圖。 第三圖係為本發明遠程配置流程示意圖。 第四圖係為本發明遠程統計訊息報告的流程圖。 第五圖係為本發明中遠程報告從交換機的狀態訊息示意圖 主要元件標號 堆疊 1 主交換機 10 用戶介® 110 、210 服務模組 120 、220 硬體概念層模組 130 、230 驅動核組 140 、240 資料庫管理模組 150 、250 交換機間通訊模組 160 ^ 260 從交換機 20 ❿Page 14 1232659_ Brief description of the drawings [Simple illustration of the drawings] The first diagram is a schematic diagram of a stacked switch. The second diagram is the schematic diagram of the internal logical architecture of the master switch and the slave switch of the present invention. The third figure is a schematic diagram of the remote configuration process of the present invention. The fourth figure is a flowchart of the remote statistical information report of the present invention. The fifth figure is the status message diagram of the remote report slave switch in the present invention. The main component numbers are stacked. 1 The main switch 10 is the user interface 110, 210 service module 120, 220, the hardware concept layer module 130, 230, and the driving core group 140. 240 database management module 150, 250 inter-switch communication module 160 ^ 260 slave switch 20 ❿

第15頁Page 15

Claims (1)

1232659_ 六、申請專利範圍 1. 一種堆疊式交換機管理方法,其包括如下步驟: (a )將複數可堆疊之交換機按照特定的拓扑結構連接 以建立一堆疊; (b )各交換機向鄰近的交換機發送包括有自身拓扑結 構資訊的拓扑檢測包,以進行拓扑檢測; (c )按照交換機的MAC及優先順序推舉其中一交換機為 主交換機,則其餘之交換機作為從交換機; (d )藉由主交換機對從交換機進行遠程配置; (e )藉由主交換機獲取從交換機的統計訊息; (f )藉由主交換機獲取從交換機的狀態訊息。 2. 如申請> 利範圍第1項所述之堆疊式交換機管理方法, 其中f驟(a )還包括按照菊花鏈型拓扑,結構來建立堆 疊之步驟。 3. 如申請專利範圍第1項所述之堆疊式交換機管理方法, 其中步驟(a )還包括按照環型拓扑結構來建立堆疊之 步驟。 4. 如申請專利範圍第1項所述之堆疊式交換機管理方法, 其中步驟(b)還包括如下步驟: (b 1 )比較所接收的拓扑檢測包中的拓扑結構資訊與自 身拓扑結 >冓資訊; (b 2 )如果自身的優先級高,則舍棄所接收到的檢測包 (b3 )如果自身的優先級低,則將自身的拓扑結構資訊 附於該拓扑檢測包中,向鄰近交換機傳送,並回饋一1232659_ VI. Scope of patent application 1. A stacked switch management method, which includes the following steps: (a) connecting a plurality of stackable switches according to a specific topology to establish a stack; (b) each switch sends to a neighboring switch A topology detection package including its own topology information is used for topology detection; (c) one of the switches is elected as the master switch according to the switch's MAC and priority order, and the remaining switches are used as slave switches; (d) the master switch pair Perform remote configuration from the switch; (e) Obtain statistics information of the slave switch by the master switch; (f) Obtain status information of the slave switch by the master switch. 2. The stacking switch management method as described in the application > Item 1, wherein step f (a) further includes the step of establishing a stack according to a daisy-chain topology and structure. 3. The stacking switch management method according to item 1 of the scope of patent application, wherein step (a) further includes the step of establishing a stack according to the ring topology. 4. The stacked switch management method according to item 1 of the scope of patent application, wherein step (b) further includes the following steps: (b 1) comparing the topology information in the received topology detection packet with its own topology node >冓Information; (b 2) If its own priority is high, discard the received detection packet (b3) If its own priority is low, attach its own topology information to the topology detection packet and send it to the neighbor The switch transmits and returns a 第〗6頁 1232659 --——. 六、申請專利範圍 讯息給原交換機。 a k i利祀圍第3項所述之堆疊式交換機管理方 八Τ步驟(C )還包紅.嗖置—六柏拙从戍s理方法’ 的步驟,吁偌+括. 換機為備用主交換媸 伴持交換機的資料庫需主交換機丄 步,當主交換機出現故障時,則;2的資料庫 成為新的主交換機。 、Λ備用主交換機 並^ ^專利靶圍第3項所述之堆疊式交換機管理方 其中步驟(d )還包括·· 、乂谀械&理方法, 父換機接收到使用的一 念層模組呼叫相】的"斗廑ί配置叩令後’其硬體概 ;^ 應的貝枓庫管理模組的應用程式介面 f料庫管理模組建立一設置命令;, 又換機間通訊模組封包該言^ ^ ^ ^ ^ ^ ^ ^ 換機; 1 P 7 ,亚將其傳送給從交 從父換機的交換機間通 封包; ° 、 接收到該封包後並進行解 從交換機的資料庫管理模組 ^ 叫相應的硬體概念層的應 ^的命令和爹數並呼 硬體概念層模組呼叫驅動^ 二"面, 從交換機的特定用途;=應用程式介面來配置該 資料庫管理模組; 日日,亚將狀態訊息返回給 從交換機的資料庫管理模組 ^ ^ ^ . 命令; 文I —页有返回代碼的響應 從交換機的交換機間通訊握 、Λ松組封包該響應命令並發送給 1232659___ 六、申請專利範圍 主交換機; 主交換機的交換機間通訊模組接收該封包並進行解封包 主交換機資料庫管理模組取出其中的返回代碼,並將其 傳送至硬體概念層模組。 7. 如申請專利範圍第5項所述之堆疊式交換機管理方法, 其中步驟(e)還包括如下步驟: 從交換機的硬體概念層模組會週期性地收集其端口的統 計訊息; 從交換機的資料庫管理模組建立一包含有統計值的報告 命令\ 從交摻機的交換機間通訊模組封包該報告命令並將其傳 送給主交換機; 主交換機的交換機間通訊模組接收該命令封包後並解封 包; 主交換機的資料庫管理模組將該報告命令中的統計訊息 存儲於硬體概念層模組的緩衝區中; 當使用者需要獲取從交換機的統計訊息時,則主交換機 會呼叫硬體概念層模組的應用程式介面以獲取統計值 , 主交換機的硬體概念層模組直接從其緩衝區返回所要求 的統計訊息。 8. 如申請專利範圍第5項所述之堆疊式交換機管理方法, 其中步驟(f )還包括如下步驟:Page 6 1232659 ------. VI. Patent Application Scope Information to the original exchange. The step 8 (C) of the stacking switch management side described in item 3 of Akiri Sakura also includes red. Set-up-Liubaizhuo follow the method of the method 'steps, call 偌 + brackets. Change the machine as a backup master The database of the exchange and supporting switch requires the master switch to step. When the master switch fails, the database of 2 becomes the new master switch. , Λ standby main switch and ^ ^ The stack switch manager described in the third item of the patent target, where step (d) also includes ..., the 乂 谀 machine & management method, the parent switch receives the used mode module Group call phase "after the configuration command is issued, its hardware profile; the application interface of the corresponding database management module; the database management module establishes a setting command; and exchanges communication between machines. The module packet ^ ^ ^ ^ ^ ^ ^ ^ ^ Switching; 1 P 7, Asia will send it to the inter-switch exchange packet from the parent to the switch; °, After receiving the packet and de-slave the switch The database management module ^ is called the corresponding command of the hardware concept layer and the number of calls and calls the hardware concept layer module call driver ^ Second, from the specific use of the switch; = application program interface to configure the Database management module; Day by day, Asia returns the status message to the database management module of the slave switch ^ ^ ^. Command; Text I — The page has a response to the return code. The response command is sent to 1232659_ __ 6. Scope of patent application Main switch; The inter-switch communication module of the main switch receives the packet and decapsulates the packet. The main switch database management module takes out the return code and transmits it to the hardware concept layer module. 7. The stacked switch management method described in item 5 of the scope of patent application, wherein step (e) further includes the following steps: The hardware concept layer module of the slave switch periodically collects statistical information of its ports; the slave switch Database management module to create a report command containing statistical values \ The inter-switch communication module of the interleaving machine encapsulates the report command and transmits it to the main switch; the inter-switch communication module of the main switch receives the command packet And decapsulating the packet; the database management module of the master switch stores the statistical information in the report command in the buffer of the hardware concept layer module; when the user needs to obtain the statistical information of the slave switch, the master switch will The application program interface of the hardware concept layer module is called to obtain the statistical value. The hardware concept layer module of the master switch returns the required statistical information directly from its buffer. 8. The stacking switch management method according to item 5 of the scope of patent application, wherein step (f) further includes the following steps: 第18頁 1232659 、申請專利範圍 從父換機的硬體概念層模組 訊息; 從父換機的資料庫管理模組建立一勺八 的報告命令; 匕含 週期性地收集其端 的狀態 有相應狀態訊息 從交換機的交換機間通訊模組封包該八 送給主交換機; σ °P々並將其傳 主交換機的交換機間通訊模組接收該命令 遠封包; 、匕後,解開 主父換機的資料庫管理模組將該報告命令 儲於本地緩衝區中· 、、、'先计值存Page 18, 1232659. The scope of patent application is from the hardware concept layer module information of the parent machine change. The spoonful report commands are created from the database management module of the parent machine change. The status message is sent from the inter-switch communication module of the switch to the master switch; σ ° P々 and passes it to the inter-switch communication module of the master switch to receive the command remote packet; The database management module stores the report command in a local buffer. 父換機會週期性地呼叫硬體概念層模組的鹿 以獲取相應狀態訊息; 、〜旌式介面 主父換機的硬體概念層模組判斷其為一遠程摂 一叫相應資料庫管理模組的應用程式介面; 其呼 貝料庫管理模組返回存儲於其本地緩衝區中的狀態气幸The parent switch periodically calls the deer of the hardware concept layer module to obtain the corresponding status information; ~ The hardware concept layer module of the master interface switch of the signal interface determines that it is a remote call corresponding database management module Application programming interface; its Hu Bei repository management module returns status stored in its local buffer 9 ·如申請專利範圍第1項所述之堆疊式交換機管理方法, 其中還包括從交換機發送事件訊息給主交換機的步驟。 •如申请專利範圍第6項戶斤述之堆疊式交換機管理方法, $中還包括主交換機交換機間通訊模組重發遠程配置命 令給從交換機之步驟,且在數次重試之後,提示失敗之 步驟。 1 1 ·如申請專利範圍第1項所述之堆疊式交換機管理方法, 其中使用者可以通過R s 2 3 2控制台、遠程T E L N E T訪問、9 · The stacking switch management method according to item 1 of the scope of patent application, further comprising the step of sending an event message from the switch to the master switch. • If the method of stacking switch management described in item 6 of the scope of patent application, $ also includes the step of resending the remote configuration command to the slave switch by the communication module between the master switch and the switch, and after several retries, it prompts failure The steps. 1 1 · The stacking switch management method described in item 1 of the scope of patent application, wherein users can access through R s 2 3 2 console, remote T E L N E T, 第19頁 1232659Page 12 1232659 20頁20 pages
TW93106156A 2004-03-09 2004-03-09 Method for managing a stack of switches TWI232659B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW93106156A TWI232659B (en) 2004-03-09 2004-03-09 Method for managing a stack of switches

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW93106156A TWI232659B (en) 2004-03-09 2004-03-09 Method for managing a stack of switches

Publications (2)

Publication Number Publication Date
TWI232659B true TWI232659B (en) 2005-05-11
TW200531478A TW200531478A (en) 2005-09-16

Family

ID=36320097

Family Applications (1)

Application Number Title Priority Date Filing Date
TW93106156A TWI232659B (en) 2004-03-09 2004-03-09 Method for managing a stack of switches

Country Status (1)

Country Link
TW (1) TWI232659B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI698104B (en) * 2018-08-28 2020-07-01 瑞昱半導體股份有限公司 Method of handling multiple forwarding information base synchronization for network switch stacking system using central learning

Also Published As

Publication number Publication date
TW200531478A (en) 2005-09-16

Similar Documents

Publication Publication Date Title
CN100370740C (en) Management method of stacked switch
CN101371474B (en) System and method for detecting and recovering from virtual switch link failures
EP1435707B1 (en) Automated voice over IP device VLAN-association Setup
CN102263704B (en) Topology construction method and device supporting layer 2 interconnection of data centers
CN103404084B (en) MAC Address forces retransmission unit and method
CN109120418A (en) The distributed multicast of endpoint
CN102938794A (en) Address resolution protocol (ARP) message forwarding method, exchanger and controller
CN101577722A (en) Method for realizing MAC forced forwarding function and device
CN101841432A (en) Port backup method, device and system for service access router
WO2006007789A1 (en) A method for realizing terminals management in the network device
JP2012015997A (en) Topology detection system and relay device
CN101014043B (en) Method for realizing interworking of VoIP service in a same OLT of EPON
CN105471601A (en) Method of managing all stacked slave switches through master switch
CN101009692A (en) Hardware address parsing method and communication processing device, and message processing method
CN102231712B (en) Method for load sharing on encoder and encoder
CN102209035A (en) Traffic forwarding method and devices
CN104579809B (en) The detection method and equipment of a kind of stacking splitting
CN101184032B (en) Stacked system equipment communication method
TWI232659B (en) Method for managing a stack of switches
US7116775B2 (en) Apparatus and method for managing BLF information in an inter-networking system
CN101232510A (en) Method and system for implementing multi-network port device multicast
JP5935859B2 (en) Topology detection system, relay device, and program
JP3794496B2 (en) Network connection method, network connection system, layer 2 switch and management server constituting the same
JP2001144793A (en) High speed/high reliability ether transmission system and i/f device
US20060067327A1 (en) Information distribution system, method and network devices

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees