EP2577917A1

EP2577917A1 - Network and expansion unit and method for operating a network

Info

Publication number: EP2577917A1
Application number: EP11723408.8A
Authority: EP
Inventors: Ralf Beyer; Harald Karl; Michael Wilding
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 2010-05-26
Filing date: 2011-05-23
Publication date: 2013-04-10
Also published as: CN102918809B; EP3024180A1; WO2011147759A1; DE102010029301B4; RU2012157096A; DE102010029301A1; US8959386B2; BR112012029987A2; RU2536370C2; CN102918809A; US20130138996A1

Abstract

The invention relates to a network (3), in particular an Ethernet network, comprising as network elements at least two network components (4A, 4B) that are interconnected by means of a network transmission line (2). According to the invention, at least one expansion unit (1) having two external ports (A, B) is disposed in the network line (2) for extending the scope thereof, wherein the expansion unit (1) forwards a failure of the network transmission line (2) at one of the ports (A, B) thereof to a port (B or A) of the next subsequent network element (network component 4A or 4B). The invention further relates to an expansion unit (1) and to a method for operating a network (3).

Description

description

Network and expansion unit and method for Operator Op ^¬ ben a network

The invention relates to a network, in particular an Ethernet network. As network ^elements , the network comprises at least two network ^components which are interconnected via a network transmission line. Furthermore designate the invention applies an expansion unit for Reichwei ^¬ tenverlängerung the network transmission line and a method for operating a network.

Due to the increasing use of network technologies, which originate from the field of office communication, z. As Ethernet, Token Ring, ATM and others, also in the field of industrial automation wins a low-cost networking of terminals over such network technologies more and more important.

In the network technologies of office communication, networking usually takes place via nodes (so-called hubs, switches), from which star-shaped point-to-point connections are made to the individual terminals. This is disadvantageous for applications in the industrial environment in which often linear structures over long distances of over 100 m are present.

In various applications with lengths over 100 m, a fiber is used as a network or transmission line (in the

Network terminology also link or path called) is not a ^¬ settable. Reasons for this may be connectors within the network transmission line, possibly even with the risk of contamination. Another reason for not using fiber optics is the increasing use of so-called hybrid cables, ie the network transmission line is integrated with other lines in a common cable. These hybrid cables can be between the endpoints the Ethernet transmission outlets and interruptions. Fiber optic cable here have the disadvantage that the Ver ^¬ connection between two optical fibers sensitive damping subject, and is difficult to produce locally.

One application where the use of fiber is also impractical is the Ethernet Train bus. He verbin ^¬ det network components of a zugübergreifenden Ethernetnetz- factory. The network components are located several 100 m apart. The network transmission line is often part of a hybrid cable and has several

Plug connector led.

When using so-called twisted-pair cables, the problem of attenuation and that of signal loss can only be solved by installing an Ethernet repeater or Ethernet switch after 100 m at the latest, which regenerates the transmission signal. Because of their simple structure, repeaters are easy to install, cost-effective and require no parameterization or configuration.

The repeaters become problematic when redundancy is required for the Ethernet transmission medium, the network transmission line. In this case, redundancy protocols such as Rapid Spanning Tree Protocol (RSTP) or MRP (Media

Redundancy Protocol) are used. Common to all protocols is the basic principle that they only activate so many of all existing network connections that all network participants form a topological tree; H. There is exactly one transmission path or path between any two network participants in the network. All other existing redundant network connections are in an inactive or passive mode; H. via them no data transmission takes place. There is only an exchange of messages of the redundancy protocols over these passive ports.

The network components or elements (such as Ethernet switches, network repeaters, network transmission lines), which If you are using a redundant network, you must support the appropriate redundancy protocol. In order to switch as quickly as possible at ei ^¬ nem failure of an active network connection to other network connections, the failure of the network transmission line (also called ether net-Link) detected. After the network transmission line failure, the Ethernet switches reconfigure the network to rebuild the topological tree. Parallel to the reconfiguration of the network transmission line, the integrity of the network connections is monitored by sending cyclic telegrams. However, as these are only sent approximately every 100 ms, fast switching can not be achieved on the basis of these telegrams; they serve only as an emergency solution if data transmission is not possible despite an active network connection (eg errors in an Ethernet switch) -ASIC).

When using Ethernet repeaters or Ethernet switches, there is a problem that if the network link or the link on one port of one of the network components fails, the network link remains active on the other ports of the network components, ie the network link of the failed network link is not routed , This prevents a rapid signaling of the failure of the network transmission line to the next network component with redundancy functionality and thus a fast switching time. The only Lö ^¬ solution that a network component is used with redundancy functionality for the Ethernet switch. However, this is expensive due to the large number of redundant components. In addition, the effort for parameterization and configuration is high. Also, the number of network components in a network with redundancy functionality is limited by the standards for the redundancy protocols.

The invention is therefore based on the object, a generic network according ^¬ and a generic method for operating a network Be ^¬ develop such a way that it once easier and with little effort even with large transmission ^¬ distances quickly and safely allows switching in case of failure of one of the network transmission lines. Furthermore, the invention has the object of providing an expansion insurance unit to provide for network transmission lines of a network which enables rapid switching in the network in case of failure of a Netzwerkübertra ^¬ supply line. With regard to the network, the object is achieved by the features specified in claim 1. With respect to the expansion unit, the object is dung OF INVENTION ^¬ according ge solves ^¬ by the features specified in claim 10. As for the method for operating a network the object is solved by the given on ^¬ in claim 15 features.

Advantageous developments of the invention are the subject of the dependent claims.

A network, in particular an Ethernet network, comprises as network elements at least two network components which are connected to one another via a network transmission line. According to the invention at least one expansion unit is arranged with two external ports in the network transmission line to de- ren range extension, the Extension C ^¬ approximation unit determines a failure of the network transmission line at one of its ports and to a port of the nächstfol ^¬ constricting network element, in particular the next ER- extender unit or the network component, forwards.

In this case, the forwarding of the failure of the network transmission line is carried out until it is received at a port ei ^¬ nes network element redundancy function. By such a forwarding of the failure of the network transmission line or the link to the port of the expansion unit to the port of the next Netzwerkele ^¬ ment, z. As a repeater, another extension a network component, such as a terminal, to a network element with redundancy function allows a rapid switching redundancy protocols. In this case, the failure of the network transmission line is forwarded independently of the redundancy protocol used within the network. Since all redundancy protocols automatically initiate a switch after a failure of a network transmission line, eliminating a complicated parameterization, Configu ^¬ ration or expensive redundancy protocol detection.

A further advantage is that due to the simp ^¬ chen construction of the expansion unit (also Link Extender ge ^¬ Nannt) without redundancy function and independent of the redundancy protocol, the number of network elements is not increased with redundancy function.

In the following, for the sake of clarity, the failure of the network transmission line or network connection is referred to as link failure.

For a simple implementation of the expansion unit and integration of this into the network, the external ports of the expansion unit are interconnected by means of a hard-wired connection ^circuit . This allows the use of expansion units in the network even in small quantities.

For simple parameterization and a particularly flexible and universal connection circuit of the ports, the expansion unit comprises an electronically programmable ^circuit . For example, the expansion unit comprises a ^so-¬-called field programmable gate array (FPGA) or an elec- tronically programmable logic device (EPLD) or a switch coupling element.

A possible embodiment for forwarding the link failure provides that in case of failure of the network transmission ^¬ line at one of the ports of the expansion unit of the other Port of the expansion unit is disabled. Characterized the link failure continues from an expansion unit advances to the Next Tier ^¬ th extension unit, for example, at distances of several kilometers to the next network component or to the next terminal. Such a simple forwarding the link failure has the advantage that it can be executed regardless of the ver ^¬ applied redundancy protocol on the network with network components and without redundancy function and is in compliance with standards. A delay caused by the detection of a link failure can be neglected.

Preferably, the other port of the expansion unit of the ^¬ art can be deactivated, that a physical layer device of the out ^¬ incurred ports switched off or switched through in a sleep mode or the network transmission line by means of the physical layer device is switched.

Alternatively, the extension unit generates at least one warning message at a Linkaus ^¬ case at one of its ports and sends it to downstream ports or receiving

Network elements. Here may persist all other network over ^¬ tragungsleitungen or links. Ie. the warning telegram is sent out only by the expansion unit or the network element which has recognized the link failure.

Conveniently, the extension unit switches at a detected link failure at one of its ports the other port in the opposite direction in a predetermined error code and generates at least a warning telegram and sends it to ports of network elements that are downstream of this port in the opposite direction. As a result, the link failure is also forwarded to network ^¬ elements in the opposite direction and reported. To detect the fault location, ie the beginning of the Linkaus ^¬ case expediently the receiving a warning telegram ^¬ catching network element at least a warning telegram with error position identifier and sends it to ports downstream ter or receiving network elements. The error position identifier is realized, for example, in a simple manner by a distance counter. The warning telegram includes a counter. Starting at the network element or at the extension unit which has first identified a link failure, the counter is incremented by one from each further receiving network element, for example. As a result, the network element receiving the warning message with error position identifier can quickly and reliably determine the location or the error position between the extension units.

The network is expediently configured in such a way that either the deactivation of the other port of the expansion unit or the sending of warning telegrams is carried out. The expansion unit is such forms being ^¬ that can be selected by a simple switching function one of the two functions, Port disabling or warning message dispatch.

For a simple embodiment of the expansion unit, it can be supplied with electrical energy via the network transmission line. Alternatively or additionally, the extension unit itself can be supplied with electrical energy by the terminal itself.

The expansion unit according to the invention for a Reichwei ^¬ tenverlängerung a network transmission line in a network comprises at least two external ports, which are connected to each other via a hard-wired connection circuit such that a failure of the Netzwerkübertragungslei ^¬ tion at one of the ports can be determined and output. Such an expansion unit according to the invention enables a simple implementation of several of these expansion units in a future or already existing network for extending the range of network connections from several hundred meters to several kilometers. For galvanic isolation, the extension unit includes input and output side each data line in each case a separating element, for. B. transformers. For an implementation of the expansion unit which is independent of the redundancy protocol or function of the network, this comprises between the separation elements two physical layer devices (called PHY for short) with an electronically programmable circuit arranged therebetween, eg. As FPGA or EPLD.

To implement the forwarding option of a link failure by sending warning telegrams, the electronically programmable circuit, z. As an FPGA, each data line a telegram generator and a controller. In the event of an identified link failure, the telegram generator automatically generates warning telegrams with or without an error position identifier and sends them via the data line. The controller serves to identify the link failure and the control of the telegram generator and the telegram transmission and the switching of the network transmission line, in particular their data lines.

An alternative embodiment provides that the expansion ^¬ tion unit comprises a switch coupling element (switch ASIC) with microcontroller. This controlled switch ASIC solution of the expansion unit executes a deactivation of the other port of the expansion unit or a generation and transmission of warning telegrams analogous to the FPGA PHY solution for forwarding a link failure at one of the ports.

Conveniently, input upon failure of the power supply and / or a faulty component of the expansion unit and the output side are provided in parallel to each of the data lines of the network transmission line, a bypass line for switching the Netzwerkübertra ^¬ supply line. Switching takes place by controlling a switching element by means of the controller. To detect the failure of one of Components of the expansion unit, this includes at least one error detection. The switching element is arranged on the input side in front of the separating element in each of the data lines and on the output side after the separating element in each of the data lines.

In this way, network availability is increased, since a wide heritage ^¬ drove the network is possible. The bypass circuit is optional and useful for a line network structure.

Embodiments of the invention are described below with reference to drawings. In the drawings: schematically in a block diagram, with partial exploded view of an expansion unit for egg ^¬ ne network transmission line to extend the range in a network without a link failure, schematically in a block diagram, with partial exploded view of the expansion unit according to Figure 1 with a link failure at one of the ports,

schematically one of the expansion unit ge ^¬ Mäss in block diagram in partial exploded view of figure 1 downstream expansion unit with a link failure at the expansion unit according to Figure 1,

schematically in block diagram with partial exploded view an alternative embodiment for an expansion unit with an Ethernet switch without link failure,

schematically in a block diagram with a partial exploded view of the expansion unit of Figure 4 with a link failure, and

schematically one of the expansion unit ge ^¬ Mäss in block diagram in partial exploded view of Figure 4 subsequent expansion unit when a link failure on the expansion unit according to FIG. 4 Corresponding parts and data and functions are provided in all figures with the same reference numerals.

FIG. 1 shows schematically in block form an extension unit 1 for a network transmission line 2. The extension unit 1 is used for range extension in a network 3.

The network 3 may be an Ethernet network or another standard network. It will be in the

Network 3 via the network transmission line 2 with data lines 2.1 and 2.2 (also called links) network components 4A, 4B, such. B. individual devices connected to each other. The network 3 can have a line structure or ring structure with a multiplicity of network components 4A, 4B connected via the data lines 2.1 and 2.2.

By means of the extension unit 1, the Netzwerkübertra ^¬ supply line 2 without transmission losses, damping and signal nalverluste individually, in particular, the maximum over ^¬ tragungsreichweite of the network, eg. For example, in an Ethernet network by 100m, be extended. In particular, a cascade connection of a plurality of expansion units 1 in a network transmission line 2 allows any distance.

Due to the simple construction of the expansion unit 1 described below, this can be subsequently introduced into network transmission lines 2 existing networks 3.

The expansion unit 1 comprises two external ports A and B for connection of the network transmission line 2, which is connected by means of conventional plug 5 with the ports A and B. In addition, the expansion unit 1 includes a not shown forth nä ^¬ internal port. Here, the data lines 2.1 and 2.2 and network components 4A, 4B in the embodiment are such confi ^¬ riert that the data traffic is in the opposite direction. Ie. By means of the data line 2.1, a data connection is made from the network component 4A connected to the port A to the network component 4B connected to the port B. By means of the data line 2.2, the data connection is reversed, ie from the network component 4B connected to the port B to the network component 4A connected to the port A.

For data-related compound of the ports A and B within the expansion unit 1, a hard-wired Verbin ^¬ decision circuit 6 is provided.

The extension unit 1, in particular the connection ^{scarf ¬} device 6 comprises for galvanic isolation port side each data line 2.1 and 2.2 as separators 7 z. B. Transformato ^¬ ren.

For a media-independent interface of the external ports A and B, the data lines 2.1 and 2.2 are connected in a first embodiment via so-called physical layer devices 8 (also called PHY for short) to the external ports A and B by means of the connection circuit 6. The physical layer de ^¬ vices 8 are for data transmission to enable physical data connections between the network components 4A, 4B, disable, maintain and transmit data.

Between the physical layer devices 8, an electronically programmable circuit 9 is connected in the connection ^¬ circuit 6. The electronically programmable circuit 9 is also shown in detail in Figure 1 in exploded view with dashed frame closer. The electronically programmable circuit 9 can be designed as so-called field programmable ge ^¬ gat (FPGA) or as Electronically Programmable Logic Device (EPLD). Alternatively, however, the circuit 9 can also be designed as a permanently programmed integrated circuit, in particular as an ASIC with an additional microcontroller. Such an electronically programmable circuit 9 represents a simple parameterization, flexible and universal configuration and control of the data traffic as well as a simple data and port as well as link monitoring.

The electronically programmable circuit 9 is formed from ^¬ such that this failure of the network transmission ^¬ line 2, also called link failure, determined at one of its external ports A or B and to an external port B or A of the next network element, eg. B. the network component 4B and 4A, forwards.

For this purpose, the electronically programmable circuit 9 per port A and B comprises a controller 10A or 10B and a telegram generator I IA or I IB for generating telegrams D. The respective controller 10A and 10B monitors the data connection between port A and port B. based on an active or inactive data signal DSA or DSB on their state.

In the exemplary embodiment according to FIG. 1, the data connection between ports A and B is active, ie the data signals DSA and DSB of the physical layer device 8 have the status "A = 1" or "B = 1". The electronically programmable circuit 9 exceeds the data telegrams carries D on the data lines 2.1 and 2.2 with a delay, since the electronically programmable circuit 9 acts as a buffer to compensate for slightly different ^¬ Liche clock frequencies of the two physical layer device. 8 The delay is about a few 100 ns.

Depending on the type of data transmission - via Layer 1 or Layer 2 of the ISO / OSI reference model - is at each transmission a tentelegramm Since ^¬ D with newly generated preamble or a further derive received data bytes on Layer 1. Since ^¬ when neither the data telegrams D nor the weitergelei ^¬ ended data bytes checked. In particular, no cyclic redundancy check (English short CRC) is performed.

In the case of failure of the power supply of the expansion unit 1 or in case of detection of a failure of one of the components, e.g. As the circuit 9, the expansion unit 1, the incoming data lines 2.1 and 2.2 of the network line 2 are switched. For this purpose parallel to the connecting circuit 6 comprises the Extension C ^¬ approximation unit 1, two bypass lines 12.1 and 12.2. For switching the data lines 2.1 and 2.2 on the bypass lines 12.1 and 12.2, the extension unit 1 comprises two switching elements 13, z. B. relay contacts. These are correspondingly controlled in the event of a power failure or in the event of an identified component fault, so that the data lines 2.1 and 2.2 are switched over to the bypass lines 12.1 and 12.2, respectively. 2 shows a block diagram of the expansion unit 1 according to FIG. 1 with link failure at the port A. The data signal DSA of the physical layer device 8 of port A delivers the state "A = 0". a so-called panic mode.

Depending on the nature and design of the panic mode, one of the following two measures can be performed: deactivation of the other port B of the expansion unit 1 or

Activation and / or control of the telegram generator IIA for the cyclical generation of warning telegrams W (also called panic messages) instead of the data telegrams D.

On the basis of the deactivated port B or the reception of the warning telegrams W, the link failure at the ports A and B of the detected downstream network component 4A and 4B. If several expansion units 1 are integrated in the network ^transmission line 2, the recognition and reporting or forwarding of a link failure corresponding to the delay of the respective expansion unit 1 is delayed.

In detail, to deactivate the port B, for example, the respective physical layer device 8 of port A of the link ^{failure is} switched off or put into a sleep mode.

Such forwarding of the link failure by Deaktivie ^¬ tion of the other external port B to port A of the next network component 4A is independent of a data and / or redundancy protocol of the network 3. disadvantage is that the generation or the fault of the link failure not determined can be.

To identify the origin of the link failure Koen ^¬ NEN, the invention provides to produce, instead of the deactivation of the other external port B, the warning messages W and send out.

By means of the generated warning telegrams W, which are sent from port A to port B or from port B to port A of the next following network component 4A or 4B, it is possible to transmit a position identifier for determining the point of origin. For this purpose, the warning telegrams comprise W as Positi ^¬ onskennung z. B. Distance data "Dist 0" for distance equal to zero, ie the link failure has occurred for the first time at expansion unit 1. For failure detection, the warning message W includes error message data "LF" for detecting a link failure.

To report the link failure also in the opposite direction of the data connection, ie from port B to port A, the controller 10B of the port B also switches to the error mode and he ^¬ generated warning telegrams W with the distance data "Dist 0" and the error message data "FEFI" , The warning telegrams W are used both by the controller 10A and the controller 10B, for example, as Ethernet telegrams with IP multicast Ethernet addresses, eg. 01-80-C2- xx.xx.xx) as the destination address. Each expansion unit 1 has in the network 3 its own Ethernet address with source address. Such generated IP multicast telegrams are not forwarded by designed as a switch network elements.

Figure 3 schematically shows in block diagram the expansion unit 1 according to Figure 1 downstream expansion ^¬ unit 1 'at a link failure on the expansion unit 1 according to Figure 1. The downstream expansion unit 1' receives the warning ^¬ telegrams W with the distance data "Dist 0" and the feh ^¬ lermeldungsdaten "LF" for link failure. Both left signals or data signals DSA and DSB are active with "A = 1" and "B = 1". The controller 10A receiving the warning telegrams W switches the port A into an error forwarding mode ("panic forward

The received distance data "Dist 0" are increased by the value "1" and the transmitted warning message W is forwarded with the distance data "Dist 1" and error message data "LF" to the next network element normal operation.

If there is a data link DSA with the status "A = 1" on port A for the detection of a valid link or a faultless data connection, then a predetermined number of warning messages W with status "LinkOK" will be sent. Subsequently, by means of the controller 9, the extension ^¬ unit 1, and thus the message generators IIA, IIB, and the controller 10A, is set in the normal operation mode 10B. Ie. received data telegrams D or data bytes are forwarded with some delay by means of the expansion unit 1. The expansion unit 1 switches to the normal operating mode when the data signals DSA and DSB are active and no more warning telegrams W are generated. In this case the link failure from port A of the Erweiterungsein ^¬ uniform 1 in the network 3 is as long as from the network element passed to the next network element to this, a network element with a redundancy function. This network element with redundancy function automatically switches the network connection to an error-free data line.

FIG. 4 schematically shows, in block form, an alternative embodiment for an expansion unit 1 "with a controlled switch coupling element 14 (also called Ethernet switch) without link failure.

In contrast to the expansion units 1 and 1 ', instead of the physical layer device 8 and the electronically programmable circuit 9, the switch coupling element 14 controlled by means of a microcontroller 15 is provided. The control of the data traffic between the external ports A and B via the data lines 2.1 and 2.2 by means of the switch coupling element 14 controlled by the microcontroller 15 via an internal port C. In this case, the microcontroller 15 via an internal port I to the switch coupling element fourteenth switched.

The switch coupling element 14 is a conventional Ethernet switch without redundancy functionality.

In normal mode, data telegrams D from port A to

Port B transferred. The microcontroller 15 continuously monitors the link status based on port-related status registers 16A, 16B, d. H. the data connection of the network line 2 and its data lines 2.1, 2.2 at the external ports A and B.

The switch coupling element 14 is designed such that the data transmission between the external ports A and C and the internal port C can be locked individually. FIG. 5 schematically shows in block form the expansion unit 1 "according to FIG. 4 with a link failure on port A. The microcontroller 15 detects the link failure at port A and blocks the data or cross traffic from port A to port B and vice versa. Ie. All data telegrams D are processed via the microcontroller 15 controlled internal port C. Port B of expansion unit 1 is deactivated.

As an alternative to port deactivation of the microcontroller generates ler 15 analogously to the embodiment according to Figures 1 to 4 warning messages W as so-called panic-telegrams with Entfer ^¬ planning data "Dist 0" and error data "LF" in the path with a link failure or "FEFI" in the path in opposite direction.

6 shows schematically in a block diagram of the expansion unit 1 '' according to Figure 4 downstream Extension C ^¬ approximation unit 1 '''at a link failure at the port A at the ER extender unit 1''according to FIG. 4

To the embodiment according to Figures 1 to 4 ^¬ which the received at port A warning messages of the microcontroller 15 W of the upstream expansion unit 1 'received analog'. The microcontroller 9 blocks direct data traffic from port A to port B and vice versa. The data message ^¬ programs D will now be processed only on the internal port C by the microcontroller 15th The microcontroller 15 sends out warning telegrams W with changed distance data "Dist 1" and the same error message data "LF", ie based on the distance data 1 "Dist 1" increased by 1, the place of origin of the link failure can be identified.

No data telegrams D are sent on port A, but the link remains active for forwarding the warning telegrams W (A = l, B = l). Alternatively, the warning telegrams W are not forwarded because they have an IP multicast address. In this case, the microcontroller 15 detects a link failure and reports it.

In a redundancy unit of the network elements, the warning message functions are analogous reali ^¬ carbonized at a link failure, so that the link failure and its origin is detected ^¬ au tomatically.

The expansion unit 1 can be designed for both a Datenübertra ^¬ supply at Layer 1 and Layer 2 of the ISO / OSI reference model. The possibilities described here of forwarding a link failure by deactivating ports, generating warning telegrams W are identical.

Claims

claims

1. Network (3), in particular Ethernet network, comprising as network elements at least two network components (4A, 4B) which are interconnected via a network transmission line (2), characterized in that in the network transmission line (2) to their range extension at least one extension unit (1) is arranged with two external ports (A, B), wherein the extension unit (1) a failure of the network transmission line (2) at one of its external ports (A or B) to an external port (B or A) of next network segment (network component 4B or 4A).

2. Network according to claim 1,

characterized in that the external ports (A, B) of the expansion unit (1) are interconnected by means of a hard-wired connection ^circuit (6).

3. Network according to claim 1,

characterized in that the extension unit (1) is designed as an electronically programmable circuit (9).

4. Network according to one of the preceding claims, characterized in that in case of failure of Netzübertra ^¬ supply line (2) at one of the ports (A or B) of the expansion unit (1) of the other port (B or A) can be deactivated.

5. Network according to any one of the preceding claims, characterized in that, if the Netzwerkübertra ^¬ supply line (2) at one of the extension unit (1) of the ports (A or B) the other port (B or A) is such deactivated vierbar, that a physical layer device (8) is switched off the ausgefal ^¬ lenen ports (a, B) or switched into a sleep mode.

6. Network according to any one of the preceding claims, characterized in that, if the Netzwerkübertra ^¬ supply line (2) at one of the extension unit of the ports (A or B) (1) said at least one warning message (W) testifies ER and to external ports (B or A) downstream network elements (network component 4A or 4B).

7. Network according to one of the preceding claims, characterized in that in case of failure of the network transmission line (2) at one of the ports (A or B) of the expansion unit (1) this the other port (B or A) in the opposite direction in a predetermined Error mode switches and at least one warning telegram (W) is generated and sent to ports of network elements (network component 4A or 4B) that are downstream of this port (B or A) in the opposite direction.

8. Network according to claim 6 or 7,

characterized in that the network element (network component 4B) receiving a warning telegram (W) generates at least one warning telegram (W) with an error position identifier and transmits it to ports of downstream network elements.

9. Network according to one of the preceding claims, characterized in that the extension unit (1) via the network transmission line (2) can be supplied with electrical energy.

10. Extension unit (1) for a range extension of a network transmission line (2) in a network (3), characterized by at least two ports (A, B), which are connected via a hard-wired connection circuit (6) so mitein ^¬ other that a Failure of Netzwerkübertra ^¬ supply line (2) at one of the ports (A or B) can be determined and output.

11. Expansion unit according to claim 10,

characterized in that input and output side each because a separating element (7) in data lines (2.1, 2.2) of the network line (2) are arranged.

12. Expansion unit according to claim 10 or 11,

characterized in that between the separating elements (7) has two Physical Layer devices (8) are connected with an intervening angeord ^¬ Neten electronic programmable circuit (9).

13. Expansion unit according to claim 12,

characterized in that the electronically programmable circuit (9) per data line (2.1, 2.2) comprises a telegram generator (IIA, IIB) and a controller (10A, 10B).

14. Extension unit according to one of claims 10 to 13, characterized in that the input and output side pa ^¬ rallel to each of the data lines (2.1, 2.2), a bypass line (12.1, 12.2) is arranged.

15. Expansion unit according to claim 14,

characterized in that input and output side between the respective data line (2.1, 2.2) and the associated bypass line (12.1, 12.2), a switching element (13) is arranged.

16. A method for operating a network (3), insbesonde ^¬ re Ethernet network, comprising as network elements at ^¬ least two network components (4A, 4B) which are interconnected via a network transmission line (2), characterized in that at least one in the network ^¬ transmission line (2) arranged to extend range extension unit (1) with two external ports (A, B) is designed such that this failure of the network ^¬ werkübertragungsleitung (2) at one of its ports (A or B) to a external port (B or A) of the next network element (network component 4B).