201023563 六、發明說明： 【發明所屬之技術領域】 本發明為一種環狀網路冗餘檢查之處理方法，尤指透過 複數環網成員發送控制封包的方式，來增加環狀網路於冗餘 檢查時之可靠度，並可達到環狀網路快速回復正常運作之功 效0 【先前技術】 按’資訊科技的日新月異及網際網路技術的蓬勃發展， 人與人之_騎絲便利，透過晴的技術，可使雙方不 受區域範圍的限制而可達成溝通的效果，如此，網路科技的 進步對民眾而言不僅增加了更纽利，對企絲說亦具有更 經濟及效率的生產方式，目前網路主要可區分為廣域網路（ Wide Area Netw〇rk，WAN)與區域網201023563 VI. Description of the Invention: [Technical Field] The present invention is a method for processing a ring network redundancy check, in particular, by adding a control packet to a plurality of ring network members to increase the ring network redundancy. Reliability during inspection and the ability to quickly return to the normal operation of the ring network. 0 [Prior Art] According to the rapid development of information technology and the rapid development of Internet technology, it is convenient for people to ride. The technology can make the two sides not reach the regional limit and achieve the effect of communication. Thus, the advancement of network technology has not only increased the New Zealand for the people, but also has a more economical and efficient production method for the company. Currently, the network can be mainly divided into wide area network (WAN) and regional network (Wide Area Netw〇rk, WAN) and regional network.
(Local Area Network)兩大部份， ★其中區域網路為單—公司、顧、辦公室或校園等較小區域 軌圍内所使料網路系統，而乙太網路（£丨h e ^ n e t )則為目前業界廣範使用的區域網路技術，且基於二二 的傳輸特性，其網路拓樸（T〇pGi()gy)為星狀或樹 狀架構，其帽狀架構之缺點樣鳴的 因此為解決上述乙太網路斷線的問題，便需要在乙太 3 201023563 網路上建立冗餘（Re d u EEE便制定了8 〇2 . 1 n d a n c y)的技術，目前^ D—生成樹協定（(Local Area Network) two parts, ★ The regional network is a single-enterprise, Gu, office or campus, and other small network areas within the network system, and Ethernet (£丨he ^ net It is a regional network technology widely used in the industry, and based on the transmission characteristics of the second and second, its network topology (T〇pGi() gy) is a star-shaped or tree-like architecture, and its hat-like architecture is disadvantageous. Therefore, in order to solve the above problem of Ethernet disconnection, it is necessary to establish redundancy on the Ethernet 3 201023563 network (Re du EEE has developed 8 〇2.1 ndancy) technology, currently ^ D - Spanning Tree agreement(
)’右將生成樹協定運用於多台乙太網路交換器所建構更大 型的網路架構，如網狀（Me sh)，該網路架構上的部份 連線便會暫畴錄料，藉赠為備援路彳t，讓網狀連 線運作如同樹狀連線—般，t網路上的節點或路線故障時， 生成樹協定會重料算後，再卿份先前暫時停止收送資料 的備援路㈣職復射收··絲，如此整個網路便 了重新連成體，此生成樹協定便使乙太網路交換器具備了 網路冗餘的功能。 然而，目前工業控制設備在實際運作上，並不需要建立 車乂為複雜的網狀乙太網路架構，㈣化成環狀網路架構，而 因考量網路減為網狀架構崎立的生成樹協定，因網路斷 線到回復的時間需要較長的時間，便不符所需，之後！E E E便新增了8 02 · 1W-快速生成樹協定（Rap 土 d"Rightly, the spanning tree protocol is applied to a larger network architecture built by multiple Ethernet switches, such as mesh (Me sh), and some connections on the network architecture will be recorded temporarily. The lending is a backup road t, so that the mesh connection operates like a tree connection. When the node or route on the t network is faulty, the spanning tree agreement will be recalculated, and then the remnant is temporarily suspended. The backup road (4) for the data is sent to receive the wire, so that the entire network is reconnected. This spanning tree protocol enables the Ethernet switch to have network redundancy. However, in the actual operation of industrial control equipment, it is not necessary to build a rut into a complex mesh Ethernet architecture, (4) into a ring network architecture, and consider the generation of the network to reduce the formation of the mesh structure. The tree agreement, because the network disconnection to the response time takes a long time, it does not meet the needs, after! E E E added 8 02 · 1W - Rapid Spanning Tree Protocol (Rap soil d
Spanning Tree Protocol，RST P)，此可透過減少連接埠狀態（p 〇 r t s t a t e) 的轉換時間及新增替代埠（Alternate p〇 r t )或備援埠（Backup Port)來減少收斂時間。 然而，因工業控制修復時間分秒必爭的情況下，原本的 生成樹協定及快速生成樹協定並無法滿足工業環境下更快速 4 201023563 的網路斷線回復&力的要求，目前解決的方式為將環狀網路 上的lx連線作為備援的線路，此備援線路只會讓特定的控 制封包通過’-般㈣料封包會被晴而無法通過，使網路 不會形成無’圈’造成資料風暴，*細者為可設定其中 一台為冗餘管理者，並透過狀期發送控㈣包至環狀網路 ’當冗餘管理者再度接收到此封包時，職示目前線路正常 如無，則會啟動備援線路，然而，此檢查方式卻容易受到 ’周路机罝的滿載及網路設備過多，而造成測試封包遺失致使 判斷錯誤的航，此触对，便令冗鮮理者驅使備援線 路不正常的啟動，此時，環狀網路將形成迴路（ L Ο ο p 1 n g)而產生廣播風暴，其表示封包被目標裝置 接收後，仍會永無止境地被循環傳送，使環狀網路形成忙碌 狀態，而無法繼續發送下一個封包，上述，則表示此種測試 方式可靠度較低，若此種冗餘機制運用於企業進行生產、製 造的環狀網路時，便具有發生無法傳輸，機器無法持續作動 之風險，如此對企業的利潤及長期利益都將受到極大影響。 另一方面’在環狀網路的架構下控制封包的傳遞通常需 要靠每一台交換器上的處理器收到前一台交換器送來的控制 封包後再轉送到下一台，然而，透過交換器上的處理器轉送 過程會造成一定時間的延遲，當環狀網路上的交換器越多， 累計的延遲時間將影響到網路斷線時恢復的快速與否。 是以’要如何解決上述習用之問題與缺失，即為從事此 201023563 ( 行業之相關廠商所硫研究改善之方向所在者。 【發明内容】 故’發明人魏於上述之問題與缺失，乃搜餘關資料 ，經由多方評錢考量，並以從事於此行業累積之多年研發 經驗，始設計驗晴騎檢查之處理方法之發明專 利誕生者。 本發明之主要目的乃在於降低環狀網路執行冗餘檢查時 ，戶斤可能發生的錯誤機率，而令企業在使用環狀網路傳輸資料 或訊號時’得以順利執行，其為_複數環網成員定期侧 連接槔之連接狀態，並於發生異常時傳輸第五控制封包至冗 餘管理者，當冗餘管理者接收到該第五控制封包時，則會將 阻斷的連接轉縣導驗態，藉峨_援_，此種利 用複數環網成員傳輸第五控制封包至冗餘管理者的方式為可 • 纽增加·檢查的可靠度，#贿雜鱗於執行傳輸時 ，達到順暢傳輸及$低錯誤機率之功效，且環網成員直接透 過實體層_連接埠的連接狀態是否發生異常之方式，為可 使冗餘管理者即時得知目前環網成員連接埠之連接狀態，進 而快速使環狀網路回復正常運作。 本發明之次要目的在於第—交換ft和第二交換器尚包括 有交換模組及微處，當第—錢^或第二交換器之交換 模組透過連接蟑接收到控制封包時，除了傳輸至本身的微處 理器做處理外，並神會直接傳送至下__台第—交換器或第 6 201023563 二交換器’此方式為可在具有多台交換 器的環狀網路架構下 ，仍有較短的冗餘機Spanning Tree Protocol (RST P), which reduces the convergence time by reducing the transition time of the connection state (p 〇 r t s t a t e) and adding an alternate 埠 (Alternate p〇 r t ) or Backup Port. However, due to the fact that the industrial control repair time is inconsistent, the original spanning tree agreement and the rapid spanning tree agreement cannot meet the requirements of the network disconnection response & force in the industrial environment. The current solution is to The lx connection on the ring network acts as a backup line. This backup line will only allow a specific control packet to pass through the '-- (four) material packet and will be cleared and unable to pass, so that the network will not form a 'circle'. Data storm, * can be set to one of the redundant managers, and through the status of the transmission control (four) package to the ring network 'when the redundant manager receives this packet again, the job is currently normal If it is not, it will start the backup line. However, this inspection method is easy to be affected by the full load of the Zhoulu machine and the excessive network equipment, which causes the test packet to be lost and the judgment is wrong. This touch makes the redundancy The driver drives the backup line to start abnormally. At this time, the ring network will form a loop (L Ο ο p 1 ng) and generate a broadcast storm, which means that the packet will never stop after being received by the target device. The ground is cyclically transmitted, so that the ring network is busy, and the next packet cannot be sent. The above indicates that the reliability of the test mode is low. If the redundancy mechanism is applied to the ring for production and manufacturing. In the case of a network, there is a risk that the transmission will not be possible and the machine will not be able to continue to operate, so that the profits and long-term interests of the enterprise will be greatly affected. On the other hand, the control packet transmission under the ring network architecture usually requires the processor on each switch to receive the control packet sent by the previous switch and then transfer it to the next station. However, The process of transferring the processor through the switch will cause a certain delay. When there are more switches on the ring network, the accumulated delay time will affect the speed of recovery when the network is disconnected. It is based on 'how to solve the above problems and lack of use, that is, to engage in this 201023563 (the relevant manufacturers of the industry to improve the direction of sulfur research. [Invention content] So the inventor Wei in the above problems and lacks, is search Yu Guan information, through the multi-party evaluation of the money, and the years of research and development experience accumulated in this industry, the birth of the invention patent for the design of the inspection method. The main purpose of the invention is to reduce the implementation of the ring network. During the redundancy check, the probability of error may occur, so that the enterprise can smoothly execute the data or signal transmitted by the ring network, which is the connection state of the regular ring connection of the members of the ring network. When the abnormality is transmitted, the fifth control packet is transmitted to the redundant manager. When the redundant manager receives the fifth control packet, the blocked connection is transferred to the county guide state, and the use of the plural is used. The way in which members of the ring network transmit the fifth control packet to the redundant manager is the reliability of the increase and check, and the smoothness of the transmission is achieved when the transmission is performed. The effect of the error probability, and the ring network member directly through the physical layer _ connection 埠 connection state is abnormal, in order to enable the redundant manager to instantly know the connection state of the current ring member connection ,, and then quickly make the ring The network resumes normal operation. The secondary purpose of the present invention is that the first exchange ft and the second switch further include a switching module and a micro-portion, and when the exchange module of the first money or the second switch is received through the connection port When the control packet is sent, in addition to being transferred to its own microprocessor for processing, and God will directly transfer to the next __ station - switch or the sixth 201023563 two switch 'this way is available in multiple switches Under the ring network architecture, there are still shorter redundant machines
【實施方式】 為顧上述目軌魏，树騎_之 Γ，咖就她之崎施例詳加·其特徵與功能 如下，俾利完全瞭解。 刀月b 請參閱第-圖所示’本發明係為一環狀網路架構，包括 有第-交換器1和複數第二交換器2，其中： 該第-父換器1為設有二個以上之連接埠丄i2。 省複數第-父換器2和第一交換器工為透過有線或無線 方式相連’此有線方式包括雙絞線（t w丨s t e d P a i r c a b 1 e )、光纖（f i b e r 〇 p t 土 c c a b 1 e)等，無線方式則為！E E E 8 〇 2 ·工工 或8 0 2 · 16等’而第二交換器2為至少設有二個以上之 連接埠21、2 2。 凊參閱第一、二圖，可由圖中清楚地看出，在穩定狀態 下，每個環狀網路僅會存在有一個冗餘管理者（[Embodiment] In order to take into account the above-mentioned visuals, the tree ride _ Γ Γ, the coffee on her saki example of the addition of its features and functions are as follows, 俾利 fully understand.刀月b Please refer to the figure--the present invention is a ring network architecture, including a first-switch 1 and a plurality of second switches 2, wherein: the first-parent 1 is provided with two More than one connection 埠丄i2. The provincial first-parent converter 2 and the first switcher are connected by wire or wirelessly. 'This wired mode includes twisted pair (tw丨sted P aircab 1 e ), fiber (fiber 〇pt ccab 1 e), etc. The wireless way is! E E E 8 〇 2 · Workers or 8 0 2 · 16 etc. and the second exchanger 2 is provided with at least two or more ports 21, 2 2 .凊Refer to the first and second figures. It can be clearly seen from the figure that in the steady state, there will only be one redundant manager in each ring network (
Ma s t e r )，而可同時具有複數個環網成員（ Member)，因此便需藉由競爭選舉（11111宮 Master Electing)定義第一交換器1和第 二交換器2之冗餘管理者和環網成員的角色，該競爭選舉處 7 201023563 * , 理流程如下： (100) 預設環狀網路上的第一交換Si和第二交換器2 皆為冗餘管理者。 (101) 定義第-交換器i之連接2的連線狀態為阻 斷（Blocking)，且定義該第一交換器 1之連接埠11、第二交換器2之連接埠2工和 參 連接埠22為導通（F〇rwa rdi ng)。 (10 2 )第-交換器1和第二交換器2透過連接埠^工、 連接埠12、連接埠21或連接埠2 2發送第一 控制封包（C F1)至環狀網路。 (10 3)第-父換器1和第二交換器2於預設時間内判斷 是否透過連接埠11、連接埠12、連接埠21 或連接埠2 2接收到第一控制封包，若否，則進 _ 行步驟（1〇 2)’若是，則進行步驟（1〇 4 )° (1〇 4)第-交換器1和第二交換器2分別利用第一控制 封包與本身的預設值相比較，藉以定義第一交換 器1或第二交換n2為冗餘管理者或環網成員。 (10 5)若判斷第一交換器1為冗餘管理者時，第二交換 器2則轉換為環網成員，此時，第二交換器2停 止發送第一控制封包至環狀網路。 (1 06)第—交換Hi定期於預設時勒透過連接璋1 1 201023563 或連接埠12發送第一控制封包至環狀網路。 (1〇 7)第二交換器2判斷第一交換器1是否仍為冗餘管 理者’若是’則進行步驟（1〇 5)，若否，則 進行步驟（1〇 8)。 〇 8)複數第二交換器2轉換為冗餘管理者並透過連接 埠21或連接埠2 2傳輸第一控制封包至環狀網 路，若否，則進行步驟（1〇 7)，若是，則進 行步驟（1〇 9)。 (1 〇 9 )根據第—控觸包及預設值設定其中之一第二交 換器2為冗餘管理者，並定期於預設時間内透過 丄，％乙伸矻乐一控制封包至環 狀網路，其它第二交換器2和第一交換器1則轉 換為環網成員，並停止發送第—控制封包。 的上述’當第二交換器2判斷第—交換器1非冗餘管理者 封方式’如預設時間内並未收到第一交換器傳輸之第一控制 ，包，或使用者定制優先條件改變等情況，此時，各第二 f換器2會轉換為冗餘管理者，而第—控制封包為包括有優 條件（P r ! 〇 r i t y )和網路實體位址㈤Μ · / d r e s s )，職_包財交鋪本細優先條件 和網路實體錄’當交換器㈣第—_封包時則會與本身 的預設值相比較，藉以決定該交換器為冗餘管理者或環網成 員，另，晴（Bl〇eking)之連結狀態為備援線路 201023563 (Backup Path)，而導通（ Forwarding)之連結狀態則為正常線路，且阻斷 之定義為允許預設封包通過之狀態，而導通之定義則為允許 任意封包通過，另步驟（103)、（106)及（1〇9 )之預設時間為設計者自行設定，該預設時間之設定值並非 用以限定本發明之申請專利範圍。 再請參閱第一、三圖，當競爭選舉結束後，若設定其中 σ第一父換器2為冗餘管理者，其餘交換器則為環網成員 ’且維持環狀網路的連線狀態不變，此時將進行迴圈檢查， 其程序如下： (2 0 0)開始。 (2〇1)環網成員之連接埠1丄、連接2、連接埠2 1和連接埠2 2為正常狀態。 (2 0 2)環網成員定期透過連接埠丄i、連接埠i 2、連 接埠21和連接埠22傳輸第二控制封包（CF 2)至冗餘管理者。 (2 0 3)冗餘管理者判斷是否從連接埠2丄和連接埠2 2 收到同一台環網成員傳輸的第二控制封包，若否 ’則進行步驟（2 〇1)，若是，則進行步驟（ 2 0 4) 〇 (2 〇 4)冗餘管理者判定環狀網路形成迴圈，將連接埠2 1設定為導通，連接埠2 2設定為阻斷，並透過 201023563 連接埠21和連接埠22傳輸第三控制封包（c F 3)至環狀網路。 (2 0 5 )環網成員透過複數連接璋收到第三控制封包，便 將阻斷狀態的連接埠轉為導通。 (2 0 6 )結束。 上述’當第二控制封包為包括有環網成員本身的網路實 體位址（Mac address)和發送的連接埠（ ρ 〇 r t)編號。 另請參閱第-、四、五圖’ #辭縣結束後，若設定 第-交換器1為冗餘管理者，第二交換器2為環網成員，且 該第-父換器1之連接埠11的連線狀態為導通，連接缚1 2則為阻斷，其檢查之處理流程如下： (3 0 0)開始。 (3 01)第二交換器2透過實體層定期檢查連接埠2工或 連接埠2 2是否發生異常，若否，則進行步驟（ 300)，若是，則進行步驟（3〇2)。 (3 0 2 )第二交換器2將發生異常的連接槔設定為崎， 透過正常的連接痒傳輸第五控制封包（c F $ ) 至第一交換器。 (3 0 3)第一交換器1透過連接埠11或連接埠12收到 第五控制封包，便將連接蟑12轉為導通狀態， 再從連接谭11和連接埠12傳輸第四控制封包 201023563 至環狀網路。 (3 0 4)第—父換器2透過連接埠21或連接蟑2 2接收 到第四控制封包後，便更新網路實體位址表（Ma ster ), but can have multiple members of the ring at the same time, so it is necessary to define redundant managers and rings of the first switch 1 and the second switch 2 by competitive election (11111 Master Electing) The role of the member, the competition election office 7 201023563 *, the process is as follows: (100) The first switch Si and the second switch 2 on the preset ring network are redundant managers. (101) Defining the connection state of the connection 2 of the first-switch i is blocking, and defining the connection port 11 of the first switch 1, the connection port 2 of the second switch 2, and the reference port. 22 is conductive (F〇rwa rdi ng). (10 2) The first switch 1 and the second switch 2 transmit the first control packet (C F1 ) to the ring network through the connection, the port 12, the port 21 or the port 2 2 . (10 3) The first parent switch 1 and the second switch 2 determine whether the first control packet is received through the connection port 11, the port 12, the port 21 or the port 2 2 within a preset time, and if not, Then, the step (1〇2) is performed. If yes, the step (1〇4) is performed. (1〇4) The first switch 1 and the second switch 2 respectively use the first control packet and its own preset value. In comparison, the first switch 1 or the second switch n2 is defined as a redundant manager or a ring network member. (10) If it is determined that the first switch 1 is a redundant manager, the second switch 2 is converted to a ring network member. At this time, the second switch 2 stops transmitting the first control packet to the ring network. (1 06) The first-switching Hi periodically sends the first control packet to the ring network through the connection 璋1 1 201023563 or the connection port 12 at the preset time. (1〇7) The second switch 2 judges whether the first switch 1 is still a redundant manager 'if yes' to perform the step (1〇 5), and if not, proceeds to the step (1〇 8). 〇8) The plurality of second switches 2 are converted into redundant managers and transmit the first control packet to the ring network through the port 21 or the port 2, and if not, the steps (1〇7) are performed, and if so, Then proceed to step (1〇9). (1 〇9) Set one of the second switch 2 as a redundant manager according to the first control packet and the preset value, and periodically pass the 丄 in the preset time, and the % B 矻 一 一 控制 control packet to the ring The second network switch 2 and the first switch 1 are converted into ring network members and stop transmitting the first control packet. The above-mentioned 'when the second switch 2 determines that the first switch 1 non-redundant manager seal mode' does not receive the first control, packet, or user customization priority condition of the first switch transmission within a preset time. Change the situation, at this time, each second f-changer 2 will be converted into a redundant manager, and the first control packet includes excellent conditions (P r ! 〇rity ) and network entity address (five) Μ · / dress ) , _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The member, and the connection status of Bl〇eking is the backup line 201023563 (Backup Path), and the link state of the Forwarding is the normal line, and the blocking is defined as the state in which the preset packet is allowed to pass. The definition of conduction is to allow any packet to pass, and the preset time of the other steps (103), (106) and (1〇9) is set by the designer, and the setting value of the preset time is not used to limit the present invention. Apply for a patent scope. Please refer to the first and third figures. After the end of the competitive election, if the σ first parent converter 2 is set as the redundant manager, the remaining switches are the ring network members' and maintain the connection state of the ring network. If it does not change, a loop check will be performed at this time. The procedure is as follows: (2 0 0) starts. (2〇1) The connection of the ring member 埠1丄, the connection 2, the connection 埠2 1 and the connection 埠2 2 are normal. (2 0 2) The ring network member periodically transmits the second control packet (CF 2) to the redundant manager through the connection 埠丄i, the connection 埠i 2, the connection port 21 and the port 埠22. (2 0 3) The redundancy manager determines whether the second control packet transmitted by the same ring network member is received from the port 2丄 and the port 2 2, and if not, the step (2 〇 1) is performed, and if so, Step (2 0 4) 〇 (2 〇 4) The redundancy manager determines that the ring network forms a loop, sets the port 1 2 1 to be conductive, and the port 埠 2 2 is set to block, and connects to the port through 201023563. And the port 22 transmits a third control packet (c F 3) to the ring network. (2 0 5) The ring network member receives the third control packet through the complex connection, and turns the connection state of the blocking state into conduction. (2 0 6 ) ends. The above-mentioned 'when the second control packet is a network address including the ring network member itself (Mac address) and the transmitted port number (ρ 〇 r t). Please refer to the figure -, 4, and 5'. After the end of the county, if the first switch 1 is a redundant manager, the second switch 2 is a ring member, and the first parent switch 1 is connected. The connection status of 埠11 is conduction, and the connection is blocked by 1 2, and the inspection process is as follows: (3 0 0) starts. (3 01) The second switch 2 periodically checks whether the connection 埠2 or the connection 埠2 2 is abnormal through the physical layer. If not, the process proceeds to step (300), and if so, the step (3〇2) is performed. (3 0 2 ) The second switch 2 sets the abnormal connection port to the state, and transmits the fifth control packet (c F $ ) to the first switch through the normal connection itching. (3 0 3) The first switch 1 receives the fifth control packet through the connection port 11 or the port 12, and turns the port 12 into a conducting state, and then transmits the fourth control packet from the connection tan 11 and the port 12: 201023563 To the ring network. (3 0 4) The first parent switcher 2 receives the fourth control packet through the port 21 or the port 2 2, and then updates the network entity address table (
Mac address table)。 (3 0 5)結束。 上述，當第一交換器丄接收到第五控制封包時，便會發 送第四控綱包至至雜網路上，通知第二交鋪2變更路 控，並更新其網路實體位址表，且當第二交換器2之連接蜂 21或連接埠22發生異常時，第二交換器2仍會於異常期 間定期檢查連接相連接狀態，#連接埠2 i或連接璋2 2 回復正常時，仍會_使連接埠2 1或連接琿2 2維持在阻 斷的連接狀態’此時’第二交換器2便會定期透過連接缚2 1和連接埠2 2傳輸第二控制封包至第一交換器1，若第一 交換器1透過連接埠11或連接琿12收到同-台第二交換 器2的第二控制封包時’便會將其中—個連接埠設定為阻斷 ，並透過連接蟑11或連接埠12發送第四控制封包，藉以 更新第二交換H 2的網路實體位址表。 另，請參閱第六圖，該第一交換器1或第二交換器2為 可傳送第-控制封包、第二控制封包、第三控制封包、第四 控制封包和第五控制封包’當上述之控制封包被第一交換器 1或第二交換器2的連接i、12、21、22接收時 連接琿11、12、21、2 2便會將控制封包送至第- 12 201023563 交換器1或第二交換器2之交換模組！ 3、23，之後交換 楔組13、2 3便會透過連接埠傳輸至下一台第一交換器丄 或第-父換H 2 ’同時會複製控觸包，並將此複製的控制 封包傳輸至第-交換器1或第二交換器2之微處理器丄4、 2 4根據本發明較佳實施例的環狀網路架構中，所設之交 換器可為2 〇 〇〜3 〇 〇台，然而，冗餘機制的執行為可小 於2 〇亳秒（m i 1 1 i s e c 〇 n d，m s),如此便可 ® 制快輕域援機制，然社述之難實施顺不因此而 偈限本發明之巾請專利細，其交換器亦可具有1〜i 〇 〇 100〜150、30Q〜500等情況，而冗餘機制的 執行時間’亦可隨環狀網路架構所設之交換器多寡而增減。 麵上所述，本發明上述環狀網路冗餘檢查之處理方法於 使用時’為具有下列之優點： _ (一）本發明利用複數環網成員發送控制封包的方式，為可 ㉟加冗餘檢查之可靠度，當雜鹏過大或網路流量 滿载時’冗餘管理者為可透過多台環網成員所傳輸的 控制封包來判斷是否啟動備援線路，此方式為可降低 控制封包遺失的風險，而有效達到環狀網路順利傳輸 之功效。 ()本發明利用環網成員直接透過實體層偵測連接埠的連 接狀態，當連接埠11、12、21、2 2之連接狀 癌發生異常’便可直接透過傳輸第五控制封包予冗餘 13 201023563 官理者，使冗餘管理者可以較快速度得知目前環網成 員連接埠之連接狀態，進而加快環狀網路回復正常運 作之速度。 (三）本發明透過第-交換器i或第二器2之交麵組 13、2 3透過連接埠直接傳送下一台第一交換器1 或第二交換器2的方式，可在具有多台交換器的環狀 網路架構下，仍可以較短的冗餘機制處理時間，藉以 達到快速啟動備援之效能。 綜上所述，本發明環狀網路冗餘檢纟之處理方法於使用 時具有顯著之功效增進’誠符合㈣翻之巾請要件，絲 法提出申請’盼審委早日賜准本案，以保障發明人之辛苦 發明’絲鈞局有任何稽疑，請不吝來函如，發明人定 當竭力配合，實感德便。 ❹ 201023563 I l 【圖式簡單說明】 第一圖係為本發明環狀網路冗餘檢查之處理方法較佳實施 例之示意圖。 第-圖係為本發明環狀網路冗餘檢查之處理方法之競爭選 舉流程圖。 第-圖係為本發明環狀網路冗餘檢查之處理方法之迴圈檢 查流程圖。 ❹ 細81係為本㈣魏鱗冗驗查之處理方法之連接埠 斷線示意圖。 第五圖係為本發明環狀網路冗餘檢查之處理方法之冗餘機 制處理流程圖。 第六圖係為本發明環狀網路冗餘檢查之處理方法之控制封 包傳遞之方塊圖 ® 【衫树符號說明】 1、第一交換器 13、 交換模組 14、 微處理器 11、 連接埠 12、 連接埠 2、第二交換器 21、連接埠 2 2、連接埠 2 3、交換模組 2 4、微處理器 15Mac address table). (3 0 5) ends. In the above, when the first switch receives the fifth control packet, it sends a fourth control packet to the hybrid network, notifies the second traffic 2 to change the road control, and updates its network entity address table. And when an abnormality occurs in the connection bee 21 or the connection port 22 of the second switch 2, the second switch 2 still periodically checks the connection phase connection state during the abnormality, and when the connection port 2 i or the connection port 2 2 returns to normal, Still _ to make the connection 埠 2 1 or the connection 珲 2 2 maintain the blocked connection state 'At this time' the second switch 2 will periodically transmit the second control packet to the first through the connection 2 1 and the connection 埠 2 2 If the first switch 1 receives the second control packet of the second switch 2 through the port 11 or the port 12, the switch 1 will be set to block and pass through. The port 11 or port 12 sends a fourth control packet to update the network entity address table of the second switch H 2 . In addition, referring to the sixth figure, the first switch 1 or the second switch 2 is a transmittable control packet, a second control packet, a third control packet, a fourth control packet, and a fifth control packet. When the control packet is received by the connection i, 12, 21, 22 of the first switch 1 or the second switch 2, the connection port 11, 12, 21, 2 2 will send the control packet to the - 12 201023563 switch 1 Or the switching module of the second switch 2! 3, 23, after which the exchange wedge groups 13, 2 will be transmitted to the next first switch or the first-female change H 2 ' through the connection, and the control packet will be copied and the copied control packet will be transmitted. The microprocessors 4 to 4 of the first switch 1 or the second switch 2, in the ring network architecture according to the preferred embodiment of the present invention, may be provided with a switch of 2 〇〇 to 3 〇〇 Taiwan, however, the implementation of the redundancy mechanism can be less than 2 〇亳 seconds (mi 1 1 isec 〇 nd, ms), so that the system can be quickly and lightly supported, but it is difficult to implement it. The towel of the present invention is patented, and the exchanger can also have the conditions of 1~i 〇〇100~150, 30Q~500, etc., and the execution time of the redundancy mechanism can also be set with the switch of the ring network architecture. Increase or decrease. As described above, the processing method for the above-mentioned ring network redundancy check of the present invention has the following advantages when used: _ (1) The present invention utilizes a method in which a plurality of ring network members send control packets, which can be used for 35 redundancy. The reliability of the remaining checks, when the jigs are too large or the network traffic is full, the redundant manager is able to determine whether to start the backup line through the control packets transmitted by the members of the ring network. This way, the control packet can be reduced. Loss of risk, and effectively achieve the smooth transmission of the ring network. () The present invention uses the members of the ring network to directly detect the connection state of the connection port through the physical layer, and when the connection cell 埠 11, 12, 21, and 2 2 is abnormal, it can directly transmit the fifth control packet to the redundancy. 13 201023563 The administrator can enable the redundant manager to know the connection status of the current ring network member connection, and speed up the normal operation of the ring network. (3) The method of the present invention can directly transmit the next first switch 1 or the second switch 2 through the interface 13 of the first switch i or the second switch 2 through the connection , Under the ring network architecture of the switch, the time can be processed with a shorter redundancy mechanism, so as to achieve the performance of quick start backup. In summary, the method for processing the redundant detection of the ring network of the present invention has a significant effect on the use of the 'sincere conformity (four) to turn over the requirements of the towel, and the silk law to apply for the 'prospective review committee to grant the case as soon as possible To protect the inventor's hard work, 'the Silk Bureau has any doubts, please don't hesitate to write a letter, such as the inventor, I will try my best to cooperate. ❹ 201023563 I l [Simple description of the diagram] The first diagram is a schematic diagram of a preferred embodiment of the method for processing the ring network redundancy check of the present invention. The first figure is a competitive selection flow chart of the processing method for the ring network redundancy check of the present invention. The first figure is a flow chart of the loop inspection of the processing method of the ring network redundancy check of the present invention. ❹ Fine 81 is the connection diagram of the connection method of the (4) Wei scale redundant inspection method. The fifth figure is a redundant processing flowchart of the processing method of the ring network redundancy check of the present invention. The sixth figure is a block diagram of the control packet transmission method for the processing method of the ring network redundancy check of the present invention. 1. [The description of the shirt tree symbol] 1. The first switch 13, the switch module 14, the microprocessor 11, and the connection埠12, port 2, second switch 21, port 2, port 2, switch module 2, microprocessor 15