Method for assigning an IP-address to a device

The invention relates to a method for assigning an IP address, to a device, and a digital storage medium, a switch and a connectable to the switch unit.

is known from the prior art, the assignment of an Internet Protocol (IP) address to a device by means of DHCP (Dynamic Host Configuration Protocol) is known. DHCP is usually in Local Area Network (LAN) environments used to IP addresses be issued by a central address server.

The choice of the device anforderndem assigned IP address meets the DHCP server to hand the local settings (Static or dynamic allocation) and in the DHCP request supplied options. Using DHCP Options (see. Website http://www.iana.org/assignments/bootp-dhcp-parameters) "# 12 hostname option", "# 43 Vendor Specific Option" "# 61 Client Identifier", "# 82 Relay Agent Information", "# 128-254 Private Use" or another newly defined Option, the DHCP server a client can with him assign stored IP address uniquely.

need to enable these unique assignment of the selected Identifier (ID) for all variants LAN-wide unique be.

This identifier must be permanently saved in the client be to a reproducible result of the address resolution to receive. As identifier for a client or a port a client can eg a UUID (Universal Unique Identifier) or a DNS (Domain Name Service) name be used. In DHCP Server multiple identifiers can (alias) for a IP address are stored.

By "# 82 Relay Agent Information" of the DHCP can the IP address requesting terminal by the port of the switching node ( "Switch") to which it is attached, accurately identified will. The Option 82 is similar to an identification by Media Access Control (MAC) address. she has the advantage that the identification of the at level 3 occurs Open System Interconnection (OSI) -Schichtenmodells supported and thus through the IP protocol (see. this IEE 48. Class of 2003 No. 11, p. 32 to 34, "Network Management for Ethernet, step in the right direction ", Frank Seufert).

The allocation of IP addresses using DHCP and "# 82 Relay Agent Information "is also well known from the U.S. Patent Applications US-A-20040010653 and US-A-20,030,101,243th

A disadvantage of the allocation of IP addresses using DHCP and Option 82 is that the switches by the elaborate filtering the DHCP requests are heavily loaded. A switch must identify all DHCP request by such a filter, and they refer to the data stream to provide additional enter information or to realize that already Entry exists .. Then, the DHCP request is again inserted into the data stream. This is particularly disadvantageous that efficient switching mechanisms such as cut-through be rendered impossible.

The invention is based on the object of providing a improved method for assigning an IP address to a creating device. the invention is also the object basis, a corresponding digital storage medium, device and to provide a switch.

The underlying basis of the invention objects are respectively solved with the features of the independent claims. Preferred embodiments of the invention are in the dependent Patent claims.

The invention provides a method for assigning a created IP address to a device, wherein an identifier of the Port to which the device is connected, from the switch to the device is transmitted. The identification of the ports is then transmitted from the device to an address server, which on the Based on the identification of the port to the device assigns an IP address. This method is applied when the device is still not received "own" identity.

Of particular advantage here is that the elaborate filtering the data packets, as for the prior art detection the DHCP request is required by the switch, deleted. Thus, the switch is released and it can efficiently Switching mechanisms such as cut-through are used in the switch.

This is especially advantageous for applications in Automation technology, especially for Industrial Ethernet and real-time Ethernet. In particular, the invention enables an efficient network management of Ethernet applications in the industrial sector, which is often a very large number of have node.

For example, a defective device by a replacement unit be replaced without manually in the network management action must be taken. The assignment of the IP address to the replacement device can automatically without any user interaction done.

According to a preferred embodiment of the invention, the Identification of the port to the device connected to the port transmitted using a neighbor discovery protocol. Suitable for this is in particular by the LLDP protocol STANDARD IEEE802.1AB. Using the mechanisms of the Slow protocol (STANDARD IEEE803.1), LLDP neighbor information in switched networks. This neighborhood information be used for the unique addressing.

According to a preferred embodiment of the invention, assigning an IP address by the DHCP protocol. After The DHCP protocol is to allocate the IP address to a device launched itself through the device by the device a corresponding DHCP request sends a DHCP address server. This DHCP request includes the identifier of the port to which the device is connected, as a basis for the assignment the IP address to the device.

According to another preferred embodiment of the invention is the Discovery and Configuration Protocol (DCP) used. Unlike DHCP this is the assignment of IP addresses not by the terminal, instead by a DCP address server initiated. This asks the question of Device, the identifier of the port connected to the device is starting to associate on this basis the IP address of the device.

For the operation described above, there are, for example, two options:<sl><li>a) It is determined which one of a particular port Switches a device to be connected. Is it not connected, the DCP address server polls the Switch to the question: "Who is connected to port x". Reports of the Switch "none" will continue to be polled. Reports of the switch device xyz, so this device is with an address and a name provided. Of particular Advantage is that in less traffic network are, because only unicasts are used.</li><li>b) It is determined which one of a particular port Switches a device to be connected. Is it not connected, the DCP address server polls the potentially existing device with a DCP Multicast with the question: "Who is connected to port x". reports no one will be polled. If the device reports itself xyz, then this device with an address and a Names provided. Of particular advantage is that announce the switch only his PortID via LLDP have to be.</li></sl>

Furthermore, preferred embodiments of the invention explained in more detail with reference to the drawings. Show it:<dl tsize="7"><dt>figure 1</dt><dd>a block diagram of an embodiment of an inventive network,</dd><dt>figure 2</dt><dd>a flow diagram of one embodiment of an inventive process,</dd><dt>figure 3</dt><dd>a block diagram of an embodiment of a Network.</dd></dl>

Figure 1 shows an automation system 100. The automation system 100 has a communications network 102, such as for example, an Ethernet. To the network 102 includes several Coupling nodes, called switches, of which one switch 104 is shown by way of example in FIG. 1

The switch 104 has a processor 106 for carrying out a Computer program 108. The computer program 108 is in the usual Manner on a digital storage medium, for example a memory stored. The switch 104 has several Ports, of which ports 110, 112, 114 exemplarily are shown in FIG. 1 Each of the ports of the Switch 104 has a port identifier, also referred to as Port ID becomes.

At the port 110 an automation device 116 is connected. The automation device 116 can be, in principle, be any device of automation technology, such as a transducer, an actuator, a controller or the like.

The automation device 116 has a processor 118 for Executing a computer program 120. The computer program 120 is in a known manner on a digital storage medium, such as a main memory, is stored.

An address server 122 via the network 102 to the port 114 of the switch 104 is connected. The address server 122 is used for Assignment of Internet Protocol (IP) addresses to the participants the automation system 100th

The procedure for assigning an IP address to the automation device 116 is as follows:

After connecting the automation device 116 to the port 110 receives the programmable controller 116 of the port 110 which Port identifier 124. The port identifier 124 is in the programmable controller 116 saved. The port identifier 124 is then from the automation device 116 via the switch 104 and the network 102 is transmitted to the address server 122nd On based on the port identification code 124 assigns the address server the PLC to 116 its IP address.

This method can at initialisation of the automation system 100 are used to all the participants the automation system to assign a IP address. A particular advantage of this method is that it also be applied when replacing a defective device can.

If, for example, the automation device 116, so it is by a corresponding functional automation device 116 exchanged. Through the connection of the exchanged automation device 116 to the port 110 the allocation of IP address is retriggered.

The algorithm for allocation of IP addresses is thereby the same result as in the assignment of IP address to the original automation device 116, since the algorithm carried out on the basis of the same port identifier 124 becomes.

For transmitting the port identification code 124 from the switch 104 to the automation device 116 is preferably a so-called Neighbor discovery protocol is used. Such Neighbor discovery protocols are known per se from the prior known art. With the help of a neighbor discovery protocol finds an automatic exchange of data between nearby devices instead of on a network.

Preferably, the proximity detection is performed by LLDP, according to Standard IEEE802.1AB (see website http://www.ieee802.org/1/pages/802.1ab.html). supplies With LLDP each Ethernet station (DTE) his name (chassis ID) and the transmitting port (port ID). Using the mechanisms Slow protocol (IEEE803.1), LLDP neighbor information in geswitchen networks.

For transmitting the port identification code 124 of the automation device 116 to the address server 122 is preferably DHCP is used. In this case, it is in the address server 122 a so-called DHCP server. After this DHCP protocol sends the automation device 116 a DHCP request to the address server 122 to the assignment of a to request IP address. The port identifier 124 as Part of the DHCP request from the automation device 116 to the address server 122 transferred, the on that basis, IP address to the automation device 116 associates.

Alternatively comes DCP used. In this case it is If the address server 122 to a print server. deviant DHCP is the initiative to assign the IP address not of the automation device 116, but by the DCP address server 122 itself. In response to a request the DCP address server 122. answers the automation device 116 with the data stored in the programmable controller 116 Port identifier 124. On this basis, the DCP address server 122 then the assignment of the IP address to the Automation device 116 before.

Preferably LLDP is used in conjunction with DHCP and Option 82 used. Since about LLDP each device knows its neighbors, this information can be used for unique addressing will. If the network 102 DHCP is used, stores the DHCP address server 122 in addition to the client ID or Host ID, the neighborhood information to an IP address. Prompts for example, the automation device 116 from the DHCP address server 122 an IP address, so it fills the optional 82 designated field with the neighbors, the ie the switch 104 obtained chassis ID and port ID (ie For example, port identifier 124). This eliminates the time-consuming Filtering the DHCP requests to the switch completely, but the DHCP address server 122 can still about the topology information, that is, the port identifier 124, an IP address assign.

Of particular advantage is that a device is replaced, for example, in case of failure of the automation device 116, without a programming device, an interchangeable storage medium such as For example, MMC, or intervention at the DHCP address server 122 possible. Furthermore, LLDP is a cost to be realized Protocol. Because of the switches of the can Automation system 100 with the same performance be realized cost-effectively.

If in the considered network 102 DCP used as stores the DCP client, that is, for example, the automation device 116, in addition to the station names, that is, the Client ID, the neighborhood information. When DCP is in Compared to DHCP, the direction is reversed with the address assignment. For example, look for the DCP address server 122 as needed active for a particular programmable controller, as For example, the automation device 116th

It can this search either via the station name, client ID, perform chassis ID or neighborhood information. Finds the DCP address server 122 the desired device 116, so the assigned IP address is assigned.

As a variant, but can also neighborhood information the DCP address server are specified 122nd In this case, search on station name, client ID or chassis ID omitted since the neighborhood information for the further Procedure is sufficient.

As a variant, only the name of the station, client ID and Chassis ID the DCP address server 122 specified. In this Case, it will search for the automation device 116 over the name of station, client ID and Chassis ID and the Stored neighborhood information.

During the commissioning phase are unique in the automation devices the automation system 100 name Station, client ID and chassis ID deposited. These are the Further uses for the search. If the desired automation device found, in addition, the neighborhood information is read and stored in the DCP address server 122nd

In this embodiment is of particular advantage that instead of a DHCP address server a much less expensive DCP address server is used. This embodiment particularly for smaller automation systems Suitable a relatively small number of participants.

Figure 2 shows a corresponding flow chart. In the Step 200 takes place during commissioning of the automation system the assignment of IP addresses to the participants the automation system. In step 202 falls during the operation of the automation system of the automation equipment out.

In the step 204, the defective automation device replaced by binding to the same port on the same switch as the original automation device connected becomes. In step 206 the port identifier is used by the switch transferred to the replacement unit. This is preferably after a neighbor discovery protocol. In the step 208 is the port identifier of the replacement device to an address server transfer. This occurs for example by DHCP or DCP. In step 210 the assignment of the IP address is to the replacement device by the address server.

Preferably, steps 206, 208 and 210 are in the same Fashion and the commissioning phase of the automation system, that is, in the step 200 is performed. This means that the steps 206-210 for the Replacement unit during operation of Automatisierüngssystems 100 are repeated, with the award algorithm for the IP address to the same conclusion as the port identifier has not changed.

In another different application scenarios of the present Invention exemplified:

Scenario 1:

All devices have received their identifier (ID) and are switched on. After switching devices send a DHCP request with the identifier to the DHCP server. This assigns the client in the reply its unique IP address to.<tables><table><tgroup cols="3"><tbody><row><entry namest="1" nameend="3">Principle:</entry></row><row><entry align="left">identifier</entry><entry align="left">Alias</entry><entry align="left">IP</entry></row><row><entry align="left">Client1.Anlage1.Werk1</entry><entry align="left">Port1.Client2.Anlage1.Werk1</entry><entry align="left">10.0.0.1</entry></row><row><entry align="left">Client2.Anlagel.Werk1</entry><entry align="left">Port3.Client1.Anlage1.Werk1</entry><entry align="left">10.0.0.2</entry></row><row><entry align="left">Client3.Anlage1.Werk1</entry><entry align="left">Port1.Client2.Anlage1.Werk1</entry><entry align="left">10.0.0.3</entry></row></tbody></tgroup></table></tables>

Scenario 2:

In a running system, a client is exchanged. This is taken from the camp and knows its identifier not. He uses the port identifier of its neighbor, the stored as an alias for the client in the DHCP Server is. For example, over the also deliverable from the DHCP Server "Host Name" is replaced by the new client and its identifier stores it retentive for further use.

Scenario 3: (DCP):

It is a plant constructed in which all neighborhoods and the connection point known of the engineering system (ES) are. After switching on the power does not have a client Identifier. The Project System (Engineering System (ES)) knows to be issued identifiers and looking over LLDP its direct neighbors. Is this Identifies so He is baptized on eg DCP with its identifier. The client with the identifier are those at about LLDP its immediate neighbors on. Thus they are in on the Scenario 2 ALIAS described addressable and also baptized. This snowballed the whole LAN / Part LAN / the whole neighborhood "baptized".

Scenario 4: (with DHCP):

It is a plant constructed in which all neighborhoods and the connection point known of the engineering system (ES) are.

After switching on the voltage no client has an identifier. The ES knows to be issued identifiers and examined its direct neighbors via LLDP. Is this Identifies, it is about as DCP with its identifier "Baptized":

The client with the identifier are those at about LLDP its immediate neighbors on. Thus these are the ALIAS described in Scenario 2 addressable via DHCP their IP address and their Identifier. This snowballed the whole LAN / Part LAN / christened the entire neighborhood.

Scenario 5: (DCP):

It is a production machine / will be several series machines constructed in a series machine all neighborhoods and the connection point of the engineering system (ES) are known.

With the method described in Scenario 3 is provided with a "Button" the entire machine to individualize. In order to is possible with little effort a fast start. In addition to the ES, this task can also be of a PLC or simple Taufgerät be taken.

Fig. 3 shows an example of a corresponding network topology.