METHOD FOR DETERMINING A NETWORK PARTICIPANT'S CHAIN ​​POSITION

Description

[0001] The invention relates to a method for determining the chain position of a network participant in an Ethernet network, wherein the network participants are connected to each other in a daisy chain and the daisy chain is coupled at both ends to a port of an Ethernet gateway. The invention further relates to an arrangement of agricultural equipment with network participants suitable for carrying out the method.

[0002] A star topology is commonly used in Ethernet networks, where multiple network devices are connected in a star topology, each to a port of a gateway. This topology allows for a simple mapping of network device identifiers, such as their assigned IP (Internet Protocol) address or their inherent MAC (Media Access Control) address, to the port (connection) to which the network device is connected. Such a query is particularly relevant when a large number of identical network devices are present in a system.

[0003] Agricultural facilities, such as milking parlors, often have a large number of initially similar network participants. A milking parlor example is a rotary milking parlor, which can have a large number of identical milking stalls arranged around its perimeter. Each stall's control unit is equipped with an Ethernet connection for monitoring and / or control purposes. A stall number is assigned to each stall. If the mapping between stall number and gateway port in a star topology is known from a wiring diagram, an identification number, such as the IP address and / or MAC address, can be easily determined for each stall.

[0004] In addition to the star topology, a daisy chain topology is now also used in Ethernet networks. Suitable network devices have two Ethernet ports, which connect them to their neighbors and, at the end of the chain, to a port on a gateway. The resulting ring structure offers the advantages of potentially simpler cabling between network devices and also a degree of redundancy, since at least one interruption of a connection can be compensated for by the fact that the network devices can be reached from different directions.

[0005] Daisy chain topology can also be used in agricultural facilities, such as the previously described milking carousel or other milking parlors, i.e., arrangements of milking stalls, particularly to reduce wiring effort.

[0006] Even with a daisy chain topology, it's easy to compile a list of network participants' identification numbers, such as IP and / or MAC addresses. However, this no longer provides information about the position of each network participant within the daisy chain. Similarly, it's no longer possible to easily map stall numbers on a milking parlor to identification numbers.

[0007] Instead, during installation, the network participant's identification number must be read locally from the network participant itself or determined via a directly connected service computer, and then assigned to the corresponding location number in an assignment table. This process could be simplified if it were possible to automatically determine which network participant occupies which position within the daisy chain. In that case, the corresponding table assigning location numbers to network participants could be generated from the existing wiring diagram.

[0008] From publication US 2020 / 0 162 285 A1, a method for determining the arrangement of Ethernet network participants connected in a daisy-chain ring is known. First, all network participants are identified using a broadcast query, and their orientation relative to an open port of a gateway is recorded. Subsequently, ports of the network participants are disabled at arbitrary points, and another broadcast query is used to identify the point of interruption in the ring. Thus, a broadcast query is required to determine the orientation of the participants, followed by another broadcast query for each participant to identify their position within the ring. Especially with larger rings, this method generates significant network traffic due to the numerous broadcast queries and the correspondingly large number of responses.

[0009] The publication EP 3 819 732 A1 describes a doubly cascaded ring arrangement of Ethernet devices, each with four ports, for coupling in the illustrated topology. To determine the topology, a packet is sent from a gateway through the network, and each network participant appends its specific address to the packet. After traversing the network, the packet is received back by the gateway, and the added entries are evaluated to derive the topology. This method can be performed with very little data overhead but requires that packets can be appended by individual participants and forwarded to the next participant.

[0010] It is an object of the present invention to provide a method for determining the chain position of a network participant within an Ethernet daisy chain topology, which can be performed with minimal additional network traffic. It is a further object to provide an agricultural arrangement with a plurality of network participants with which this method can be carried out.

[0011] This task is solved by a process or agricultural facility with the characteristics of the respective independent claim. Advantageous design and further development are the subject of the dependent claims.

[0012] A method according to the invention of the type mentioned at the outset is characterized by the following steps: In a first step (a), one of the two ports of the Ethernet gateway is deactivated, and then in a step (b), a status message from one of the network participants is recorded, which concerns a port of the network participant in question that, as a result of step (a), has no active connection. From the status message, an identification number of the network participant in question is extracted in a step (c) and assigned to a first chain position. In a step (d), the other port of the network participant in question is then deactivated.Subsequently, steps (b) to (d) are repeated until all network participants are identified by their identifier, with subsequent chain positions, in particular ascending or descending positions, being assigned in the repetitions of step (c), and with the status message in step (b) relating to a port of the network participant in question which does not have an active connection as a result of step (d).

[0013] In order to compensate for the failure of a network line in an Ethernet daisy chain topology, the network participants have the ability to detect a port that is not connected to an active port of another network participant and report this to the switch. In a method according to the invention, the network participants are further configured to deactivate a port upon request. The method according to the invention then uses these capabilities to query information about the positions of the network participants within the daisy chain by selectively deactivating ports. The chain is successively shortened from one end to the other by deactivating ports and thus "making invisible" network participants, and the network participant located at the end of the chain can be identified.Because the port deactivation is only temporary, the entire chain can be reactivated after all network participants have been identified and is then ready for regular network operation again.

[0014] The gateway and network participants are directly involved in carrying out the process. The process can be coordinated by a controller within the gateway itself, for example, upon request from a higher-level controller or by a service computer connected to the gateway via a network. It is understood that the higher-level controller or the service computer can be connected to the gateway via an intranet or the internet. The process can also be coordinated by an external entity, such as the higher-level controller or the service computer.

[0015] In an advantageous embodiment, the deactivation of a network participant's port can be reversed by the network participant itself after a predetermined time following the deactivation of the port, e.g. by the network participant starting a timer after deactivating a port, after which the port is reactivated.

[0016] In a further embodiment, it can be provided that the deactivation of a port by the network participant is immediately terminated when it detects that a connection to another network participant or the Ethernet gateway has been re-established on its other port. In this case, the entire daisy chain can be successively reactivated without waiting times due to the expiration of individual timers by reactivating the previously deactivated port of the Ethernet gateway.

[0017] In a further advantageous embodiment of the procedure, the actions in steps (b) and / or (d) can be performed independently by the network participant in question. Alternatively, the status message in step (b) and / or the deactivation of the port in step (d) can also be performed by the network participant upon request from the gateway or an externally connected entity.

[0018] As a result of this process, a table can be created, for example by the Ethernet gateway or an external unit, that uniquely identifies network participants and their positions within the daisy chain, starting from the initially deactivated port. Using a predefined wiring diagram of the setup, the identification numbers of the network participants can then be uniquely assigned to the actual network participants.

[0019] An arrangement according to the invention, comprising agricultural devices, each having at least one network interface connected as a network participant to a gateway in an Ethernet daisy chain topology, is configured to execute the aforementioned method. Advantageously, the arrangement comprises a plurality of identical devices and is, for example, an arrangement of a plurality of milking stalls, in particular a milking carousel. The advantages described in connection with the method result.

[0020] The invention is explained in more detail below with reference to exemplary embodiments and the accompanying figures. The figures show: Fig. 1 a schematic representation of an arrangement of network participants and a gateway in an Ethernet daisy chain topology; Figs. 2a-2e each a schematic representation of the arrangement according to Fig. 1in various operating states during the execution of a method according to the invention; Figs. 3-4 each show a schematic representation of an arrangement of a plurality of milking stalls with network participants in a milking parlor; and Fig. 5 shows a schematic representation of an arrangement of a plurality of milking stalls with network participants in several milking parlors.

[0021] In Figure 1 The diagram shows an exemplary arrangement of several devices, here four as examples, 2.1-2.4, which are coupled in a daisy chain topology with a gateway 1.

[0022] Gateway 1 can be, for example, an Ethernet switch or other network device capable of connecting the daisy chain of devices 2.1-2.4 to a higher-level network. Gateway 1 itself can be configured to coordinate the procedure described below. The execution of this procedure can be initiated by a higher-level unit connected to Gateway 1 via the network, such as a milking parlor control unit. Alternatively, the procedure can be coordinated by the higher-level unit itself.

[0023] Gateway 1 has two ports (network connections) 11a and 11b, which are connected to the ends of the daisy chain. Within the daisy chain itself, devices 2.1-2.4 are arranged serially. Each of these devices 2.1-2.4 has a network interface 21.1-21.4, which, like Gateway 1, provides two ports connected to the daisy chain. Figure 1The ports are designated with the reference numbers 22.1a-22.4a for a first port and 22.1b-22.4b for a second port. Ports with the index "a" are each connected to the port with the index "b" of their neighbor via network connections 3, resulting in the loop arrangement shown in Figure 1.

[0024] In the arrangement shown, for example, devices 2.1-2.4 are milking stations of a milking parlor, whose control devices have the network connections shown for monitoring and / or control purposes.

[0025] After connecting the components shown, each network participant (21.1-21.4) is assigned an IP address, which is displayed in the Figure 1 is identified by the reference number 23.1-23.4. Furthermore, each network participant 21.1-21.4 has a permanently assigned MAC address 24.1-24.4, which is also in the Figure 1As shown. After the components are wired, gateway 1 knows how many network participants 21.1-21.4 are present and what IP address 23.1-23.4 or MAC address 24.1-24.4 each of them has.

[0026] However, Gateway 1 or the higher-level unit does not know the order in which these detected network participants 21.1-21.4 are located within the daisy chain.

[0027] The example IP addresses 23.1-23.4 are given in the Figure 1 This indicates, for example, that these addresses are not assigned successively and in ascending order in the sequence of the network participants 21.1-21.4 within the chain, but rather that this assignment is made rather randomly.

[0028] In the operation of the in Figure 1In the arrangement shown, a break is typically deliberately created at one point in the loop by deactivating one of ports 11a, b or 22.1a, b - 22.4a, b, so that each network participant 21.1-21.4 is only reachable from one direction to prevent data collisions. For this purpose, the technology RSTP (Rapid Spanning Tree Protocol), well-known in Ethernet networks, can be used, which automatically creates a tree-like network structure during operation. If a network connection 3 actually fails or one of the aforementioned ports 11a, b or 22.1a, b - 22.4a, b becomes non-functional, the previously deactivated port is automatically reactivated, so that the actual fault location represents the break point in the loop. A fault location can thus be compensated for.

[0029] A method according to the invention for detecting the position of a network participant within the daisy chain is described below using the following example: Fig. 2a-2e explained. These figures each show what was already explained in Figure 1 Block diagram shown in different process steps at different states of the involved ports 11a, b and 22.1a, b - 22.4a, b.

[0030] The status of ports 11a, b and 22.1a, b - 22.4a, b is indicated by the symbols "0", "1+", and "1-" next to each port. The symbol "0" indicates a disabled port. The symbol "1+" indicates an enabled port that recognizes it is connected to an enabled and functioning port of a neighboring network device or the gateway. The symbol "1-" indicates an enabled port that recognizes no connection to an active and functioning port of a neighboring network device.

[0031] In a preparatory step of the process, Gateway 1 itself or the higher-level entity can send a command via Gateway 1 into the daisy chain, signaling the start of the process and, if necessary, placing network participants 21.1-21.4 into an initialization mode. In response, network participants 21.1-21.4 acknowledge receipt and send their current IP address 23.1-23.4 and their MAC address 24.1-24.4 back to Gateway 1 or the higher-level entity. If a device table has not yet been created, this is done now by storing the number of network participants 21.1-21.4 and the mapping of IP addresses 23.1-23.4 to MAC addresses 24.1-24.4.

[0032] In a first procedural step (a), gateway 1 deactivates one of the two ports 11a, 11b, for example, the first port 11a. This is in the Figure 2aThis is represented by the symbol "0" on port 11a of gateway 1. As a result, network participant 21.1 recognizes that its port 22.1b has no active connection and sets its status to "1-". All other ports in the system have the status "1+".

[0033] In the next step (b), network participant 2.1, which has detected the status change on its port 22.1b, sends this status change to gateway 1 (or the higher-level entity). This informs gateway 1 that device 2.1, with network participant 21.1 and the IP and MAC addresses 23.1 and 24.1 transmitted with the status report, is the device directly connected to port 11a, i.e., the device with position number 1 in the daisy chain. Network participants 21.1-21.4 can be configured to automatically send a corresponding status report to gateway 1 or the higher-level entity when they detect a status change on one of their ports 22.1a, b - 22.4a, b. Alternatively, it can be provided that Gateway 1 or the higher-level unit sends a request into the daisy chain and asks the network participants 21.1-21.4 to report the status of their ports 22.1a, b - 22.4a, b.

[0034] In a next step (c), network participant 21.1 deactivates the previously active port 22.1a. The resulting state is described in the Figure 2bThe deactivation can occur either after an immediate request from Gateway 1 or the higher-level entity in a direct message to network participant 21.1, since its IP address 23.1 is known after receiving the status message. Alternatively, Gateway 1 or the higher-level entity can send a request to all network participants 21.1-21.4, instructing those network participants 21.1-21.4 with one port in the state "1-" to deactivate their other port. Sending such a request would also cause network participant 21.1 to deactivate port 22.1a. Another alternative is that a network participant 21.1-21.4 in initialization mode could automatically deactivate its other port after detecting the unconnected port with the status "-1" and sending the corresponding status message.

[0035] After this step, step (b) is repeated, namely that the deactivated port – now port 22.1a of network participant 21.1 – is recognized by network participant 21.2. In the Figure 2b This is indicated by the "1-" state on port 22.2b. Network participant 21.2 then sends the corresponding message, indicating a connection interruption on port 22.2b, to gateway 1 or the higher-level device. This gateway can then assign the transmitted IP address 23.2 and MAC address 24.2 to network participant 21.2 and its second position in the daisy chain. Device 2.2 is then instructed to deactivate its other port, port 22.2a, or it deactivates it automatically, leading to the following: Figure 2c The depicted state results.

[0036] Steps (b) and (c) are then repeated until finally the last device 2.4, directly connected to Gateway 1 at port 11b, has deactivated its port 22.4a with network participant 21.4, which Gateway 1 detects at port 11b. This in Figure 2d This is displayed. The gateway now knows that all network participants 21.1-21.4, including their position within the daisy chain, have been detected. Gateway 1, or the higher-level device, has thus completed the device table to include the position within the daisy chain, also called the "chain position".

[0037] The detection process is complete with this step, after which the daisy chain can be returned to a normal network operating state. This can be achieved, for example, by device 2.1 or its network participant 21.1 changing the state of port 22.1a from "disabled" back to "enabled." For this purpose, a timer can be implemented in each network participant 21.1-21.4, which is started when the respective port 22.1a-22.4a is disabled, and the port is re-enabled after the timer expires. As a consequence, the connected port, in this process state port 22.2b of network participant 21.2, also recognizes that it is connected to an active port again and also sets its state to "1+". This is in Figure 2e reproduced.

[0038] Once the timers for devices 2.2-2.4 have expired, the entire daisy chain is ready for use again. To speed up the process, it can be implemented that when an active connection is detected on one port, the other port is automatically reactivated as well. In that case, network device 2.2 would immediately start using the connection detected on the other port. Figure 2e The depicted state (i.e., after the timer of device 2.1 has expired and port 22.1a has been reactivated) also activates its port 22.2a. In a chain reaction, this activation then extends almost immediately to network devices 2.3 and 2.4.

[0039] In an alternative configuration, this chain reaction can also be triggered directly by Gateway 1 by activating port 11a, which would cause device 2.1 and subsequently devices 2.2-2.4 to also activate their deactivated ports, even if no timer is provided in the network devices.

[0040] The device table thus created, for example by Gateway 1 or the higher-level unit, can now be used by any application that benefits from or depends on a location assignment. It can also be used, for example, to instruct a DHCP (Dynamic Host Configuration Protocol) server to assign desired, predefined IP addresses, possibly reflecting the location number, to the individual devices 2.1-2.4 or their network participants 21.1-21.4. Furthermore, names can be automatically assigned to the IP addresses in the network by a DNS (Domain Name Service) using this device table; these names also depend on or reflect the location number. In the event of connection problems between devices 2.1-2.4, the created table can also be used to specify precisely which devices 2.1-2.4 are experiencing the problem, thus simplifying troubleshooting for service technicians.

[0041] In Figure 3 A block diagram or wiring diagram of a milking carousel is shown as an example of a milking parlor arrangement in which the described procedure can be used.

[0042] A plurality of devices 2 are connected to each other in a daisy chain with connections 3 and to two ports 11a, 11b of a gateway 1. The devices 2 exhibit, as in the example of the Figure 1 each has a network interface (corresponding to network participants 21.1 - 21.4 of the Figure 1 ) with two ports, which are not shown in detail here for the sake of clarity. The milking carousel shown comprises a total of 16 units 2. The units 2 correspond to the milking stalls of the milking parlor layout and are therefore referred to as milking stalls 2 in the following.

[0043] By carrying out the procedure described above, it is possible to record the milking stations 2, including an assignment of the IP or MAC addresses to the individual stations along the milking carousel, starting from port 11a.

[0044] Assuming that milking parlor 2 connected to port 11a is "stall number 1" of the milking carousel, the recorded position of each milking stall 2 in the daisy chain directly represents its stall number in the milking carousel. If "stall number 1" is assigned to another milking stall 2, a corresponding offset results compared to the determined position number in the daisy chain. In this simple way, the inventive method enables the automatic recording of the IP and / or MAC addresses of the individual milking stalls 2 and their stall numbers.

[0045] Figure 4shows a comparable milking carousel with a plurality of milking stations 2, which, in contrast to the exemplary embodiment of the Figure 3 They are not connected to each other in a daisy chain, but in three separate daisy chain chains. The milking stations are designated as equipment groups 4, 5, and 6 in the Figure 4 characterized, with the milking stations of each of the equipment groups 4-6 being coupled by a daisy chain.

[0046] Device group 4 has a daisy chain of six milking stations 2 connected to ports 11a and 11b of gateway 1. Device group 5 comprises three milking stations 2 connected in a daisy chain to ports 12a and 12b. Finally, the third device group 6 comprises seven milking stations 2 connected in a daisy chain to ports 13a and 13b of gateway 1.

[0047] The previously described procedure for determining the assignment of an IP and / or MAC address to a device 2's position in a daisy chain can now be performed separately for each of the three device groups 4-6, i.e., each of the daisy chains. The wiring diagram shows that, with the depicted wiring, the first detected device 2 of the second daisy chain (device group 5) connects to the last device of the first daisy chain (device group 4). Furthermore, the first device of the third daisy chain (device group 6) connects to the last device of the second daisy chain (device group 5). Using this information from the wiring diagram, the three separately created device tables for the individual daisy chains can be combined, and an assignment of IP or MAC addresses to stall numbers on the milking carousel can be performed.Again, only one assignment needs to be specified here, which concerns the milking stall with stall number 1.

[0048] It should be noted that the three device groups 4-6 do not necessarily have to be connected to each other in the same way as in the example shown. Figure 4 This is the case. The groups can also be wired with different orientations and / or intertwined; however, the information on how the groups are related in order to be assigned a continuously ascending place number is then required when assigning them to an actual physical place on the milking carousel.

[0049] Figure 5 Finally, an arrangement of a total of six equipment groups 4-9 of milking stations 2 shows that in this case not a common milking parlor, but (as with the milking carousel of the Figure 3 and 4) are not assigned to one group of milking parlors, but rather to six different milking parlors. All these milking parlors, or equipment groups 4-9, are operated from a common gateway 1. Gateway 1 has twelve ports 11a-16a and 11b-16b, each paired with one of the daisy chain chains. The equipment groups 4-9 can, for example, be clusters of milking stalls 2, which are also spatially close to each other within the groups, so that the grouping reflects, to some extent, the spatial positioning on a farm. This also applies to the example of the Figure 5 The inventive method can be carried out independently for each of the groups 4-9 in order to determine the position of the individual milking stations 2 within one of the equipment groups 4-9 and within a daisy chain. Reference sign

[0050] 1Gateway 11a first port 11b second port 12a-16a second first port 12b-16b second second port 2, 2.1-2.4 Device (Milking station) 21.1-21.4 Network participant 22.1a-22.4a First port 22.1b-22.4b Second port 23.1-23.4 IP address 24.1-24.4 MAC address 3 Network connection 4-9 Equipment group (milking parlor)

Claims

1. Method for determining a chain position of a network subscriber (21.1-21.4) in an Ethernet network, in which the network subscribers (21.1-21.4) are connected to one another in a daisy chain and the daisy chain is coupled at its two ends to a respective port (11a-16a, 11b-16b) of an Ethernet gateway (1), comprising the following steps: (a) deactivating one of the two ports (11a-16a, 11b-16b) of the Ethernet gateway (1); (b) acquiring a status message from one of the network subscribers (21.1-21.4), wherein the status message concerns a port (22.1a-22.4a, 22.1b-22.4b) of the respective network subscriber (21.1-21.4) which, as a result of step (a), has no active connection; (c) extracting at least one identification number from the status message of the respective network subscriber (21.1-21.4) and assigning a first chain position to the identification number; (d) deactivating the other port (22.1a-22.4a, 22.1b-22.4b) of the respective network subscriber (21.1-21.4); and (e) repeating steps (b) to (d) until all network subscribers (21.1-21.4) are acquired with their identification number, wherein subsequent chain positions are assigned in the repetitions of step (c) and wherein, in step (b), the status message concerns a port (22.1a-22.4a, 22.1b-22.4b) of the respective network subscriber (21.1-21.4) which, as a result of step (d), has no active connection.

2. Method according to claim 1, in which the determined chain positions are acquired in a device table together with the identification numbers.

3. Method according to claim 1 or 2, in which the identification numbers are IP addresses (23.1-23.4) and / or MAC addresses (24.1-24.4).

4. Method according to one of claims 1 to 3, in which the status message in step (b) is issued independently by the relevant network subscriber (21.1-21.4) after detection of the non-connected port.

5. Method according to one of claims 1 to 3, in which the status message in step (b) is issued upon request by the respective network subscriber (21.1-21.4).

6. Method according to one of claims 1 to 5, in which the deactivation of the other port (22.1a-22.4a, 22.1b-22.4b) in step (d) is carried out independently by the respective network subscriber (21.1-21.4) after detecting the non-connected port.

7. Method according to one of claims 1 to 5, in which the deactivation of the other port (22.1a-22.4a, 22.1b-22.4b) in step (d) is carried out after detecting the non-connected port upon request by the respective network subscriber (21.1-21.4).

8. Method according to one of claims 1 to 7, in which after step (e) the network subscribers (21.1-21.4) and / or the gateway (1) reactivate the deactivated ports (11a-16a, 11b-16b, 22.1a-22.4a, 22.1b-22.4b) again.

9. Method according to one of claims 1 to 8, in which the network subscribers (21.1-21.4) independently reactivate the ports (22.1a-22.4a, 22.1b-22.4b) after a predetermined time after deactivation.

10. Method according to one of claims 1 to 8, in which the network subscribers (21.1-21.4) independently reactivate the ports (22.1a-22.4a, 22.1b-22.4b) when they detect a connection again on the other of the ports (22.1a-22.4a, 22.1b-22.4b).

11. Arrangement of agricultural devices (2), each having at least one network interface, which are connected as network subscribers (21.1-21.4) to a gateway (1) in an Ethernet daisy chain topology, characterized in that the arrangement is adapted for carrying out a method according to one of claims 1 to 10.

12. Arrangement according to claim 11, in which the devices (2) are milking places, in particular of a milking carousel.