Chapter 4, NetWare Networking with Windows 2000

Chapter 4, Lesson 1

Introducing IPX and NWLink

1. Introduction

A. Novell developed IPX specifically for use with its NetWare network operating system.

1. Unlike TCP/IP, the IPX standards were not cooperatively developed or published and remain the sole property of Novell.

B. When Microsoft began integrating networking capabilities into Windows, NetWare compatibility was considered a crucial factor.

1. Originally, it was agreed that Novell would supply a Windows client for NetWare that would be included with Windows.

2. After this client failed to materialize, Microsoft developed its own NetWare clients, including a set of protocols it called NWLink IPX/SPX/NetBIOS Compatible Transport Protocol (NWLink), which was designed to be fully compatible with IPX.

2. Introducing IPX

|1| A. In its basic architectural structure, the IPX protocol suite is somewhat similar to TCP/IP.

1. The IPX protocol for which the suite is named is a connectionless network layer protocol that performs most of the same functions that Internet Protocol (IP) does in the TCP/IP suite.

2. Like IP, IPX encapsulates transport layer data, addresses it, and routes it to its destination on the network.

B. Scale is the primary difference between TCP/IP and IPX.

1. TCP/IP was designed to support any type of computer on any type of network.

2. IPX was designed for use on local area networks (LANs) because NetWare is strictly a LAN operating system.

|2| C. The most obvious indication that IPX was designed for LANs is that it does not have its own self-contained address space, as IP does.

1. IPX uses node addresses to identify network interfaces on a network and network addresses to identify LANs.

a. The node addresses are the data-link layer hardware addresses coded into the computers’ network interface adapters.

b. NetWare randomly assigns network addresses during installation, or the network administrator can assign them manually.

2. There is no need for a network registrar or a separate node addressing system with IPX because it is limited to private LANs.

D. The greatest shortcoming of IPX is its incompatibility with the Internet.

1. When NetWare was first developed, the Internet was a small, experimental project.

2. When the Internet became popular in the mid–1990s, most software manufacturers modified their product lines to make them Internet-compatible.

a. Microsoft, for example, adopted TCP/IP as the default networking protocols for Windows, replacing NetBIOS Extended User Interface (NetBEUI).

3. However, Novell stuck with its IPX protocols for several more years.

4. NetWare 5.0 enabled the operating system to run natively using TCP/IP, eliminating IPX from the network completely.

a. Many networks still run earlier versions of NetWare that need IPX, and NetWare compatibility is still a Windows issue.

5. If you have NetWare servers on your network, you must install additional software components on your Windows systems for them to access resources on those servers.

3. IPX Protocols

A. Like TCP/IP, the IPX protocol suite consists of several protocols operating at different layers of the Open Systems Interconnection (OSI) reference model, all of which work together to provide a nearly complete protocol stack.

|3| B. Data-link layer protocols

1. IPX has no data-link layer protocols of its own; it relies completely on the standard LAN protocols used at the data-link layers, such as Ethernet and Token Ring.

2. The only important factor when combining Windows and NetWare on the same LAN is the selection of the correct frame type in the NetWare clients and servers.

3. NetWare supports four Ethernet frame types:

a. Ethernet 802.3. The original default frame type for all NetWare versions up to and including 3.11; also known as “raw Ethernet”

(1) This frame lacks any means of identifying the network layer protocol that generated the information in the frame’s Data field.

(2) Consequently, this frame type can only be used on a NetWare network running only IPX at the network layer.

(3) If you have Windows computers running NWLink on your network, you should not use this frame type.

b. Ethernet 802.2. Conforms more precisely to the Ethernet protocol standards published by the Institute of Electrical and Electronics Engineers (IEEE); is the default frame type for NetWare versions 3.12 and later

(1) You can use this frame type with any products that support IPX at the network layer (including Windows).

(2) However, this frame type still lacks a network layer protocol identifier, which means that you cannot use it on a network running more than one protocol at the network layer, such as a mixed NetWare/Windows network running both IPX and TCP/IP.

c. Ethernet II. Defined by the original DIX (Digital Equipment Corporation, Intel Corporation, and Xerox Corporation) Ethernet standard

(1) Includes an Ethertype field that specifies what network layer protocol generated the information in the frame’s payload

(2) You can use this frame type on a network running multiple protocols at the network layer, including IPX and TCP/IP.

(3) This is the most commonly used frame type on mixed Windows/NetWare networks.

d. Ethernet SNAP. Uses the same format as Ethernet 802.2, except that it includes an additional Subnetwork Access Protocol (SNAP) header in the Data field

(1) The SNAP header has a field that identifies the network layer protocol that generated the data carried in the frame, enabling you to use the frame type on networks running TCP/IP, AppleTalk, or both, in addition to IPX.

4. In most NetWare clients, the selection of a frame type is automatic and usually reliable.

5. When you troubleshoot network communication problems on a Windows/NetWare network, you should check that the computers are using an appropriate frame type and that they are all using the same frame type.

C. Internetwork Packet Exchange

1. Like IP, IPX is a networking protocol that provides a connectionless datagram transfer service and controls the addressing and routing of data packets within and between networks.

|4| 2. The IPX data encapsulation process adds a 30-byte header to a payload received from a transport layer protocol. The IPX datagram format is shown in Slide 4.

|5| 3. Functions of the IPX header fields

a. Checksum (2 bytes). A vestige of the now-unused protocol from which IPX was derived

b. Length (2 bytes). Specifies the length of the datagram in bytes, including the IPX header and the data

c. Transport Control (1 byte). Specifies the number of routers the datagram has passed through on the way to its destination; also known as the hop count

(1) The computer generating the packet sets the value of this field to 0, and each router increments the field by 1 as it processes the packet.

(2) If the value reaches 16, the packet is discarded.

d. Packet Type (1 byte). Identifies the upper layer protocol that generated the information carried in the Data field

e. Destination Network Address (4 bytes). Identifies the network on which the destination system is located, using a value assigned by the operating system or the administrator during NetWare installation

|6| f. Destination Node Address (6 bytes). Identifies the network interface adapter in the computer to which the data is to be delivered, using the data-link layer protocol hardware address

g. Destination Socket (2 bytes). Identifies the application or process on the destination system for which the information in the Data field is intended

h. Source Network Address (4 bytes). Identifies the network on which the source system is located, using a value assigned by the operating system or the administrator during NetWare installation

i. Source Node Address (6 bytes). Identifies the network interface adapter in the transmitting computer, using the data-link layer protocol hardware address

j. Source Socket (2 bytes). Identifies the application or process on the transmitting system that generated the information in the Data field

k. Data (variable). Contains the data generated by the upper layer protocol

4. IPX is responsible for end-to-end transmissions across an internetwork and is therefore responsible for network layer addressing and routing, just as IP is.

a. Both these tasks are somewhat simpler than they are in IP because IPX is working only with private internetworks of limited size.

(1) For example, IPX discards datagrams after 16 hops, assuming them to be circulating endlessly around the network because of a routing problem.

(2) IP datagrams generated by a computer running Windows have a Time to Live (TTL) value that enables them to travel as many as 128 hops, because processing by literally dozens of routers is a distinct possibility on the Internet.

5. As with IP, routers on IPX networks use dynamic routing protocols, such as Routing Information Protocol (RIP), to share their routing information.

|7| D. Sequenced Packet Exchange

1. SPX is a connection-oriented transport protocol that is often thought of as a frequent partner to IPX, just as TCP is to IP.

a. The IPX protocol suite is often referred to as the IPX/SPX protocols.

b. In actuality, NetWare uses the combination of IPX and SPX far less frequently than the TCP/IP protocols use the combination of TCP
and IP.

(1) For example, NetWare file sharing, which generates much of the traffic in a typical NetWare network, uses the NetWare Core Protocol (NCP).

c. SPX is relegated to communications between network print devices, remote console sessions, network backups, and other relatively infrequent activities.

2. Like TCP, SPX provides reliable delivery by assigning sequence numbers to packets and acknowledging specific sequences.

a. A computer receiving SPX data verifies successful delivery of the packets by generating acknowledgment messages and returning them to the sender.

(1) If an acknowledgment request brings no response within a specified time, SPX retransmits the request as many as eight times.

(2) If no response is received, SPX assumes the connection has failed.

b. Unlike TCP, the original SPX implementation required a separate acknowledgment for each data packet.

3. Novell has modified SPX over the years, adding a packet burst mechanism.

a. Packet burst, or burst mode, enables a computer to transmit multiple data packets without requiring that each packet be sequenced and acknowledged individually.

b. By enabling the receiving system to acknowledge multiple packets at once, burst mode can reduce network traffic on most IPX networks.

c. Additionally, the packet burst mechanism monitors dropped packets and retransmits only the missing packets.

(1) In Windows 2000, SPX burst mode is enabled by default.

4. SPXII improves on SPX by enabling it to perform better on high-bandwidth networks.

a. SPXII allows a greater number of outstanding packets than SPX.

(1) SPX cannot have more than one outstanding unacknowledged packet at any time.

(2) SPXII can have as many outstanding packets as the communicating computers negotiate during connection establishment.

b. SPXII allows larger packets than SPX does.

(1) SPX has a maximum packet size of 576 bytes.

(2) SPXII can use the maximum packet size of the data-link layer protocol.

(a) For example, on an Ethernet network, SPXII can use packets as large as 1518 bytes.

5. Because SPX is a connection-oriented protocol, it performs a handshake before transmitting any application data.

|8| 6. The header that the protocol adds to the information generated by the application layer protocol is 12 bytes long. The SPX header format is shown on Slide 8.

|9| 7. Functions of the SPX header fields

a. Connection Control (1 byte). Contains a code that helps to regulate the bidirectional flow of data between the computers

(1) Indicates, for example, that the packet contains the end of a message, requires an acknowledgment, or is a system packet

b. Datastream Type (1 byte). Indicates the function of the data in the message and the upper layer process for which it is intended

c. Source Connection ID (2 bytes). Contains a unique value used by the source computer to identify this connection, because two computers can have multiple connections open between the same sockets simultaneously

d. Destination Connection ID (2 bytes). Contains the unique value used by the destination computer to identify this connection

|10| e. Sequence Number (2 bytes). Contains a value that identifies this message’s place in the sequence of messages that make up the transaction

(1) The destination system uses these values to place the incoming messages in the proper sequence.

f. Acknowledgment Number (2 bytes). Contains the sequence number of the next message that the system expects to receive from the connected system, tacitly acknowledging all the packets with lower sequence number values

g. Allocation Number (2 bytes). Provides flow control by specifying the number of packet receive buffers available on the system

h. Data (variable). Contains the information generated by an application layer protocol

|11| E. NetWare Core Protocol

1. NCP is NetWare’s primary upper layer protocol and is responsible for the majority of the traffic traveling between clients and servers on most NetWare networks.

2. Carries messages that perform a wide variety of services, including file sharing, printing, directory services communications, messaging, and data synchronization

3. Is difficult to place in the OSI reference model because its functions span the transport, session, presentation, and application layers

4. NCP messages are carried within IPX datagrams, just like those of SPX.

5. There is also a variation on the protocol, called the NetWare Core Packet Burst Protocol (NCPB), which enables a computer to transmit multiple data packets without requiring each one to be acknowledged individually.

|12| 6. NCP uses headers for request and reply messages that are slightly different from each other. The NCP Request message format is shown on Slide 12.

|13| 7. Functions of the NCP Request header fields

a. Request Type (2 bytes). Specifies the message’s basic function

b. Sequence Number (1 byte). Contains a value that identifies this message’s place in the sequence of messages that make up the transaction

(1) The destination system uses these values to place the incoming messages in the proper sequence.

c. Connection Number Low (1 byte). Contains a value that indicates the number of the client’s connection to the server

d. Task Number (1 byte). Contains a unique value that the computers use to associate request messages with replies

|14| e. Connection Number High (1 byte). Not used

f. Function (1 byte). Contains a code that indicates the message’s specific function

g. Subfunction (1 byte). Contains a code that further defines the message’s function

h. Subfunction Length (2 bytes). Specifies the length of the Data field

i. Data (variable). Contains information specific to the type of request, such as the name and location of a file

|15| 8. Functions of the NCP Reply header fields (the NCP Reply message format is shown in Slide 15)

|16| a. Reply/Response Type (2 bytes). Specifies the type of reply

b. Sequence Number (1 byte). Contains a value that identifies this message’s place in the sequence of messages that make up the transaction

(1) The destination system uses these values to place the incoming messages in the proper sequence.

c. Connection Number Low (1 byte). Contains a value that indicates the number of the client’s connection to the server

d. Task Number (1 byte). Contains a unique value that the computers use to associate request messages with replies

|17| e. Connection Number High (1 byte). Not used

f. Completion Code (1 byte). Specifies whether the associated request succeeded or failed

(1) A value of 0 indicates success; a value other than 0 indicates failure.

g. Connection Status (1 byte). Indicates whether the connection between the client and the server is still active

(1) A value of 0 indicates that the connection is active; a value of 1 indicates that it is not active.

h. Data (variable). Contains information transmitted by the server in response to the associated request

|18| F. Windows 2000 and NetWare compatibility

1. Windows 2000 includes a complete collection of NetWare compatibility components in addition to its Windows networking components.

a. NetWare compatibility components included with Windows 2000 include NWLink, a client module, and a Directory Service Migration Tool.

b. When you install NWLink and an appropriate NetWare client on a computer running Windows, the computer can access files, printers, and other services on your NetWare servers.

2. NWLink is the IPX counterpart to the Internet Protocol (TCP/IP) module.

a. NWLink provides all network and transport layer services, plus additional upper layer functions.

b. NWLink is written to the same interfaces as TCP/IP, that is, the Network Device Interface Specification (NDIS) interface at the boundary between the data-link and network layers, and the transport driver interface (TDI) at the top of the transport layer.

(1) Therefore, NWLink can operate parallel to the TCP/IP module, using the same NDIS driver below it and the same client above.

|19| (2) One network interface adapter, therefore, can enable Windows 2000 to participate on both Windows and NetWare networks.

|20| 3. In addition to NWLink, Windows 2000 also includes a NetWare client, which performs roughly the same services as Client for Microsoft Networks, but for NetWare.

a. Microsoft Windows 2000 Professional includes Client Service for NetWare (CSNW), and Microsoft Windows 2000 Server includes Gateway Service for NetWare (GSNW).

b. Both clients provide basic client access to NetWare servers, but GSNW includes additional gateway capabilities that CSNW lacks.

c. You must install NWLink if you want to use CSNW or GSNW to connect to NetWare servers.

4. The Windows 2000 NetWare networking components are not the only alternative for NetWare connectivity.

a. Novell produces its own NetWare client software for Windows, called Novell Client for Windows NT/2000.

b. The Novell client includes its own IPX protocol module and Open Data-link Interface (ODI) network adapter drivers that you can use instead of Windows’ own NDIS drivers.

(1) ODI drivers are not compatible with the Windows protocol and client modules, so you can use them only on a NetWare-only network.

(2) NDIS drivers are compatible with both the Windows and Novell networking components, thereby providing both Windows and NetWare connectivity.

5. NetWare versions 5.0 and later include native support for TCP/IP, which enables you to eliminate IPX from the network entirely.

a. However, Windows CSNW and GSNW do not work with TCP/IP; they require NWLink.

b. If you use CSNW and GSNW, you must also enable IPX on your NetWare servers.

c. To access NetWare 5 servers using TCP/IP, you must use Novell Client for Windows NT/2000.

d. Only the Novell client enables you to run NetWare’s primary network administration tool, NetWare Administrator.

Chapter 4, Lesson 2
Using the Windows 2000 NetWare Clients

1. Introduction

A. To install the NetWare networking components included with Windows 2000, you use Network And Dial-Up Connections in Control Panel.

B. When you install CSNW, Windows 2000 automatically installs NWLink, which is required for CSNW to function.

C. You can also install NWLink by itself and use it for Windows communications, but this is rarely done.

2. Installing Client Service for NetWare

A. Before you can install CSNW on a computer running Windows 2000 Professional, you must have a network interface adapter and the appropriate device driver installed.

1. This creates a Local Area Connection icon in the Network And Dial-Up Connections window.

2. To install CSNW, you need Administrator rights to the computer running Windows 2000 Professional.

|21| B. To install CSNW:

1. Click Start, and then select Network And Dial-Up Connections from the Settings menu to open the Network And Dial-Up Connections window.

2. Right-click the Local Area Connection icon, and then select Properties from the shortcut menu to open the Local Area Connection Properties dialog box.

3. In the General tab, click Install to open the Select Network Component Type dialog box.

4. Select Client, and then click Add to open the Select Network Client dialog box.

5. Click Client Service For NetWare, and then click OK. When asked if you want to restart the computer, click No to return to the Local Area Connection Properties dialog box.

a. In the Local Area Connection Properties dialog box, notice that the NWLink NetBIOS and NWLink IPX/SPX/NetBIOS Compatible Transport Protocol modules have been automatically installed in addition to CSNW.

6. Click OK to complete the installation and close the Local Area Connection Properties dialog box.

7. Restart the computer.

a. To confirm that NWLink is working properly, at the command prompt, type ipxroute config to display a table with information about the bindings for which NWLink is configured.

C. If you plan on running a NetWare-only network—that is, one in which computers running Windows access only shared NetWare resources—you can do one of the following:

1. Remove the CSNW and Internet Protocol (TCP/IP) modules from the Local Area Connection Properties dialog box.

2. Unbind the modules from the network adapter.

3. Configuring NWLink

A. In most cases you can use NWLink immediately after installing it, with no additional configuration.

1. Sometimes you might have to modify the default NWLink settings, so it is a good idea to know what they represent.

|22| B. To configure NWLink:

1. Click Start, and then select Network And Dial-Up Connections from the Settings menu to open the Network And Dial-Up Connections window.

2. Right-click the Local Area Connection icon, and then select Properties from the shortcut menu to open the Local Area Connection Properties dialog box.

3. Select NWLink IPX/SPX/NetBIOS Compatible Transport Protocol, and then click Properties to open the NWLink IPX/SPX/NetBIOS Compatible Transport Protocol Properties dialog box.

4. Specify values for any or all of the Internal Network Number, Frame Type, and Network Number settings, and then click OK to close the NWLink IPX/SPX/NetBIOS Compatible Transport Protocol Properties dialog box.

5. Click OK to close the Local Area Connection Properties dialog box.

C. The internal network number is used for internal routing when the computer running Windows 2000 is also hosting IPX services.

1. When calculating the best possible route for transmitting packets to a specified computer, multiple routes with the same route metrics can create ambiguities for computer hosts.

2. When you specify a unique internal network number, you create a virtual network inside the computer.

a. This creates a single optimum path from the network to the services running on the computer.

|23| D. The frame type defines the way the computer’s network interface adapter formats the NWLink data for transmission over the network.

1. To communicate with NetWare servers, a computer running Windows 2000 must be configured to use the same frame type as the servers.

2. By default, NWLink is configured to automatically detect the frame type used on the network and configure itself accordingly.

3. During the autodetection process, NWLink tries each of the frame types for the data-link layer protocol used by the network interface adapter.

a. For example, on an Ethernet network, NWLink tries the Ethernet 802.2, Ethernet 802.3, Ethernet II, and Ethernet SNAP frame types to see which ones it can use to communicate with the other IPX systems on the network.

b. Computers using Token Ring or Fiber Distributed Data Interface (FDDI) use two different frame types.

4. When NWLink receives a response from a NetWare server with one of the frame types, it also receives the network number associated with that frame type for the network segment on which the client resides.

5. NWLink then rebinds, using the frame types from which it received responses.

6. If NWLink detects no network traffic or detects multiple frame types in addition to the 802.2 frame type, NWLink sets the frame type to 802.2.

7. Occasionally, the autodetection mechanism selects an inappropriate network number and frame type combination for the adapter.

a. Because the mechanism uses the responses it receives from computers on the same network segment, it might select an incorrect frame type and network number if the computers responded with incorrect values.

(1) Incorrect values are usually caused by an incorrect manual setting on another computer on the network.

8. If the autodetection mechanism selects an inappropriate frame type and network number for a particular adapter, you can manually set a frame type in the NWLink IPX/SPX/NetBIOS Compatible Transport Protocol Properties dialog box.

a. Changing the contents of the Frame Type box to any value but Auto Detect activates the Network Number box, enabling you to specify the network number currently used by the network segment to which the computer is connected.

b. The external network number is a unique number that represents a specific network segment and associated frame type.

9. All computers on the same network segment that use a given frame type must have the same external network number, which must be unique for each network segment.

a. The IPX frame type and network number are set during the initial NetWare server configuration, so it is best to check a NetWare server, rather than another client workstation, to determine the correct Network Number value.

|24| 4. Configuring Client Service for NetWare

A. When you install CSNW in the Local Area Connection Properties dialog box, the Select NetWare Logon dialog box appears.

1. In this dialog box, you can specify either a preferred NetWare server to which the client will always connect first, or a default Novell Directory Services (NDS) tree and context that the client will use to log in to the network.

a. A preferred server is the NetWare server to which you are automatically connected when you log on if your network does not use NDS.

b. The tree and context define the position of the user object for the user name you use to log on to an NDS tree.

(1) You set a default tree and context only in an NDS environment; otherwise, you set a preferred server.

2. You can also specify whether the client should execute the login scripts associated with the user’s account.

a. You do not have to specify login parameters for the client during the client installation, however.

(1) You can access the configuration parameters for CSNW at any time through the CSNW Control Panel created during the client installation process.

B. When you double-click the CSNW icon in the Control Panel, the Client Service For NetWare dialog box appears.

1. You can set the Preferred Server, Default Tree And Context, and Run Login Script Options parameters, as well as configure the client’s printing behavior.

|25| 5. Using Gateway Service for NetWare

A. Introduction

1. GSNW included with Windows 2000 Server is a superset of CSNW.

2. The server computer can access NetWare file and print resources just like a workstation running CSNW, but it can also function as a gateway to those resources, enabling computers running Windows to access NetWare resources through the Windows 2000 server.

a. GSNW is a bridge between NetBIOS used by the Windows network and NCP used by the NetWare network.

b. When you enable a gateway on the network, computers running Microsoft client software can access NetWare files and printers without having to run NetWare client software locally.

|26| 3. For file access, the gateway server maps one of its own drives to the NetWare volume and then shares that drive with other Microsoft clients.

a. The gateway uses a NetWare account on the computer running Windows 2000 Server to create a validated connection to the NetWare server.

b. This connection appears on the computer running Windows 2000 Server as a normal mapped drive.

c. When you share the mapped drive, it becomes like any other shared Windows network resource.

d. Other computers on the network can access the drive like any other shared Windows resource, without being aware that the shared files and folder are actually stored on a NetWare server.

4. After the gateway connection between the Windows 2000 server and the NetWare server is established, it is disconnected only if

a. The computer running Windows 2000 Server is turned off

(1) GSNW runs continually, so logging off the computer running Windows 2000 Server does not, by itself, disconnect the gateway.

b. The administrator disconnects the shared resource or disables the gateway

c. A network problem prevents access to the NetWare server

5. Accessing NetWare resources through a gateway is generally slower and less efficient than accessing them directly using a NetWare client.

a. This is particularly true when you have many clients accessing NetWare resources through the gateway, because all the access requests have to go through one GSNW computer and be transmitted again to the NetWare server.

b. GSNW is intended for the following circumstances:

(1) Servicing clients that require only occasional access to NetWare resources

(2) When you are gradually migrating your network from NetWare to Windows 2000 and want to provide users with temporary access to resources on a NetWare server before you move them to a Windows server

6. Advantages of GSNW

a. You do not have to install a NetWare client and IPX or NWLink on every computer accessing the NetWare resources.

b. You do not have to maintain separate accounts for all your users on both Windows and NetWare.

B. Installing Gateway Service for NetWare

1. You can install GSNW when you install Windows 2000 Server, or you can install GSNW later.

a. You must be logged on as a member of the Administrators group to install and configure GSNW.

b. Before you install GSNW on a computer, remove any existing client software that is compatible with NCP, including any Novell client software, from the computer.

|27| 2. To install GSNW after the operating system installation:

a. Click Start, and then select Network And Dial-Up Connections from the Settings menu to open the Network And Dial-Up Connections window.

b. Right-click the Local Area Connection icon, and then select Properties from the shortcut menu to open the Local Area Connection Properties dialog box.

c. In the General tab, click Install to open the Select Network Component Type dialog box.

d. Select Client, and then click Add to open the Select Network Client dialog box.

e. Click Gateway (And Client) Service For NetWare, and then click OK. When asked if you want to restart the computer, click No.

(1) As with CSNW, installing GSNW also causes Windows 2000 to install the NWLink NetBIOS and NWLink IPX/SPX/NetBIOS Compatible Transport Protocol modules, if they have not been installed already.

f. Click OK to complete the installation and close the Local Area Connection Properties dialog box.

g. Restart the computer.

C. Configuring Gateway Service for NetWare

1. As with CSNW, when you install GSNW, Windows 2000 prompts you to specify a default NDS tree and context or a preferred server in the Gateway Service For NetWare dialog box.

2. To set a preferred server later, open the GSNW Control Panel to display the Gateway Service For NetWare dialog box, click Preferred Server and, in Select Preferred Server, type the name of the NetWare server you want to use.

a. If you do not want to set a preferred server, click None.

(1) You are then logged on to the nearest available NetWare server, and your interaction with the NetWare network is through that server.

b. If you do not set a preferred server, you are prompted to set one each time you log on.

3. If you are running NDS, you can specify a default tree and context in place of the preferred server; you cannot configure both.

a. If you select a default tree and context, you can still access NetWare servers that use bindery security.

b. To set a default tree and context, open the GSNW Control Panel, click Default Tree And Context, and then type the tree name and the context where your user object is located in the Tree and Context boxes.

D. Creating a gateway

1. Before you can create a gateway to NetWare resources using GSNW, you must create a group called NTGATEWAY, either on your preferred NetWare server or in the NDS tree and context you specified.

a. The NTGATEWAY group must also have a user account as a member with the permissions needed to access the NetWare resources you want to share with Windows users.

b. The NetWare user account you use to enable gateways can be either an NDS account or a bindery account.

(1) If the server will have gateways to both NDS and bindery resources, you should use a bindery account (which can connect to NDS resources using bindery emulation).

(2) If you intend to create gateways only to NDS resources, you can use an NDS account.

E. Enabling gateways in Windows 2000

1. Creating a gateway on a computer running Windows 2000 Server is a two-step process.

a. First, you enable the gateway, which provides the means for the client to log on to the NetWare resources.

(1) When you enable a gateway, you must type the name and password of the user account that has access to the NetWare server and is a member of the NTGATEWAY group.

(2) You do this only once for each server that will act as a gateway.

b. Next, you activate a gateway for each volume or printer you want to share.

|28| 2. To enable a gateway in Windows 2000 Server:

a. Click Start, point to Settings, click Control Panel, and then click Gateway Service For NetWare.

b. Click Gateway to display the Configure Gateway dialog box.

c. Select the Enable Gateway check box.

(1) This activates the Gateway Account, Password, and Confirm Password boxes.

d. In the Gateway Account box, type the name of the account you created in the NetWare NTGATEWAY group.

e. In the Password and Confirm Password boxes, type the password for the gateway account, and then click OK to close the dialog box.

3. You can now share NetWare file and printing resources over a Windows–based network.

F. Activating gateways

1. The second step in creating the gateway is to activate a gateway for each volume or printer you want to share.

a. When you activate a gateway, you identify the NetWare resource to be shared and specify the share name that Windows clients will use to access the resource.

b. To activate a gateway for a volume, you use the Gateway Service for NetWare Control Panel.

c. To activate a gateway for a printer, you use the Add Printer Wizard.

2. If you are activating a gateway to an NDS resource and the gateway user is a bindery account, use a bindery context name to identify the resource.

a. If you are using an NDS user account and you do not plan on also creating gateways to bindery resources, specify the NDS resource name.

|29| 3. To activate a gateway to a NetWare file resource:

a. Click Start, point to Settings, click Control Panel, and then click Gateway Service For NetWare.

b. Click Gateway to open the Configure Gateway dialog box.

c. Click Add to display the New Share dialog box.

d. In the Share Name box, type a share name that Microsoft clients will use to access the NetWare resource.

e. In the Network Path box, type the network path of the NetWare volume or directory you want to share.

(1) When specifying the value for the Network Path field, you use Universal Naming Convention (UNC) notation, as in \\netwareserver\netwarevolume.

|30| f. In the Use Drive box, type the default drive you want to map to the share, if necessary.

g. Click Unlimited, and then click OK to add the share to the list in the Configure Gateway dialog box.

(1) You can also click Allow, enter a maximum number of concurrent users, and then click OK.

h. Repeat steps c–g (steps 3–7 on Slides 29–30) to create additional shares, or click OK to close the Configure Gateway dialog box.

i. Click OK to close the Gateway Service For NetWare dialog box.

4. To activate a gateway to a NetWare printer, you create a network printer in the Printers window in the usual manner, specifying the name of a NetWare printer instead of a Windows printer, and then share the printer from its Properties dialog box.

G. Securing gateway resources

1. Security for gateway resources is provided on two levels:

a. On the computer running Windows 2000 Server and acting as a gateway, you can set share-level permissions for each of the resources made available through the gateway.

b. On the NetWare server, the NetWare administrator can assign trustee rights to the user account that is used for the gateway or to the NTGATEWAY group.

(1) These rights are enforced for all Microsoft client users who access the resource through the gateway.