OS/2 Arsenal

home *** CD-ROM | disk | FTP | other *** search

/ OS/2 Arsenal / OS2 Arsenal v1.0 (Disc 1)(Arsenal Computer).ISO / os2_inet / tcp20c4.exe / BASEOC1.ZIP / DOC / readme.bas

Wrap

BASIC Source File | 1994-06-14 | 146.4 KB | 3,411 lines

TCP/IP Version 2.0 for OS/2 Base Kit Note: This TCP/IP kit has Corrective Service UN56401 and UN50382 already applied. This README file reflects new and changed functions for the TCP/IP 2.0 for OS/2: Base kit, and contains changes introduced in the original product as well as changes due to Corrective Service Diskettes (CSDs) UN56401 and UN50382. Contents of this README file include: - Previous CSD Information - Obtaining CSDs - CSD UN56401 Contents - CSD UN50382 Contents - Fixed APARs ************************************* ***************** Previous CSD Information ******************* ************************************* This CSD incorporates the corrective service and new functions that have been released in all available CSDs for TCP/IP Version 2.0 for OS/2. This file contains information about all functions described in the following CSDs: - UN56401 - UN50382 ************************************* **************** Obtaining CSDs ******************* ************************************* You can obtain Corrective Service Diskettes (CSDs) electronically by: o Using FTP to access SOFTWARE.WATSON.IBM.COM as Anonymous. NOTE: Fixes for TCP/IP for OS/2 are located in the pub\tcpip\os2 directory. o Establishing a modem connection to the NCSBBS bulletin board at 1-919-517-0001. NOTE: Set your modem settings to: none,8,1. The latest CSDs for TCP/IP Version 2.0 for OS/2 are shown in the following table: +=============================+===========================+ | KIT NAME | PACKAGE NAME | +=============================+===========================+ | Applications Kit | APP20CSD.PKG | +-----------------------------+---------------------------+ | Base Kit | TCP20CSD.PKG | +-----------------------------+---------------------------+ | Domain Name Server Kit | DNS20CSD.PKG | +-----------------------------+---------------------------+ | DOS/Windows Access Kit | DOS20CSD.PKG | +-----------------------------+---------------------------+ | Extended Networking Kit | XNT20CSD.PKG | +-----------------------------+---------------------------+ | NetBIOS Kit | NET20CSD.PKG | +-----------------------------+---------------------------+ | Network File System Kit | NFS20CSD.PKG | +-----------------------------+---------------------------+ | OSF/Motif Kit | MOT20CSD.PKG | +-----------------------------+---------------------------+ | Programmer's Tool Kit | PGM20CSD.PKG | +-----------------------------+---------------------------+ | X Window System Client Kit | XCL20CSD.PKG | +-----------------------------+---------------------------+ | X Window System Server Kit | PMX20CSD.PKG | +-----------------------------+---------------------------+ ************************************* ***************** UN56401 CORRECTIVE SERVICE DISKETTE ******************* ************************************* Topics included in CSD UN56401 are: 1. Remote Installation (CID) 2. RUNNING TCP/IP FOR OS/2 FROM A FILE SERVER 3. Autostart Using INETD 4. Packet Tracing Programs 5. Changes to FTP, FTPD, and FTPPM 6. Configuring KEEPALIVE Servers: the INETCFG Command 7. TelnetPM Cursor Modification 8. SLIP 9. SNMP Enhancements 10. Default 3270 Emulator Keyboard Mappings 11. The DPKT Utility 12. Configuration Notebook 13. Configuring a Token-Ring Adapter using BOOTP 14. SETTERM 15. SendMail Server and Client 16. REXEC and RSH Redirection 17. Programming Support 18. Applications Kit / ANYNET Users Information -------------------------------------------------------------------- 1. Remote Installation (CID) If you are using CID technology to install TCP/IP, the statements in the response file that contain the INSTALL_NAME and the LINK_NAME must preceed the statements that contain the EXEC, DEPENDENCY, or LINK_INFO. In addition, since CSDs are cumulative, you must erase or remark out the lines in your DEFAULT.RSP file that pertain to the previous CSD level for the component you want to enable. -------------------------------------------------------------------- 2. RUNNING TCP/IP FOR OS/2 FROM A FILE SERVER If you want to conserve your local disk space, you can run TCP/IP for OS/2 from a file server using a requester (such as IBM's OS/2 LAN Requester or TCP/IP's NFS client). To run TCP/IP for OS/2 from a server, ensure the following: 1. You have the required local files 2. Your PATH statement and environment variables contain the correct information 3. You have access to the file server -------------------------- 2.1 REQUIRED LOCAL FILES Although many of the files used by TCP/IP are common and can be accessed on the file server, certain files contain configuration information and are unique to each installation. These files must be located in directories on your workstation. In the tcpip\ETC subdirectory on your workstation, ensure that you have copies of all the files from the \ETC subdirectory on the file server. In the tcpip\BIN subdirectory on your workstation, ensure that you have the following files: o CNTRL.EXE o SETUP.CMD o TCPSTART.CMD o PMX.EXE o TCPIPCFG.EXE o TELNET.EXE In the \IBMCOM\PROTOCOL subdirectory on your workstation, ensure that you have all LAPS files, including the following files: o INET.SYS o IFNDIS.SYS o TCPIP.NIF If you are using NFS to access the server, also ensure that you have the following files in your tcpip\BIN subdirectory: o NFS200.IFS o NFSCTL.EXE o NFSSTART.CMD o ARP.EXE o ROUTE.EXE o IFCONFIG.EXE o MOUNT.EXE o QMOUNT.EXE o NFSWAIT.EXE And in your tcpip\DLL subdirectory: o TCPIPDLL.DLL o CNBAPP.DLL o CNBNFS.DLL o CNBPMX.DLL o CNBSTK.DLL In addition, you need the following empty subdirectories: o TCPIP\DOC o TCPIP\TMP -------------------------- 2.2 SET PATH STATEMENT AND ENVIRONMENT VARIABLES To run TCP/IP from a server, the PATH statement and the environment variables must point to the proper locations on the server and on your workstation. The PATH statement must point to the directory on your workstation that contains the local TCP/IP batch files and executables and to the directory on the file server that contains the TCP/IP executables. For example, if you access a network drive that contains TCP/IP as your X drive, your PATH statement would look similar to this: PATH=...;C:TCPIP\BIN;X:\TCPIP\BIN; The LIBPATH statement must point to the directory on your workstation that contains the local TCP/IP dynamic link libraries (DLLs) and to the directory on the file server that contains the TCP/IP DLLs. For example, if you access a network drive that contains TCP/IP as your X drive, your LIBPATH statement would look similar to this: LIBPATH=...;C:TCPIP\DLL;X:\TCPIP\DLL; The ETC environment variable must point to the ETC subdirectory on your workstation. For example: SET ETC=C:\TCPIP\ETC -------------------------- 2.3 ACCESS THE SERVER If you are using NFS to access the file server, you will need to start TCP/IP and NFS before you access the file server. However, no other TCP/IP services should be started until you access the file server. If you are using other methods to access the file server, you should access the file server before using any services of TCP/IP. -------------------------------------------------------------------- 3. Autostart Using INETD If you use INETD to autostart TCP/IP services, any parameters that you specify will be ignored. However, for TFTPD, you can add the following environment variable to your CONFIG.SYS to specify the default path: SET TFTPDPATH=<path> For REXECD, you can add the following environment variable to your CONFIG.SYS to specify the timeout: SET REXECDTIME=<time> -------------------------------------------------------------------- 4. Packet Tracing Programs TCP/IP for OS/2 provides a set of utilities that you can use to trace packets. Tracing packets is useful in diagnosing problems at your workstation. These utilities allow you to: o Initiate a trace o Convert trace information to human-readable output o Convert trace information to network analyzer data -------------------------- 4.1 Initiating a Trace The utility, IPTRACE, traces all packets received from and sent to an interface. The syntax of the IPTRACE command is: >>--iptrace--.-----.--.--------------.------------------------>< '- -i-' | <----------< | '-- interface--' -i Specifies that only IP packets are to be traced. The default is to include all information (such as hardware type). Certain interfaces (for example, X25 and SNALINK) require this parameter. interface Specifies an interface to be traced (for example, lan0 or sl0). If not specified, all interfaces are traced. NOTES: 1. IPTRACE writes data to IPTRACE.DMP in the directory from which you initiated the IPTRACE command. IPTRACE records all traffic sent and received, but does not check for sufficient disk space to record that information. As a result, running IPTRACE can impact your workstation's performance as the IPTRACE.DMP file continues to grow larger. 2. IPTRACE is not a network monitor. It can trace only data received by and sent from the specified interfaces. 3. IPTRACE provides a time stamp, recording when the packet was sent or received. 4. To stop IPTRACE, press Enter, Ctrl-Break, or Ctrl-c. 5. Use IPFORMAT to convert the IPTRACE.DMP file into a user-readable format. -------------------------- 4.2 Converting Trace Information The IPFORMAT utility converts the data in the IPTRACE.DMP file and the SLIPTRC.DMP file (if the file was created without VJ header compression on) to either human-readable format, which is displayed to the screen, or to data to be used as input to a network analyzer. If you choose to convert the data into human-readable format, you can redirect the output to a file. IPFORMAT reads the header information in the trace to determine the type of packet received; for example, token-ring (TRC) or ethernet (ENC). It then separates the data by the IP, TCP, UDP, and ICMP layers. The rest of the packet is displayed as hexadecimal output. The syntax of the IPFORMAT command is: .- -f IPTRACE.DMP-. >>--ipformat--.-----.--.-----.--+-----------------+--.-----.---> '- -a-' '- -d-' '- -f filename----' '- -h-' >---.-----.---.---------------.--.-----.--.-------------.----->< '- -n-' '- -s hwaddress-' '- -x-' '- > filename-' -a Do not format ARP or RARP packets. -d Do not display the data portion of a packet. -f filename Specifies the input file name. The default is IPTRACE.DMP. -h Display the raw data packet after the formatted information. -s hwaddress Format data only for the specified hardware address. 'hwaddress' is the 12-digit hexadecimal address for the Ethernet or token-ring adapter. You can use the NETSTAT -N command to display this address. -n Do not display hexadecimal data for unknown data type. -x Converts IPTRACE data to a format that can be read by a Network General Sniffer. > filename Redirect the output to the specified file. -? Display help information. -------------------------------------------------------------------- 5. Changes to FTP, FTPD, and FTPPM -------------------------- 5.1 IMAGE Subcommand for FTP FTP now supports an image transfer type. The IMAGE subcommand sets the file transfer type to image. The image file transfer type is useful for image transfers, such as executable files. Using the IMAGE subcommand is the same as issuing the TYPE subcommand with the binary parameter. You can issue the IMAGE subcommand with the abbreviation IM. >>--image-------------------------------------------------------->< -------------------------- 5.2 IMAGE Parameter for TYPE Subcommand for FTP The TYPE subcommand now supports the IMAGE parameter. >>--type--.---------.-------------------------------------------->< |- ascii--| |- binary-| '- image--' image Specifies the file transfer type as image (binary). Specifying the TYPE subcommand with the image parameter is the same as issuing the IMAGE subcommand. -------------------------- 5.3 FTPD Trace Facility FTPD now includes a trace facility. The FTPD command uses the FTPD.EXE program to start the FTP server. It runs as a task until you shut down the server. SYNTAX >>--ftpd--.--------------.--.------------.--.-----.-------------->< '- -c codepage-' '- -p portno-' '- -t-' -c codepage Specifies the ASCII codepage used by the remote host. If you do not specify a codepage, the data is sent untranslated. Possible values are: TCPDECMU DEC Multinational codepage TCP8859 ISO 8859 codepage NONE PC codepage 850 -p portno Specifies the port that you want FTPD to use. If you omit this option, FTPD uses its well-known port assignment. -t Specifies that FTPD is to trace control connection commands and replies. FTPD places the information into FTPD.TRC in the subdirectory specified by the ETC environment variable. Tracing will continue until you terminate FTPD. -------------------------- 5.4 TRACING FTPPM COMMANDS/REPLIES If you select the Trace Commands/Replies option, from the Options pull-down menu, you can trace your FTPPM session activities. The commands you issue and the replies you receive are written to a file named FTPPM.TRC in the directory specified by the ETC environment variable. The trace file is overlayed (information is not appended) each time the trace option is selected. -------------------------- 5.5 Ending an FTPPM Session You can end an FTPPM session and disconnect from all remote hosts in one of the following ways: o Select the Close all Connections option from the Remote pull-down menu. You will be prompted to confirm that you want to end all connections. o Press the F3 key to exit. You will not be prompted to confirm that you want to exit the session. o Close the FTPPM window. You will not be prompted to confirm that you want to exit the session. NOTE: Options set for the current session are not saved for future sessions. -------------------------------------------------------------------- 6. Configuring KEEPALIVE Servers: the INETCFG Command The INETCFG command configures TCP/IP parameters. >>--inetcfg---- keepalive=value------------------------------->< value The time in minutes until the KEEPALIVE timer expires on a TCP socket. Valid values are from 0 to 120. If you specify 0, INETCFG sets the timer to 120. Note: To use the INETCFG command, Telnetd cannot first be started with INETD. -------------------------------------------------------------------- 7. TelnetPM Cursor Modification You can use the Options menu to alter the way the cursor is displayed during your TelnetPM session. Your cursor can be defined as follows: o Character - Underscore (Default) - Solid block - Frame block o Blink - On (Default) - Off o Intensity - Normal (Default) - Dim To select cursor values for your session: 1. Select (using mouse button 1 to click on the field) Options from the menu bar. 2. Select the Cursor option. (The current default is indicated by a checkmark to the left of the option on the drop-down menu.) 3. Select the desired option. -------------------------------------------------------------------- 8. SLIP -------------------------- 8.1 The SLIP Command The SLIP command is used to start the main slip driver. Since the majority of users only use a single interface, command line options for the sl0 interface have been added to simplify making a single interface SLIP connection. These options override the default values in the configuration file and cannot be set using the Configuration Notebook (TCPIPCFG.EXE). .- -f SLIP.CFG----. >>--slip--.--------.--+-----------------+--.------------------.--> '- -comn-' '- -f config_file-' '- -hangup command-' >--.--------------------------.--.-----------------.-------------> '- -ifconfig ipaddr ipdest-' '- -modem command-' >--.------------.--.-----.--.----------.--.--------------.-------> '- -mtu size-' '- -p-' '- -rtscts-' '- -speed baud-' >--.----------------------.--.------------.--.------.----------->< '- -d--.-------------.-' '- -t--.---.-' '- -vj-' '- > filename-' |-i-| '-o-' Displaying SLIP Help >>--slip-- -?--------------------------------------------------->< -comn Specifies the communications port (n) to be used for the sl0 interface. -f config_file Specifies the name of the SLIP configuration file. The default is SLIP.CFG. -hangup command Specifies a command to send to the sl0 communications port upon exit. -ifconfig ipaddr ipdest Specifies IP addresses to use to configure sl0. -modem command Specifies a command to send to the sl0 communications port. -mtu size Sets the sl0 interface MTU to the specified size. -p Specifies that SLIP communications port and INET interface threads be placed in the highest foreground server priority. This prevents packet loss on slow machines or in heavily loaded environments. -rtscts Specifies asynchronous card-to-modem flow control via the RTS (request-to-send) and CTS (clear-to-send) handshaking signals (RTS/CTS hardware flow control). At baud rates above 2400, this option should be enabled to prevent modem buffer overruns and packet loss. -speed baud Specifies the baud rate to which to set the sl0 communications port. -d Enables debugging output. NOTE: Because this option can produce large amounts of output that can degrade performance, you may want to redirect the debugging output to a file. > filename Specifies a file to which to redirect debugging output. -t Specifies that SLIP begin tracing as follows: -t Traces both input and output packets. -ti Traces only input packets. -to Traces only output packets. Data is written to SLIPTRC.DMP in the directory from which SLIP was started. To stop the trace, press Ctrl-c or Ctrl-Break. This option produces packets in INET trace format. You can read the output using IPFORMAT. NOTES: 1. As the SLIPTRC.DMP file increases in size, the performance of your workstation can be affected. 2. SLIP does not check to ensure that there is sufficient hard disk space to record the data. -vj Specifies that SLIP is to enable Van Jacobsen (VJ) header compression for the sl0 interface. NOTE: If VJ header compression is on, you will not be able to use IPFORMAT to read the SLIPTRC.DMP file. -? Displays the list of parameters and their meanings. -------------------------- 8.1.1 SLIP Examples Example 1: slip -rtscts -modem atdt9,999-9999 -ifconfig 9.67.111.1 9.67.111.2 -speed 38400 -com2 -mtu 1006 -vj This command: - Starts SLIP using the sl0 interface - Sets RTS/CTS hardware flow control on - Sends the command string "atdt9,999-9999" to the modem (dial out) attached to comport 2 - Configures the interface with local IP address 9.67.111.1 and remote peer IP address 9.67.111.2 - Sets the baud rate to 38.4K bits - Sets the communications port to com2 - Sets the mtu to 1006 bytes - Enables Van Jacobson Header compression Example 2: slip -rtscts -modem ats0=2s7=30 -ifconfig 9.67.111.2 9.67.111.1 -speed 38400 -com1 -mtu 1006 -vj -p This command: - Starts SLIP using the sl0 interface - Sets RTS/CTS hardware flow control on - Sends the command string "ats0=2s7=30" to the modem (dial in) attached to communications port 1 - Configures the interface with local IP address 9.67.111.2 and remote peer IP address 9.67.111.1 - Sets the interface baud rate to 38.4K bits - Sets the communications port to com1 - Sets the mtu to 1006 bytes - Enables Van Jacobson Header compression - Schedules SLIP execution using time critical server priority -------------------------- 8.2 The SLIPTERM Command The SLIPTERM command now has the -f and interface parameters. .- -f SLIP.CFG----. >>--slipterm--.-------------.--+-----------------+-------------> '- -w---.---.-' '- -f config_file-' '-n-' .-sl0-. >--+-----+--.-----.------------------------------------------->< '-sln-' '- -d-' -w Invokes the SLIPWAIT utility. n Specifies the number of seconds to wait (the default is 30). -f config_file Specifies the name of the SLIP configuration file. The default is SLIP.CFG. sln Specifies the interface to be addressed. The default is SL0. -d Enables debugging mode. -------------------------- 8.3 The SLIPHOLD Command The SLIPHOLD command now has the -f and interface parameters. .- -f SLIP.CFG----. .-sl0-. >>--sliphold--+-----------------+--+-----+-------------------->< '- -f config_file-' '-sln-' -f config_file Specifies the name of the SLIP configuration file. The default is SLIP.CFG. sln Specifies the interface to be addressed. The default is SL0. -------------------------- 8.4 SLIP Correction The description of the MODE command in the Installation and Administration manual is incorrect. The speed parameter is the async-to-modem bit rate. <speed> is the async-to-modem bit rate (1200 to 38400). The async card UART (universal asynchronous receiver/transmitter) rate is not the same as the modem line rate. If the modem rate and async communications port rate are different, then data flow control is used to coordinate the async card and the modem; to prevent one from overrunning the other. This is called "pacing". Pacing is required for newer modems to use data compression. If the two rates are the same, then the modem tries to adjust its rate to that of the async communications port. This is called "bit rate adjustment". It is recommended that you use as high a bit rate as is supported by the hardware. If you are connecting to another modem, the modems need to negotiate the highest line rates possible. To do this, start SLIP with the -rtscts parameter. -------------------------- 8.5 New Sample Attachment Scripts Three new sample attachment scripts have been added for SLIP: 1. ATDIAL.CMD is a sample attachment script that can be used for a basic SLIP client that does not require a user ID and password prior to starting a SLIP connection. 2. DIRECT.CMD is a sample attachment script that can be used for a direct connection between two asynchronous lines. 3. ATANS.CMD is a sample attachment script that can be used for putting your modem into auto answer mode to enable others to dial into your host. These new samples support multiple SLIP interfaces. -------------------------- 8.6 Changes to the Way Parameters Are Passed to SLIP Attachment Scripts If the ipdest and ipaddress keywords appear in your SLIP configuration file, the interface name (slx), IP address, destination address, and dial commands are passed to the REXX attachment scripts. If the ipdest or ipaddress keywords do not appear in your SLIP configuration file, only the interface name and dial commands are passed to the REXX attachment scripts. -------------------------- 8.6.1 Example of Configuring SLIP for More Than One Interface Following is an example of a SLIP configuration file for 3 concurrent dial-in sessions: interface sl0 { mtu=296 # Use 296 byte MTU for interface 0. compression=off # Disable VJ compression. device = com1 # Use com1 for interface sl0. ipaddress = 9.67.1.1 # Local IP address. ipdest = 9.67.10.1 # Peer IP address. attachcmd = atans # Rexx command to place modem in auto # answer mode using AT command set and # ifconfig the interface. attachparms = "com1 38400" # Parameters passed to REXX # command script. } interface sl1 { mtu=296 compression=off device = com2 ipaddress = 9.67.1.2 ipdest = 9.67.10.2 attachcmd = atans attchparms = "com2 38400" } interface sl2 { mtu=296 compression=off device = com3 ipaddress = 9.67.1.3 ipdest = 9.67.10.3 attachcmd = atans attparms = "com3 38400" } -------------------------------------------------------------------- 9. SNMP Enhancements -------------------------- 9.1 The SNMP Command The Simple Network Management Protocol (SNMP) is a protocol used by network elements (for example, hosts and bridges) to exchange information about network management. The snmp command provides SNMP manager function to query SNMP agents for network management information. SYNTAX RECEIVING TRAPS .- -h localhost-. >>--snmp--.---------------------------.--+---------------+--------> | <-----------------------< | '- -h host------' | .- 1-----------. | '--- -d--+--------------+---' '- debug_level-' .- -c public---------. >--+--------------------+-- trap--------------------------------->< '- -c community_name-' SETTING MIB VARIABLE VALUES .- -h localhost-. >>--snmp--.---------------------------.--+---------------+---------> | <-----------------------< | '- -h host------' | .- 1-----------. | '--- -d--+--------------+---' '- debug_level-' .- -c public---------. .- -t 6-------. .- -p 161--. >--+--------------------+--+-------------+--+----------+-- set-----> '- -c community_name-' '- -t seconds-' '- -p port-' <------------------------------< >--- mib_variable-.-------.- value------------------------------->< '- type-' GATHERING INFORMATION ABOUT MIB VARIABLES .- -h localhost-. >>--snmp--.---------------------------.--+---------------+--------> | <-----------------------< | '- -h host------' | .- 1-----------. | '--- -d--+--------------+---' '- debug_level-' .- -c public---------. .- -t 6-------. .- -p 161--. >--+--------------------+--+-------------+--+----------+----------> '- -c community_name-' '- -t seconds-' '- -p port-' >--.- get-----.-- mib_variable----------------------------------->< |- getnext-| '- walk----' Displaying SNMP Help >>--snmp-- -?---------------------------------------------------->< -d Starts debugging. debug_level Specifies the debug level. There are currently 2 debug levels (1 and 2). Level 2 provides more detailed debug and trace information than does level 1. -h host Specifies the destination host to which you want to send a request. This can be either an internet protocol address or a host name. If you do not specify a host, the default is your local host. -c community_name Specifies the community name used to access the specified variables on the destination SNMP agent. If you do not specify a community, the default is public. NOTE: Community names are case sensitive. trap SNMP waits for a trap from the specified host. If no host is specified, SNMP waits for a trap from any host. -t seconds Specifies the amount of time (in seconds) that the SNMP client waits for a reply from the SNMP agent. The default is 6 seconds. -p port Specifies the destination (agent) port. The default is 161. set Sends a request to the SNMP agent to set a specific MIB variable. SNMP then waits for a response, a trap, or a timeout. mib_variable Specifies the MIB object, using its object descriptor (textual name), object identifier using ASN.1 notation, or a combination of the two. When used with WALK, this is the MIB object prefix. A prefix can be any leading portion of the complete object identifier. type The type of value being set. To complete an SNMP SET request, the SMI_type must be known. If no type is specified, SNMP searches the compiled MIB and the MIB2.TBL to determine the type. If the variable is not found, an error is returned. If a type is specified, the type takes precedence over any type that may be assigned in the MIB. The type and value must be compatible. For example, if you specify a type of "number" and a value of "foo," an error will be returned because "foo" is not a number. Possible variable types are: Types From MIB2.TBL ASN.1 SMI_types =================== =============== number or INTEGER display object or OBJECT_IDENTIFIER internet or IPAddress counter gauge ticks or TimeTicks Opaque NULL value Specifies the value to be set by the SET function. If white space is needed in the value, you must use double quotes ("). If you want to set a variable to a value that is also a type, you must specify the type. get Sends a request to an SNMP agent for a specific management information base (MIB) variable. SNMP then waits for a response, a trap, or a timeout. getnext Sends a request to an SNMP agent for the next MIB variable that lexicographically follows the mib_variable specified. SNMP then waits for a response, a trap, or a timeout. walk Issues a GETNEXT request for a specified prefix, then continues to issue GETNEXT requests for as long as there are variables that match the specified prefix. A prefix can be any leading portion of the complete object identifier. -? Displays help information. -------------------------- 9.2 The SNMPGRP Command The snmpgrp command provides SNMP manager function to query SNMP agents for a collection of information contained in tables and other related MIB variables. This information is retrieved using the name of the group or table of MIB variables. SYNTAX .- -h localhost-. >>--snmpgrp--.---------------------------.--+---------------+-----> | <-----------------------< | '- -h host------' | .- 1-----------. | '--- -d--+--------------+---' '- debug_level-' .- -c public---------. .- -t 6-------. .- -p 161--. >--+--------------------+--+-------------+--+----------+----------> '- -c community_name-' '- -t seconds-' '- -p port-' <-------------< >---.-----------.------------------------------------------------>< |- sys------| |- snmp-----| |- ip-------| |- icmp-----| |- udp------| |- tcp------| |- tcptab---| |- udptab---| |- iproute--| |- ipaddr---| |- arptab---| |- iftab----| '- mediatab-' Displaying SNMPGRP Help >>--snmpgrp-- -?------------------------------------------------->< -d debug_level Specifies the debug level. There are currently 2 debug levels (1 and 2). Level 2 provides more detailed debug and trace information than does level 1. If you do not specify a debug level, the default is 1. -h host Specifies the host to which you want to send a request. You can specify either an internet protocol address or a host name. If you do not specify a host, the default is your local host. -c community_name Specifies the community name of the simple network management protocol agent on the destination host. If you do not specify a community name, the default is public. NOTE: Community names are case sensitive. -t seconds Specifies the amount of time (in seconds) that SNMPGRP waits for a reply. The default is 6 seconds. - p port Specifies the destination (agent) port. The default is 161. NOTE: The following list represents the names of the groups from which you can choose. If you request more than one group, your output will be displayed in the order in which the groups appear in the list. sys Requests the System group of MIB variables. snmp Requests the SNMP group of MIB variables. ip Requests the scalar MIB variables in the IP group. icmp Requests the ICMP group of MIB variables. udp Requests the scalar MIB variables in the UDP group. tcp Requests the scalar MIB variables in the TCP group. tcptab Requests the TCP connection table in the TCP group. udptab Requests the UDP listener table in the UDP group. iproute Requests the IP routing table in the IP group. ipaddr Requests the IP address table in the IP group. arptab Requests the AT table in the address translation group. iftab Requests the Interface table in the interfaces group. mediatab Requests the IP address translation table in the IP group. -? Displays help information. -------------------------- 9.3 Changes to SNMP The SNMP and SNMPGRP commands now search an internal table first to find the requested information. If the mib_name does not exist in the internal table, SNMP searches the MIB2.TBL file for the information. The internal table contains all the textual names defined in the following RFCs: o RFC1155 o RFC1213 o RFC1231 o RFC1285 o RFC1315 o RFC1398 In the MIB2.TBL, the ASN.1 name should now always end in a dot. No instance IDs of 0 (zero) should be used. The SNMP commands have been changed to comply with the SNMP RFCs. Prior to this CSD, trailing zeros (instance 0) were used in the MIB2.TBL. Therefore, if you issued the following command: snmp getnext sysDescr the value of the sysObjectID.0 would be returned. With the new SNMP command, sysDescr is no longer obtained from the MIB2.TBL but from the internal table. In the previous example, the outgoing request is for sysDescr (without instance 0). Since the next lexicographical variable is the instance 0 of sysDescr, the value of sysDescr.0 will be returned. The new SNMP commands require you to specify the instance. For example, if you want to perform a GET or GETNEXT on the instance 0 of the sysDescr object, you would use the following commands: snmp get sysDescr.0 snmp getnext sysDescr.0 The SNMP GET command will return the value of the sysDescr object. The SNMP GETNEXT will return the value of the sysObjectID, which is the next object that lexicographically follows instance 0 of sysDescr). The following rules apply to object IDs in ASN.1 notation: o The object ID consists of 1 to 128 subIDs, which are separated by dots. o Each subID is a positive number. No negative numbers are allowed. o The value of each number cannot exceed 4294967295 (which is 2 to the power of 32 minus 1). o The valid values of the first subID are: 0, 1, or 2. o If the first subID has a value of 0 or 1, the second subID can only have a value of 0 through 39. -------------------------- 9.4 Enhancements to SNMPD TCPIPCFG (the Configuration Notebook) now includes an autostart option for SNMPD. -------------------------------------------------------------------- 10. Default 3270 Emulator Keyboard Mappings The following table lists the default keyboard mappings for PMANT and TN3270. +---------------------------------+--------------------------------+ | WORKSTATION KEY | 3270 FUNCTION | +---------------------------------+--------------------------------+ | Alt-1 | PA1 | +---------------------------------+--------------------------------+ | Alt-2 | PA2 | +---------------------------------+--------------------------------+ | Alt-3 | PA3 | +---------------------------------+--------------------------------+ | Backspace | Backspace | +---------------------------------+--------------------------------+ | Control-End | Erase-EOF | +---------------------------------+--------------------------------+ | Control-Delete | Kill-Word | +---------------------------------+--------------------------------+ | Control-F1 | PA1 | +---------------------------------+--------------------------------+ | Control-F2 | PA2 | +---------------------------------+--------------------------------+ | Control-F3 | PA3 | +---------------------------------+--------------------------------+ | Control-Left | Move backward a word | +---------------------------------+--------------------------------+ | Control-Right | Move forward a word | +---------------------------------+--------------------------------+ | Delete | Delete | +---------------------------------+--------------------------------+ | Down Arrow | Down | +---------------------------------+--------------------------------+ | End | End | +---------------------------------+--------------------------------+ | Enter | Enter | +---------------------------------+--------------------------------+ | Escape | Clear | +---------------------------------+--------------------------------+ | F1 | PF1 | +---------------------------------+--------------------------------+ | F2 | PF2 | +---------------------------------+--------------------------------+ | F3 | PF3 | +---------------------------------+--------------------------------+ | F4 | PF4 | +---------------------------------+--------------------------------+ | F5 | PF5 | +---------------------------------+--------------------------------+ | F6 | PF6 | +---------------------------------+--------------------------------+ | F7 | PF7 | +---------------------------------+--------------------------------+ | F8 | PF8 | +---------------------------------+--------------------------------+ | F9 | PF9 | +---------------------------------+--------------------------------+ | F10 | PF10 | +---------------------------------+--------------------------------+ | F11 | PF11 | +---------------------------------+--------------------------------+ | F12 | PF12 | +---------------------------------+--------------------------------+ | Home | Home | +---------------------------------+--------------------------------+ | Insert | Insert | +---------------------------------+--------------------------------+ | Left Arrow | Left | +---------------------------------+--------------------------------+ | LeftAlt (PMANT only) | Gives focus to the menu bar | +---------------------------------+--------------------------------+ | NewLine | Enter | +---------------------------------+--------------------------------+ | PageDown | PF8 | +---------------------------------+--------------------------------+ | PageUp | PF7 | +---------------------------------+--------------------------------+ | Right Arrow | Right | +---------------------------------+--------------------------------+ | RightAlt (PMANT only) | Gives focus to the menu bar | +---------------------------------+--------------------------------+ | RightControl | Enter | +---------------------------------+--------------------------------+ | ScrollLock (PMANT only) | PA3 | +---------------------------------+--------------------------------+ | Shift-End | Erase-EOF | +---------------------------------+--------------------------------+ | Shift-Enter | NewLine | +---------------------------------+--------------------------------+ | Shift-F1 | PF13 | +---------------------------------+--------------------------------+ | Shift-F2 | PF14 | +---------------------------------+--------------------------------+ | Shift-F3 | PF15 | +---------------------------------+--------------------------------+ | Shift-F4 | PF16 | +---------------------------------+--------------------------------+ | Shift-F5 | PF17 | +---------------------------------+--------------------------------+ | Shift-F6 | PF18 | +---------------------------------+--------------------------------+ | Shift-F7 | PF19 | +---------------------------------+--------------------------------+ | Shift-F8 | PF20 | +---------------------------------+--------------------------------+ | Shift-F9 | PF21 | +---------------------------------+--------------------------------+ | Shift-F10 | PF22 | +---------------------------------+--------------------------------+ | Shift-F11 | PF23 | +---------------------------------+--------------------------------+ | Shift-F12 | PF24 | +---------------------------------+--------------------------------+ | Shift-NewLine | NewLine | +---------------------------------+--------------------------------+ | Shift-Tab | Backtab | +---------------------------------+--------------------------------+ | Tab | Tab | +---------------------------------+--------------------------------+ | Up Arrow | Up | +----------------------------------+-------------------------------+ | NOTE: LeftAlt, RightAlt, and ScrollLock are PMANT defaults | | only. They are not valid for TN3270 sessions. | +------------------------------------------------------------------+ -------------------------------------------------------------------- 11. The DPKT Utility The DPKT utility will aid in servicing TCP/IP. >>--dpkt-- n------------------------------------------------->< n Specifies a number indicating a particular packet within the IPTRACE.DMP file created by the IPTRACE.EXE utility. The number you specify must represent a valid packet number. -------------------------------------------------------------------- 12. Configuration Notebook -------------------------- 12.1 The TCPIPCFG Command The TCPIPCFG command opens the TCP/IP configuration notebook. With the configuration notebook, you can configure your TCP/IP network connections and customize your TCP/IP applications according to your preferences. >>--tcpipcfg--.-----.----------------------------------------->< '- -d-' -d Specifies that tracing will begin, and upon exiting, creates a TCPIPCFG.LOG text file containing debug data. Data recorded in the TCPIPCFG.LOG includes: o Opening files o Reading-in and initialization of data o Writing data to files o Closing files -------------------------- 12.2 Changes to the Configuration Notebook The Configuration Notebook now includes the following additions and changes: 1. You can now specify that SNMPD is to autostart. 2. You can now specify parameters when you autostart REXECD. 3. The default for 'Codepage' is now 'None'. 4. You can now specify 5-digit values for the MTU size field. -------------------------------------------------------------------- 13. Configuring a Token-Ring Adapter using BOOTP To configure a token-ring adapter using BOOTP, ensure that the corresponding entry in the BOOTPTAB file has a hardware type (ht) of token-ring. For example: frank.tcp.raleigh.ibm.com: tc=subnet22.dummy: ht=token-ring:\ ha=10005a2f341b: ip=9.67.111.130: hn: -------------------------------------------------------------------- 14. SETTERM The -f parameter description has been clarified. >>--setterm--.-------.--.-----------.-->< '- -fs -' '- -f cfile-' -f cfile Instructs SetTerm to access the cfile.CFG configuration file in the ETC subdirectory for the SetTerm definitions and display a menu to allow changes to the file. If you issue the SETTERM command without this parameter, you access the TELNET.CFG file in the ETC subdirectory. -------------------------------------------------------------------- 15. SendMail Server and Client SendMail can be used in one of two modes, as a server or as a client. The server mode is used for receiving mail, and the client mode is used for sending mail. -------------------------- 15.1 Server If both modes of SendMail are needed (which normally is the case), the SendMail server must always be started first. Following are examples of commands to start the SendMail server: sendmail -bd start the SendMail server sendmail -q30m -bd start the SendMail server and process the mail queue every 30 minutes sendmail -d1.1 -q30m -bd start the SendMail server, process the mail queue every 30 minutes, and log connections in the SENDMAIL.LOG file Note: The SENDMAIL.LOG is not a message file. You should use the -d1.1 parameter only for debugging purposes, because it records all details of the client-server handshaking procedure (as described in RFC 821) in the SENDMAIL.LOG. The SendMail server functions as follows: 1. The command with a -bd flag starts a SendMail server which sets up the SendMail environment, copies the SENDMAIL.CF file to a shared memory and then waits for requests from SendMail clients to establish connections and to deliver mail. 2. When a request for connection arrives, the server initiates another instance of SendMail (a SendMail child process) which initializes itself from the configuration information in the shared memory, and then services the connection and receives mail. 3. Once the mail has been received and properly delivered to its destination, the SendMail child process closes the connection and exits. The number of SendMail child processes that can run at any one time is controlled by a Load Limiting variable which by default is 6, but can be changed by the OX option in the SENDMAIL.CF file. -------------------------- 15.2 Client The SENDMAIL client command now has additional parameters. The SendMail client mode is usually started by a User Mail Agent, but can also be started from an OS/2 Command line. SYNTAX >>--sendmail--.-----.--.------.--.----------------.---------------> '- -q-' '- -bt-' '- -Cmyconfig.CF-' >-- -af filename--.- -f user@localhost user@remotehost-.----------> '- -t--------------------------------' >--.--------.---------------------------------------------------->< |- -d1.1-| '- -d----' -q Instructs SendMail to process the mail queue. If you include other parameters, SENDMAIL will ignore them. -bt Instructs SendMail to run in rule-testing mode. You can use this parameter to test changes to rules in the configuration file. In rule-testing mode, SendMail prompts you to enter a rule-set number and address, and then displays how the rule-set causes the address to be rewritten. SendMail Version 1.3.0 always calls rule-set 3 first and then the entered rule-set number. In rule-testing mode, SendMail does not deliver any mail. This parameter cannot be used with any other parameter. -Cmyconfig.CF Specifies the name of a configuration file to use instead of SENDMAIL.CF. -af filename Specifies the name of the file that contains the mail message. -f user@localhost Identifies the user and host name of the sender. user@remotehost Identifies the user and host name of the receiver (or destination). NOTE: OS/2 is not a multiuser operating system. The user field of an address in an item of mail sent to an OS/2 system is not significant; however, it is required. The significant part of the address is the host name, with the domain name expansion if domain names are used. -t Specifies that SENDMAIL retrieves the to and from information for the mail from the body of the file rather than from the command line. -d1.1 Creates only the SENDMAIL.LOG file. The SENDMAIL.LOG file is placed in the ETC directory. NOTE: The SENDMAIL.LOG is not designed to keep track of messages you send and receive. You should use the -d1.1 parameter only for debugging purposes, because it records all details of the client-server handshaking procedure (as described in RFC 821) in the SENDMAIL.LOG. -d Writes detailed debugging information to the SENDMAIL console and creates a SENDMAIL.LOG file. NOTE: The -d parameter produces a great deal of information, and for performance reasons, should be used only for debugging. Examples of the SendMail client startup command are: sendmail -af file -f sender recipient (Send a message using the SendMail client) sendmail -d1.1 -af file -f sender recipient (Send a message using the SendMail client and log the connection in the SENDMAIL.LOG file) sendmail -Cmyconfig.cf -af file -f sender recipient (Send a message using the SendMail client and also using MYCONFIG.CF instead of SENDMAIL.CF) The SendMail client also uses shared memory for Configuration information, if present, otherwise it gets this information by reading the SENDMAIL.CF file. -------------------------- 15.3 Considerations If SendMail is used for sending a message to several recipients on the same workstation, the number of recipients should be limited to 150. A message for 150 recipients will be received as several files with lesser number of recipients in each (the number of recipients is limited by the header buffer space; recipients would be added to a header until that header ran out of room, then the process would begin again with a new header). If sending a message to 30 or fewer recipients, the whole file may be transmitted in one transaction. The processing time for sending a message to several recipients is much longer than that for a single recipient. Potentially, the processing time may exceed the time-out of the client. Time-out is determined by the idle time on the socket, and is typically measured from the last activity on the socket. On a UNIX system, the client time-out is usually 2 minutes; on an OS/2 system, it is usually 5 minutes. If a client times out, the connection is not closed properly, and SendMail is not made aware of the completion status of the message. Sending a message to more than 150 recipients can cause the network connection to time out. Therefore, the number of recipients for a given message should be limited to 150. On a busy server handling multiple transactions, you run the risk of not servicing a given connection fast enough, thereby allowing the client that established the connection to time out. The client, not aware of the completion status of the message, then requeues the message for later delivery even though it may have been delivered already. At a later time, the client delivers the same message again, and if the connection times-out again, the client requeues the message again. To prevent this undesirable looping, a network connection cannot be allowed to time out. It is, therefore, imperative to use a server that has sufficient processing capability to handle the mail delivery load. To send mail to large numbers of recipients, the minimum size of machine you should consider is a 33 Megahertz machine with 16 MB RAM, HPFS and DISKCACHE to enhance the disk performance. -------------------------------------------------------------------- 16. REXEC and RSH Redirection A redirection facility has been added for REXEC and RSH. -------------------------- 16.1 REXEC The REXEC command sends a single command to the remote host for execution. >>--rexec-- host--.---------------.--.--------------.--.-----.---> '- -l loginname-' '- -p password-' '- -k-' >--.-----.--.-----.--.-----.-- command--.-------------.--------->< '- -n-' '- -b-' '- -i-' '- < filename-' -b Transfers data to and from the remote host using binary mode. If you do not use this switch, REXEC transfers the data using ASCII mode. If you omit this switch, REXEC will strip the CR from each CRLF pair when sending data to the remote host and will add a CR to each LF on data received from the remote host. -i Uses interactive mode for input. REXEC reads input directly from the keyboard instead of from standard input (stdin). < filename Specifies the name of a file containing input to be used by the command being issued on the remote host. Your filename can also include the path. In cases where a command requires or accepts additional input, you can use this parameter to specify that the input comes from a file rather than from the keyboard. -------------------------- 16.2 RSH The RSH command issues a specified command at the remote host. .- -l user environment variable-. >>--rsh--host--+-------------------------------+-----------------> '- -l loginname-----------------' .- -u value of -l argument-. >--+--------------------------+--.-----.--.-----.--.-----.-------> '- -u local_loginname------' '- -n-' '- -b-' '- -i-' >--.----------.--.-------------.-------------------------------->< '- command-' '- < filename-' -b Transfers data to and from the remote host using binary mode. If you do not use this switch, RSH transfers the data using ASCII mode. If you omit this switch, RSH will strip the CR from each CRLF pair when sending data to the remote host and will add a CR to each LF on data received from the remote host. -i Uses interactive mode for input. RSH reads input directly from the keyboard instead of from standard input (stdin). < filename Specifies the name of a file containing input to be used by the command being issued on the remote host. Your filename can also include the path. In cases where a command requires or accepts additional input, you can use this parameter to specify that the input comes from a file rather than from the keyboard. -------------------------------------------------------------------- 17. Programming Support -------------------------- 17.1 REXX File Transfer Protocol API The REXX File Transfer Protocol (FTP) Application Program Interface (API) package provides access to the OS/2 TCP/IP FTP APIs. It is assumed you are familiar with the basic FTP. The REXX FTP API function names are similar to the FTP subcommands. For more information and installation instructions, view RXFTP. -------------------------- 17.2 REXX Socket Support The REXX Socket Support package provides access to the OS/2 TCP/IP socket APIs. It is assumed you are familiar with the basic socket APIs and that you can reference the OS/2 specific ones. For more information and installation instructions, view RXSOCKET. -------------------------------------------------------------------- 18. Applications Kit / ANYNET Users Information If you are an Applications Kit or ANYNET user: 1. SO32DLL.DLL and TCP32DLL.DLL are required to run your applications. 2. These files will be shipped with the NEXT version of ANYNET. 3. You can unzip the files from BASEOC1.ZIP on disk 4 of CSD UN56401. ************************************* ***************** UN50382 CORRECTIVE SERVICE DISKETTE ******************* ************************************* Contents of CSD UN50382: 1. Running TCP/IP on ValuePoint systems 2. Using addtional autostart files 3. Using NR/2 and LaMail 4. Using an MTU greater than the default 5. SLIP information returned by NETSTAT 6. Outbound FAX support 7. LPR port driver filter support 8. The tcpinst Command 9. The tcpdsktp Command 10. New Response File Options 11. Changes to the CID Installation Process 12. Adding Additional Device Drivers 13. The ftpd Command 14. The rexec Command 15. The rsh Command 16. SENDMAIL.CF Addition 17. The slip Command 18. 3270 Emulator Keyboard Mappings 19. TN5250 Emulator Changes 20. The telnetd Command 21. Packet Tracing Programs 22. Local Setup to Run TCPIPCFG from a LAN 23. Network Security Program Support -------------------------------------------------------------------- 1. Running TCP/IP on ValuePoint Systems If you run the TCP/IP protocol stack in a ValuePoint system using an IBM Token-Ring 16/4 Adapter II, you may receive a TRAP 000D error in IFNET.SYS. The protocol stack will not activate. This is an error in the MAC driver for the adapter. If this error occurs, contact your IBM representative and request the latest MAC driver for the adapter. -------------------------------------------------------------------- 2. Using Additional Autostart Files You can use two command files to start your own processes from the TCP/IP autostart command file (TCPSTART.CMD). One of the command files (B4TCP.CMD) is called before TCP/IP starts its applications, the other command file (TCPEXIT.CMD) is called after TCP/IP starts. You can create these files using a system editor. They should be stored in the TCPIP\BIN subdirectory. NOTE: As these files are separate from the TCPSTART.CMD file, they are not overwritten when you reinstall TCP/IP or when you modify TCPSTART.CMD. B4TCP.CMD: Use this command file to start any processes you want to run before the TCP/IP applications are started. For example, you could use this file to start another protocol stack. TCPEXIT.CMD: Use this command file to start any processes that require TCP/IP to be running. For example, you could use this file to start an OFFLOAD procedure. -------------------------------------------------------------------- 3. Using NR/2 with LaMail You might experience problems when you save an NR/2 file to a LaMail folder. The function will appear to work (no error message is displayed), but the post does not appear in the folder. The files may appear in the root directory of the drive where NR/2 and LaMail are installed (for example, in a file named 61664846.NR2). A temporary solution for this is: 1. Close NR/2 2. From the LaMail action bar, select the Set Application Options item from the Options pull-down menu. a. Take note of the paths in the Folders and In Basket fields. b. Change these two paths to any valid path. c. Select Ok. If the paths are not valid, an error message will be displayed. 3. Then again from the LaMail action bar, select the Set Application Options item from the Options pull-down menu. a. Enter the paths that you noted earlier in the Folders and In Basket fields. b. Select Ok. 4. Restart NR/2. -------------------------------------------------------------------- 4. Using an MTU Greater than the Default You can configure an interface maximum transmission unit (MTU) for a Token-Ring adapter that is greater than the default of 1500. This allows TCP to send packets that have an MSS larger than 1460 (1500-40, where 40 is the maximum TCP/IP header size). For performance reasons, TCP sets the MSS for a connection to be a multiple of 2048 (2K) bytes of data. If you want configure TCP to use larger packets, use the IFCONFIG command to specify an MTU of at least 40 bytes greater than a multiple of 2048. For example: MSS + HDR = MTU ----------------------- (1 * 2048) + 40 = 2088 (2 * 2048) + 40 = 4136 In the first example, if you specify an MTU greater than or equal to 2088, but less than 4136, TCP will try to use an MSS of 2049 (2K). In the second example, if you specify an MTU greater than or equal to 4136, TCP will try to use an MSS of 4096 (4K). For TCP/IP to be able to send these packets, you must also configure your Token-Ring adapter to support transmission of buffers that are at least the size that you specified as the MTU. You can do this either through LAPS, or by editing the PROTOCOL.INI file to set the XMITBUFSIZE parameter equal to or greater than the MTU. -------------------------------------------------------------------- 5. SLIP Information returned by NETSTAT If you start SLIP at 76800 baud using the SIO.SYS and VSIO.SYS COM port drivers, the NETSTAT command can return incorrect information. Under these circumstances, the "netstat -n" command will return an incorrect response of 1200 BPS for the SLIP speed. -------------------------------------------------------------------- 6. Outbound FAX Support TCP/IP Version 2.0 for OS/2 provides outbound FAX support through the use of an LPR Port Driver filter. Any application that can print to an OS/2 Print Object can FAX it's printed output with no additional software at the client. SOFTWARE REQUIREMENTS: To use the outbound FAX support, you must have a copy of FaxWorks for OS/2 at each server. At each client workstation, install the IBM Proprinter Printer Driver, which is supplied with the OS/2 operating system. CONFIGURING THE LPR PORT DRIVER: To configure an LPR port driver, do the following: 1. Create an OS/2 Work Place Shell print object using the printer icon in the Templates folder. 2. Associate the IBM Proprinter Printer driver with that print object. 3. Specify an LPD Port on the Output page of the Settings notebook. 4. Configure the LPD Port icon to specify the name and printer of an OS/2 LPD server running FaxWorks for OS/2. You will also need check the 'Enable data filter' check box, and fill in the 'Filter' entry with a PMFAXW command. For more information, see the PMFAXW.EXE filter in the section entitled "LPR Port Driver Filter Support" later in this file. Once this configuration is complete, any application that can print to that printer object can produce outbound faxes through an OS/2 LPD server. LIMITATIONS: The following limitations exist with the FAX support: o Only outbound FAX through a server is supported. There is no support for receiving a FAX and distributing it to the corresponding client. o The print resolution using the IBM Proprinter printer driver is not as fine the resolution you can get using the FaxWorks for OS/2 Printer Object directly. -------------------------------------------------------------------- 7. LPR Port Driver Filter Support NOTE: To view the figures in this section, please select the System Monospace font for your editor. If you use the E editor, select Set font from the Options pull-down menu. In the EPM editor, select Settings from the Preferences item on the Options pull-down menu. Select the System Monospace font on the Fonts page. Two new fields have been added to the LPD Port Icon Settings window. These fields control the use of a LPR Port Driver Filter. An LPR Port Driver Filter allows the user to specify a program that can filter the data that is sent to the LPD Server specified in the 'Print Destinations' section of the Settings window. These two fields consist of a checkbox with the label 'Enable data filter', and an entry field labeled 'Filter'. ┌──┬──────────────────────────────────────────────────────────────────────┐ │..│ \PIPE\LPD<x> - Settings │ ├──┴──────────────────────────────────────────────────────────────────────┤ │ │ │ ┌─┐ │ │ ┌─ Print Destination ───────────┐ │ │ Display port settings on print │ │ │ ┌────────────────┐│ └─┘ │ │ │ LPD server │ ││ ┌─┐ │ │ │ └────────────────┘│ │ │ Enable data filter │ │ │ ┌────────────────┐│ └─┘ ┌────────────────────────┐ │ │ │ LPD printer │ ││ Filter │ │ │ │ │ └────────────────┘│ └────────────────────────┘ │ │ └───────────────────────────────┘ │ │ A filter specifies a command run by the LPD Port Driver. This command receives the print data as input (through standard input). The output that the command produces (to standard output) is then sent to the LPD Server. This allows the user to specify programs that convert one form of data to another for use by a specific device. The filter program doesn't have to be an executable; it can be anything that the command interpreter (CMD.EXE) can execute. For example, you can use the following REXX command to add a time and date stamp to the beginning of the data being printed: /* DATESTMP.CMD: Adds a time and date stamp to the beginning of a file */ say 'Printing: 'date('W') date('N') 'at' time('C') do while stream('STDIN:','S') == 'READY' say linein() end exit 0 Note that the program ends in 'exit 0'. This is necessary because if the filter returns anything other than a '0', the LPR Port Driver interprets it as an error and will not send the job to the server to be printed. To specify an LPD Port Filter, select the 'Enable data filter' checkbox in the Settings window for the corresponding LPD Port icon, and enter a filter command in the Filter field. You can enter parameters for the program specified in the Filter field. You can also pass any keywords that are defined for use in the 'Additional Control Cards' field (see the "Users Guide" for more information about 'Additional Control Cards' keywords). For example, if you specify 'DATESTMP <document>', and modify the above DATESTMP.CMD with the following changes: /* DATESTMP.CMD: Adds a time and date stamp to the beginning of a file */ parse arg filename say 'Printing: 'filename' on 'date('W') date('N') 'at' time('C') do while stream('STDIN:','S') == 'READY' say linein() end exit 0 You will get a timestamp that also has the actual name of the file being printed. ┌──┬──────────────────────────────────────────────────────────────────────┐ │..│ \PIPE\LPD<x> - Settings │ ├──┴──────────────────────────────────────────────────────────────────────┤ │ │ │ ┌─┐ │ │ ┌─ Print Destination ───────────┐ │ │ Display port settings on print │ │ │ ┌────────────────┐│ └─┘ │ │ │ LPD server │print_server_1 ││ ┌─┐ │ │ │ └────────────────┘│ │x│ Enable data filter │ │ │ ┌────────────────┐│ └─┘ ┌────────────────────────┐ │ │ │ LPD printer │text_printer ││ Filter │DATESTMP <document> │ │ │ │ └────────────────┘│ └────────────────────────┘ │ │ └───────────────────────────────┘ │ │ We are providing two filters: NULFLTR.EXE Pass through filter that adds whatever parameters are passed to it at the beginning of the data file. PMFAXW.EXE A filter which supports remote fax'ing through a server running OS/2 TCP/IP's LPD.EXE and FaxWorks OS/2 V1.30 from SofNet. When using this filter, you must select IBM Proprinter as the corresponding Printer Object's printer driver. This filter will pop up a dialog box when you print a job, and will allow you to specify both 'To' and 'From' information. To: Name - Name of person to receive fax, which appears on the cover page. Company - Company name of person receiving fax, which appears on the cover page. Fax # - Fax number to fax file to. This is a required field, and is where you specify the phone number of the destination fax machine. From: Name - Your name which appears on the cover page. Company - Your companies name which appears on the cover page. Phone # - Your phone number, which appears on the cover page. Fax # - Your fax number, which appears on the cover page. Comment - Cover page comment. Header - If specified, defines the page header, and implicitly enables page headers. By passing parameters to PMFAXW.EXE, you can specify defaults for any or all of the above fields. Following is a list of parameters that can be passed to PMFAXW: >>TO=<name>,<company>,<faxnumber> >>FROM=<name>,<company>,<phonenumber>,<faxnumber> >>INFO=<comment>,<heading> By passing any or all of the above parameters you can default the corresponding field in the PMFAXW dialog box. If you do not wish to specify all of the fields on a particular parameter, you can leave that entry blank, but the number of comma's (',') must stay the same. Example: PMFAXW ">>FROM=JP Doe,Widget's,,(xxx)yyy-zzzz" ">>INFO=,<document>" The previous example produces the following dialog box when ever you go to print a file to the corresponding Printer Object. NOTE: The field with <document> would be filled in with the corresponding document name actually being printed. ┌────────────────────────────────────────┐ │ Fax Works Filter │ ├────────────────────────────────────────┤ │ │ │ ┌─ To ────────────────────────────┐ │ │ │ ┌──────────────────┐ │ │ │ │ Name │ │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Company │ │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Fax # │ │ │ │ │ │ └──────────────────┘ │ │ │ └─────────────────────────────────┘ │ │ │ │ ┌─ From ──────────────────────────┐ │ │ │ ┌──────────────────┐ │ │ │ │ Name │JP Doe │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Company │Widget's │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Phone # │ │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Fax # │(xxx)yyy-zzzz │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Comment │ │ │ │ │ │ └──────────────────┘ │ │ │ │ ┌──────────────────┐ │ │ │ │ Header │<document> │ │ │ │ │ └──────────────────┘ │ │ │ └─────────────────────────────────┘ │ │ │ │ ┌───────────┐ ┌───────────┐ │ │ │ OK │ │ CANCEL │ │ │ └───────────┘ └───────────┘ │ │ │ └────────────────────────────────────────┘ -------------------------------------------------------------------- 8. The tcpinst Command The TCPINST command has changes to let you: o Specify that TCP/IP start from either the startup folder or the STARTUP.CMD file o Specify individual services that will start when TCP/IP starts. o Specify where to write logging information. o Specify that TCPINST will make the necessary changes to the CONFIG.SYS file. >>──tcpinst──┬───────┬───────────────────────────────────────────> └─ /sf──┘ >──┬──────────────────────────────────────────┬──────────────────> | └─ /srv:"service1,service2, ... servicen"──┘ >──┬───────────────────────────────┬─────────────────────────────> └─ /l1:path:\filename.extension─┘ >──┬───────────────────────────────┬──┬─────┬──┬──────┬─────────>< └─ /l2:path:\filename.extension─┘ └─ /c─┘ └─ /a──┘ /sf- Specifies that TCPSTART will be added to the STARTUP.CMD file rather than to the startup folder. If you omit this option, TCPSTART is added to the startup folder and not the STARTUP.CMD file. Note: This option causes the /sf- option to be passed to TCPDSKTP, which then adds TCPSTART to the STARTUP.CMD file. /srv:"service1,service2, ... servicen" Enter /srv: followed by a list of names of TCP/IP services, enclosed in quotation marks, that you want started from TCPSTART.CMD. If you enter more than one service name, separate the service names with commas. You can select service names from this list: inetd ftpd tftpd pmx talkd portmap nfsd nfsstart routed nr2 lamail lprportd sendmail /l1:path\filename.extension Writes logging information to the specified file. For example: tcpinst /l1:d:\tmp\tcpip.log Writes logging information to the file d:\tmp\tcpip.log. Typically the path points to a location on a code server so that a network administrator can access the log if a failure occurs. /l2:path\filename.extension Writes logging information to the specified file. For example: tcpinst /l2:d:\tmp\tcpip2.log Writes logging information to the file d:\tmp\tcpip2.log. /c Causes TCPINST to make the necessary changes to CONFIG.SYS, but does not actually install the product. This is useful if your CONFIG.SYS gets erased during the installation of OS/2. (Note that the installation exits must be in SOURCE_PATH because the exits actually modify CONFIG.SYS.) /a- Specifies that the installation will be done unattended. The TCP/IP installation window will be displayed at the target workstation, but no action is required of the user. -------------------------------------------------------------------- 9. The tcpdsktp Command The TCPDSKTP command lets you start TCP/IP from either the startup folder or the STARTUP.CMD file. >>──tcpdsktp──┬───────┬─────────────────────────────────────────>< └─ /sf──┘ /sf- Specifies that TCPSTART will be added to the STARTUP.CMD file rather than to the startup folder. If you omit this option, TCPSTART is added to the startup folder and not the STARTUP.CMD file. -------------------------------------------------------------------- 10. New Response File Options There are new options you can put in a response file. TCP_SERVICES Lets you specify which TCP/IP services to start automatically when you start TCP/IP itself. This is the same as coding the /srv: option on the TCPINST command. STARTUP_FOLDER=Y|N Lets you specify whether TCP/IP starts from the STARTUP.CMD file or the startup folder. This is the same as using the /sf- option on the TCPINST command. And there is a change to the EXEC=BASE line in the default response file to work with these new options. -------------------------- 10.1 TCP_SERVICES=service1,service2, ... servicen Use this option in a response file to specify TCP/IP services to start when TCP/IP starts. You can select service names from this list: inetd ftpd tftpd pmx talkd portmap nfsd nfsstart routed nr2 lamail lprportd sendmail -------------------------- 10.2 STARTUP_FOLDER=Y|N You can add the STARTUP_FOLDER=Y|N option to the response file to specify where TCP/IP starts from. STARTUP_FOLDER=N The STARTUP_FOLDER=N option in a response file adds TCPSTART to the STARTUP.CMD file. While this option: STARTUP_FOLDER=Y adds TCPSTART to the Startup folder. -------------------------- 10.3 EXEC=BASE Line If you will install this CSD using a CID installation, be sure to merge your current default response file with the default response file shipped on this CSD. Then look at your default response file to ensure that the TCP_SERVICES and STARTUP_FOLDER options have been added to the EXEC=BASE entry like this: EXEC = BASE call basext BOOT_DRIVE TARGET_PATH MPTN_INSTALLED NTS2_DIR HOSTNAME IP_ADDR NETMASK ROUTE DNSDOMAIN NAMESERVER STARTUP_FOLDER TCP_SERVICES -------------------------------------------------------------------- 11. Changes to the CID Installation Process TCP/IP Version 2.0 for OS/2 has added support for installing TCP/IP remotely using a software distribution manager (SDM). These changes include: o Changes to the TCPINST and TCPINST2 commands o Changes to the DEFAULT.RSP used by SDM o Corrections to "Installation and Configuration Parameters" -------------------------- 11.1 Changes to the TCPINST and TCPINST2 Commands To enable you to install TCP/IP using an SDM, the following parameters have been added to the TCPINST and TCPINST2 commands: o /a- Specifies that the install is to be performed on an unattended basis. The TCP/IP Installation window will be displayed at the target workstation, but no action will be required on the part of the user. o /l1:<fully-qualified_path> Specifies the fully-qualified path (drive, subdirectories, and file name) to which TCP/IP will write a log. The path typically points to a location on the code server so an administrator can access the log in case of a failure. -------------------------- 11.2 Changes to the DEFAULT.RSP Used by SDM Some of the parameters included in the TCP/IP DEFAULT.RSP file are incompatible with the SDM parameters. Source path and response path are determined at the time of installation. Also, software distribution managers use a common set of parameters to specify variables such as a log path. Therefore, if you install TCP/IP using an SDM, use the TCPINST command line equivalent parameters to specify the following: RSP_FILE SOURCE_PATH LOG_PATH LAPS_EXEC_PATH LAPS_LOG_PATH -------------------------- 11.3 Corrections to "Installation and Configuration Parameters" This section discusses corrections to the information in the section titled "Installation and Configuration Parameters" of the TCP/IP Version 2.0 for OS/2: Installation and Configuration manual. In the printed book, this section starts on page 15. In the BookManager book, this section is topic 2.6.3. The following is a corrected description of the LAPS parameter: LAPS_RSP_FILE=<laps_response_file> Specifies the fully-qualified path (drive, subdirectories, and file name) on the code server of the response file to be used when LAPS is installed. This is optional. You can specify this information in your CONFIG.SYS or as a parameter of the installation command. The following is a list of parameters that were not documented: HPFS_NEEDED=<name> Specifies the abbreviated name of the kit or component being installed that requires a high performance file system (HPFS) drive. More than one kit or component can be specified, separated by spaces. This parameter is required if you are installing kits that require HPFS. For a list of kit and component names, see the table of Abbreviated Names in the section titled "Setting up the Code Server" of the TCP/IP Version 2.0 for OS/2: Installation and Configuration manual. EXEC=<name> CALL <exit_routine> BOOT_DRIVE TARGET_PATH Specifies the exit routine to be used for each kit or component. The exit routines are named: <name>XT.EXE, where <name> is the abbreviated name of the kit or component. This parameter is required for each kit being installed. For a list of kit and component names, see the table of Abbreviated Names in the section titled "Setting up the Code Server" of the TCP/IP Version 2.0 for OS/2: Installation and Configuration manual. For additional information about installing a program using an SDM, see the documentation for the software distribution manager that you are using. -------------------------------------------------------------------- 12. Adding Additional Device Drivers Use this procedure to add additional Drivers to TCP/IP. You can use this procedure for any card not listed in the default LAPS configuration. 1. Install LAPS. Select one of the device drivers listed and save it. 2. From the adapter-card vendor, get the NDIS Drivers (the *.OS2 and *.NIF files), and manually copy them into the IBMCOM\MACS subdirectory. 3. Rerun LAPS to configure it and select the added driver. 4. If the new adapter is not listed in LAPS, edit the .NIF file to the same format as one supplied by IBM, or contact the adapter-card vendor and get one that will work with IBM's TCP/IP. -------------------------------------------------------------------- 13. The ftpd Command The FTPD command now has a -c option, that lets you specify a codepage, and a -p option that lets you specify a port number. >>──ftpd──┬──────────────┬──┬────────────────┬──────────────────>< └─ -p <portno>─┘ └─ -c <codepage>─┘ -p <portno> Specifies the port that you want FTPD to use. If you omit this option, FTPD uses its well-known port assignment. -c <codepage> Specifies the ASCII codepage used by the remote host. If you do not specify a codepage, the data is sent untranslated. The possible values are: TCPDECMU for the DEC Multinational codepage TCP8859 for ISO 8859 codepage NONE for PC codepage 850 -------------------------------------------------------------------- 14. The rexec Command The REXEC command adds the -b and -i switches. >>──rexec──host──┬───────────────┬──┬──────────────┬──┬─────┬────> └─ -l loginname─┘ └─ -p password─┘ └─ -k─┘ >──┬─────┬──┬─────┬──┬─────┬──command───────────────────────────>< └─ -n─┘ └─ -b─┘ └─ -i─┘ -b Transfers data to and from the remote host using binary mode. If you do not use this switch, REXEC transfers the data using the ascii mode. If you omit this switch, RSH will strip the CR from a CRLF pair when sending data to the remote host and will add a CR to the LF on data received from the remote host. -i Uses interactive mode for input. REXEC reads input directly from the keyboard instead of from standard input (stdin). -------------------------------------------------------------------- 15. The rsh Command The RSH command adds the -u, -b, and -i switches. ┌─ -l user environment variable─┐ >>──rsh──host──┼───────────────────────────────┼─────────────────> └─ -l loginname─────────────────┘ ┌─ -u value of -l argument─┐ >──┼──────────────────────────┼──┬──────┬──┬──────┬──┬──────┬────> └─ -u local_login_name─────┘ └─ -n ─┘ └─ -b ─┘ └─ -i ─┘ >──┬──────────┬─────────────────────────────────────────────────>< └─ command─┘ -u local_login_name Specifies the user ID on the local host. If you omit this value, the value of the -l argument is used by default. For example, to use RSH with a VM remote host, you would use -l for the VM password and -u for the VM user ID. -b Transfers data to and from the remote host using binary mode. If you do not use this switch, RSH transfers the data using the ascii mode. -i Uses interactive mode for input. RSH reads input directly from the keyboard instead of from standard input (stdin). -------------------------------------------------------------------- 16. SENDMAIL.CF Addition You can add the OT option to the Options section of your SENDMAIL.CF file. YOU MUST use an editor (such as the E.EXE editor) that preserves tabs! Use the OT option to tell the SENDMAIL program at your workstation how to process mail to be delivered to a host that is currently not receiving. If you omit the OT option, the mail is deleted from the queue when you next start SENDMAIL. You can use the OT option to tell SENDMAIL to keep the mail on its queue, and to continue trying to resend it, for a specified period of time. The format of the OT option is: OTxt OT Code the characters OT as shown. x This value tells SENDMAIL for how long to keep the mail in its queue for re-transmission. Replace the "x" with a number specifying the number of hours, days, minutes, or weeks. t This value tells SENDMAIL how often to resend to a server that failed. Replace the "t" with a unit of time. w The x value is the number of weeks. d The x value is the number of days. h The x value is the number of hours. m The x value is the number of minutes. Here is an example of the OT option that tells SENDMAIL to keep the undelivered mail in its queue, and to keep trying to resend it, for 8 hours before deleting it from the queue. ##################################################################### # Options Timeout parameter: # # Causes notes, sent to a server that was down, to be queued # # for retry. OTxt Where x is a number and t is the timeout char: # # h-hours, d-days, m-minutes. # # Without the OTxt specification, the message does not get queued. # ##################################################################### OT8h -------------------------------------------------------------------- 17. The slip Command The SLIP command now has a -t operand to start packet tracing. >>──┬─────┬──┬─────┬────────────────────────────────────────────>< └─ -d─┘ └─ -t─┘ -t Specifies that the SLIP command start tracing packets. The -t option starts SLIP tracing all packets received from and sent to the modem. The data is written to SLIPTRC.DMP in the directory from which SLIP was started. To stop the trace, press the Ctrl-c or Ctrl-Break keys. Because SLIP will record all traffic sent and received, the SLIPTRC.DMP file can grow quite large. There is no error checking to see if there is enough hard disk space available to continue recording the information. You should use this trace option in a limited fashion so that it does not impact the performance of the workstation where it is being run. -------------------------------------------------------------------- 18. 3270 Emulator Keyboard Mappings The User's Guide lists the OS/2 keys that you can remap to different functions. Following are changes to the names of the OS/2 keyboard keys that you can remap. -------------------------- 18.1 Key Name You need to use the full name of the key for these keys. Change To Del Delete Esc Escape Ins Insert PgDn Pagedown PgUp Pageup Scrlock Scroll-lock -------------------------- 18.2 TN3270 Emulator KEYS YOU CANNOT REMAP: For TN3270, you cannot remap these OS/2 keys: Backtab Space ScrLock Pause Printscreen All numeric keypad keys except the Enter key KEYS YOU CAN REMAP: The User's Guide incorrectly lists the OS/2 keys that you can remap. It neglects to add the various states of the OS/2 function keys. The following is a corrected list of the OS/2 keys and key combinations that you can remap. o F1 through F12 o Control-F1 through Control-F12 o Alt-F1 through Alt-F12 o Shift-F1 through Shift-F12 o Control-'unshifted_keypress' o Alt-'unshifted_keypress' o Shift-OS/2_function_key o Alt-OS/2_function_key o Ctrl-OS/2_function_key o Any OS/2_function_key For TN3270, these are the OS/2_function_keys that you can remap: Backspace Left-Alt Delete Left-Control Down Newline (Enter key on alpha keypad) End Pagedown Enter (Enter key on numeric keypad) Pageup Escape Right F1-F12 Right-Alt Home Right-Control Insert Up Left Tab -------------------------- 18.3 PMANT Emulator KEYS YOU CANNOT REMAP: For PMANT, you cannot remap these OS/2 keys: Backtab Printscreen All numeric keypad keys except the Enter key KEYS YOU CAN REMAP: The User's Guide incorrectly lists the OS/2 keys that you can remap. It neglects to add the various states of the OS/2 function keys. The following is a corrected list of the OS/2 keys and key combinations that you can remap. o F1 through F12 o Control-F1 through Control-F12 o Alt-F1 through Alt-F12 o Shift-F1 through Shift-F12 o Control-'unshifted_keypress' o Alt-'unshifted_keypress' o Shift-OS/2_function_key o Alt-OS/2_function_key o Ctrl-OS/2_function_key o Any OS/2_function_key For PMANT, these are the OS/2_function_keys that you can remap: Backspace Left-Control Delete Newline (Enter key on alpha keypad) Down Pagedown End Pageup Enter (Enter key on numeric keypad) Pause Escape Right F1-F12 Right-Alt Home Right-Control Insert Scroll-lock Left Up Left-Alt Tab -------------------------------------------------------------------- 19. TN5250 Emulator Changes In the TCP/IP 2.0 for OS/2 User's Guide, we list the OS/2 keys that you can remap to different functions. The syntax for remapping the keyboard is: os2_key whitespace 5250_function ;optional comment Following are changes to the names of the OS/2 keyboard keys that you can remap. The valid 'os2_key' names are: o F1 through F12 o Control-F1 through Control-F12 o Alt-F1 through Alt-F12 o Shift-F1 through Shift-F12 o Control-'unshifted_keypress' o Alt-'unshifted_keypress' o Control-os2_function_key o Alt-os2_function_key o Shift-os2_function_key o Any os2_function_key The valid 'os2_function_key' names are: BACKSPACE LEFT-CONTROL DELETE NEWLINE DOWN PAGEDOWN END PAGEUP ENTER (On numeric keypad) RIGHT ESCAPE RIGHT-ALT F1-F12 RIGHT-CONTROL HOME SCROLL-LOCK INSERT TAB LEFT UP LEFT-ALT The valid '5250_function' names are: ATTN FORWARD-WORD AUTOENTER HELP BACKSPACE HOME BACKTAB INSERT BACKWARD-KILL-WORD KILL-WORD BACKWARD-WORD LEFT CENT NEWLINE CLEAR PF1-PF24 DELETE PRINT DOWN RECBACK DUP RESET END RIGHT ENTER ROLLDOWN ERASE-EOF ROLLUP ERASE-INPUT SPBACKSPACE FIELD-MARK SYSREQ FIELDEXIT TAB FIELDMINUS TESTREQ FIELDPLUS UP Additional valid '5250_function' names (work only in documents and mail): TEXTDPBACKSPACE TEXTPAGEEND TEXTCARRIERRETURN TEXTDUP TEXTENDATTRIBUTE TEXTBOTTOMOFPAGE TEXTFASTLEFT TEXTBEGINOFLINE TEXTFASTRIGHT TEXTALTD TEXTINTENSIFY TEXTTOPOFPAGE TEXTNEWLINE TEXTFIELDBACKSPACE TEXTRQSPACE TEXTSTOPCODEFUNCTION TEXTRQTABFUNCTION TEXTCENT TEXTSYMBOLSCMDPROMPT TEXTSTOPCODEADVANCE TEXTUNDERSCORE -------------------------------------------------------------------- 20. The telnetd Command The TELNETD command now has a -c operand to specify a codepage. >>──telnetd──┬──────────┬──┬─────┬──┬─────┬──┬─────┬─────────────> └─ -p port─┘ └─ -b─┘ └─ -d─┘ └─ -l─┘ ┌─ -e 50──────────┐ ┌─ -u telnet_password─┐ >──┼─────────────────┼──┼─────────────────────┼──────────────────> └─ -e escapedelay─┘ └─ -u password────────┘ >──┬────────────────┬───────────────────────────────────────────>< └─ -c <codepage>─┘ -c <codepage> Specifies the ASCII codepage used by the remote host. If you do not specify a codepage, the data is sent untranslated. The possible values are: TCPDECMU for the DEC Multinational codepage TCP8859 for ISO 8859 codepage NONE for PC codepage 850 -------------------------------------------------------------------- 21. Packet Tracing Programs This kit includes documentation for a set of utilities that aid in diagnosing problems at your workstation. These files are installed in the <tcpip>\BIN directory. IPTRACE.EXE TR2SNIF.EXE IPPARSE.EXE IPFORMAT.EXE -------------------------- 21.1 IPTRACE.EXE IPTRACE will trace all packets received from and sent to the Ethernet or token-ring adapter. There are no parameters associated with IPTRACE. >>──iptrace─────────────────────────────────────────────────────>< The data is written to IPTRACE.DMP in the directory from which IPTRACE was started. To stop IPTRACE, press the Ctrl-c keys. Because IPTRACE will record all traffic sent and received, the IPTRACE.DMP file can grow quite large. There is no error checking to see if there is enough hard disk space available to continue recording the information. You should use IPTRACE in a limited fashion so that it does not impact the performance of the workstation where it is being run. IPTRACE is not a network monitor. It can trace only data received by and sent from the adapter. Also, IPTRACE does not provide a time stamp: it does not record when the packet was sent or received. After you start IPTRACE, it traces all adapters at the workstation. You can not select one adapter when several are installed. -------------------------- 21.2 TR2SNIF.EXE TR2SNIF converts the IPTRACE.DMP file to a format that can be viewed on a network monitor trace workstation (a network analyzer). This file will not contain any time information. Start TR2SNIF from the same directory where the IPTRACE.DMP file resides. You can code: >>──tr2snif──┬─e─┬──────────────────────────────────────────────>< └─t─┘ e Specifies Ethernet output. t Specifies token-ring output. -------------------------- 21.3 IPPARSE.EXE IPPARSE reads the IPTRACE.DMP file and produces a HEX dump for the screen. You can redirect this information to a file that you can browse with an editor. Start IPPARSE from the same directory where the IPTRACE.DMP file resides. You can enter IPPARSE with no parameter to dump the data to the screen, or you can enter IPPARSE <filename> to dump the data to the named file. >>──ipparse──┬───────────┬──────────────────────────────────────>< └─>filename─┘ >filename Specifies a file into which the data is dumped. -------------------------- 21.4 IPFORMAT.EXE IPFORMAT tries to format the IPTRACE.DMP and the SLIPTRC.DMP files into a user-readable format and display it to the screen. You can redirect the output to a file for browsing with a editor. If IPFORMAT cannot determine what type of packet was received, a HEX representation of the data is displayed. IPFORMAT will separate the IP, TCP, UDP, and ICMP layers. The rest of the packet will be displayed as HEX output. If IPFORMAT encounters a problem with the file, you may receive a General Protection Fault message. Customers should run IPPARSE and see if it is able to to parse the file completely or if it returns "error reading from iptrace.dmp". In either case, IPFORMAT will format the data until it reaches a problem. ┌─ -f IPTRACE.DMP─┐ >>──ipformat──┬─────┬──┬─────┬──┼─────────────────┼──┬─────┬─────> └─ -a─┘ └─ -d─┘ └─ -f filename────┘ └─ -h─┘ >──┬─────┬──┬───────────────┬──┬───────────┬── -? ──────────────>< └─ -n─┘ └─ -s hwaddress─┘ └─>filename─┘ -a Do not format ARP or RARP packets. -d Do not display the data portion of a packet. -f filename Specifies the input fine name. The default is IPTRACE.DMP. -h Display the raw data packet after the formatted information. -s hwaddress Format data only for the specified hardware address. Replace hwaddress with the 12-digit hexadecimal address for the Ethernet or token-ring adapter. You can use the NETSTAT -n command to display this address. -n Do not display hexadecimal data for unknown data type. > filename Redirect the output to the specified file. -? Display help information. -------------------------------------------------------------------- 22. Local Setup to Run TCPIPCFG from a LAN Here are some things you should do to set up your local system to run the TCPIPCFG program when TCP/IP resides on a LAN server. Make sure the following are setup on the local workstation: -------------------------- 22.1 For the Base Kit Set up: 1. A local <tcpip>/bin directory, and specify it in the PATH statement before the remote <tcpip>/bin directory. The local directory should contain: inet.sys, telnet.exe, and tcpipcfg.exe. 2. A local <tcpip>/dll directory, and specify it in the LIBPATH statement before you specify a remote <tcpip>/dll directory. The local directory should contain: cnbstk.dll and cnbapp.dll. -------------------------- 22.2 For the NFS Kit Set up: 1. A local <tcpip or nfskit>/bin directory, and specify it in the PATH statement before the remote <tcpip or nfskit>/bin directory. The local directory should contain nfsd.exe. 2. A local <tcpip or nfskit>/dll directory, and specify it before the remote <tcpip or nfskit>/dll directory in the LIBPATH statement. The local directory should contain cnbnfs.dll. -------------------------- 22.3 For the X Windows System Server kit Set up: 1. A local <tcpip or pmxkit>/bin directory, and specify it in the PATH statement before the remote <tcpip or pmxkit>/bin directory. The local directory should contain pmx.exe. 2. A local <tcpip or pmxkit>/dll directory, and specify it before the remote <tcpip or pmxkit>/dll directory in the LIBPATH statement. The local directory should contain cnbpmx.dll. -------------------------- 22.4 For the Extended Networking Kit Set up: 1. A local <tcpip or xntkit>/bin directory, and specify in the PATH statement before the remote <tcpip or xntkit>/bin directory. The local directory should contain x25io.exe. 2. A local <tcpip or xntkit>/dll directory, and specify it before the remote <tcpip or xntkit>/dll directory in the LIBPATH statement. The local directory should contain cnbx25.dll. -------------------------- 22.5 For the DOS/Windows Access kit Set up: 1. A local <tcpip>/bin directory, and specify it in the PATH statement before the remote <tcpip>/bin directory. The local directory should contain vdostcp.vdd. -------------------------------------------------------------------- 23. Network Security Program Support This kit includes a Telnet server that can use the authentication services of the Network Security Program (NetSP). You can get the NetSP by ordering program number 5622-061. The following information tells you how to setup Telnet using the authentication feature provided by the Network Security Program (NetSP), and includes: o An introduction to Telnet using the NetSP program o Installation and Configuration instructions o How to use Telnet with NetSP This file does not contain the information for setting up NetSP. For more information about how to set up and use NetSP, please refer to the Network Security Program for AIX, OS/2, and DOS: Developer's Guide (SC31-6500). -------------------------- 23.1 Introduction The Network Security Program (NetSP) provides a trusted third-party security program that authenticates the credentials of both a user at the Telnet client workstation and the server program at the Telnet server workstation. This trusted third party is called the Authentication Server, and provides the means by which the user's program and application program can identify and validate each other. The Telnet server and client identify the Telnet user at the client workstation and at the Telnet server workstation using the Authentication Server provided by NetSP. This third party authentication process is efficient because it reduces the number of administrative tasks when you want to add, change, or delete users and their access to programs. After you define a user to the Authentication Server, and after that user gets security credentials from the Authentication Server, that user can be authenticated to use any application program known to that Authentication Server. In effect, it takes only one administrative action to add a new client to all application servers in the network. Telnet, using the authentication feature provided by NetSP, no longer requires the user to enter the Telnet password of the target host for authorized Telnet connection. This yields Telnet connections that are more convenient and more secure. -------------------------- 23.2 Installation and Configuration In order to use the Telnet with NetSP, you must install both TCP/IP and NetSP on the Telnet server and client workstations, and configure those workstations to use the authentication feature. The following is a brief overview of the installation and configuration processes for Telnet with NetSP: 1. Use TCPINST to install TCP/IP V2.0 for OS/2 on the Telnet server and client workstations. 2. Follow NetSP installation process to install NetSP on your Telnet server and client workstations. 3. Have your Authentication Server administrator add the Telnet server name to the Authentication Server. The Telnet server name should contain the character string "TELNET." followed by the complete host name of the Telnet server workstation. For example, if the host name is test.raleigh.ibm.com, the Telnet server name would be TELNET.test.raleigh.ibm.com. 4. Have your Authentication Server administrator add the Telnet user's name to the Authentication Server. 5. Ensure that you have the <tcpip>\bin subdirectory on your PATH statement, and that the KK.DLL and KKTelnet.DLL files are in a directory that is in your LIBPATH statement. 6. On the Telnet server workstation, use the SETTCPPF command to set the Telnet authentication option in the TCP/IP profile (TCPIP.INI) to optional (1) or required (2). The SETTCPPF command has the following format: >>──settcppf──┬─ -Telnet ─┬─0─┬──┬──────────────────────────────>< │ ├─1─┤ │ │ └─2─┘ │ ├─ -display────────┤ └─ ?───────────────┘ -Telnet <value> Specifies 0 Sets authentication off. 1 Sets authentication to optional. The server will attempt to authenticate the user, but will complete the login even if the authentication fails. 2 Sets authentication to mandatory. The server will attempt to authenticate the user, and will complete the login only if the authentication is successful. -display Displays the authentication status at the workstation. ? Displays the syntax of the settcppf command. -------------------------- 23.3 Authenticating a Telnet Connection Before using Telnet with NetSP, create a master-key cache file on both Telnet server and client workstations. To create master-key cache file on the Telnet server workstation, use KKINSTK (a NetSP function) to enter a master key into the cache file. You should use the KKINSTK command with -s option to initialize a cache file on the Telnet server workstation. (For information about KKINSTK, see the Network Security Program for AIX, OS/2, and DOS: Developer's Guide.) To create master-key cache file on the Telnet client workstation, use KKLOGIN (another NetSP function) or KKINSTK to enter a master key into the cache file. (For information about KKINSTK and KKLOGIN, see the Network Security Program for AIX, OS/2, and DOS: Developer's Guide.) After you create master-key cache files on both Telnet client and server workstations, you can issue the Telnet command to start Telnet. telnet <targetName> If the Telnet authentication option is set to optional (1) or required (2) on the Telnet server workstation, an authentication message is displayed on the Telnet client when authentication is started. When authentication process ends, an authentication success or failed message is displayed on the Telnet client. If authentication fails, and the Telnet authentication option on the server workstation is set to optional (1), the normal login process is performed as usual. ************************************* ***************** Fixed APARs ******************* ************************************* 1. APARs Fixed The following is a list of APARs fixed since this release became available. APARs are grouped numerically by CSD. -------------------------- 1.1 APARS Fixed in CSD UN56401 +==========+===============+=======================================+ | | MODULES | | | APAR | AFFECTED | DESCRIPTION | +==========+===============+=======================================+ | PN46622 | TCPINST.EXE | Fixed: When performing a CID Attended | | | TCPINST2.EXE | Install, the INSTALL program did not | | | | select the adapter or the protocol. | +----------+---------------+---------------------------------------+ | PN47008 | SENDMAIL.EXE | Fixed: Multiple Sendmail causing | | | | system to hang. | +----------+---------------+---------------------------------------+ | PN47308 | TELNETPM.EXE | Fixed: When opening multiple telnetpm | | | | windows, cursor is not displayed on | | | | the first telnetpm window. | +----------+---------------+---------------------------------------+ | PN47373 | TELNETPM.EXE | Fixed: When using TelentPM to log | | | | onto an OS/2 machine using VT220 mode,| | | | if a command is entered that will | | | | scroll the screen, TelnetPM either | | | | forces the user off or stops | | | | displaying the results on the screen. | +----------+---------------+---------------------------------------+ | PN48022 | PMANT.EXE | Fixed: PMANT and TN3270 to a COMTEN | | | TN3270.EXE | that is connected to MVS doesn't | | | | display logon screen on the console. | +----------+---------------+---------------------------------------+ | PN48051 | SNMPD.EXE | Fixed: SYSUPTIME is set to zero in | | | | SNMPTRAP. | +----------+---------------+---------------------------------------+ | PN48241 | SNMPTRAP.EXE | Fixed: SNMP TRAP addressing exception | | | SNMPTRAP.HLP | occurs when using enterprise specific | | | SNMP.EXE | trap. | +----------+---------------+---------------------------------------+ | PN48275 | RSHD.EXE | Fixed: RSHD does not accept commands; | | | | "Unauthorized Request Rejected" error | | | | message displays. | +----------+---------------+---------------------------------------+ | PN48444 | TN3270.EXE | Fixed: Issuing TN3270 using syntax | | | | shown in the Command Reference with | | | | the -ext parameter before the port | | | | parameter results in message that the | | | | remote host name is missing. | +----------+---------------+---------------------------------------+ | PN48454 | NR2.EXE | Fixed: When NR/2 is accessing the News| | | | Server, it will stop at a News Group | | | | and issue the message, "The News | | | | Server does not handle XHDR." NR/2 | | | | stops processing. | +----------+---------------+---------------------------------------+ | PN48509 | NR2.EXE | Fixed: While running NR/2, and | | | NR2.HLP | attempting to add multiple News | | | | Groups, the function fails to | | | | complete. | +----------+---------------+---------------------------------------+ | PN48570 | TELNET.EXE | Fixed: PRINT THRU does not work | | | TELNETPM.EXE | properly with TELNET (VT100/VT220) | +----------+---------------+---------------------------------------+ | PN48742 | TELNET.EXE | Fixed: When using VT220 to DEC | | | TELNETPM.EXE | machine, all tab fields in which | | | | the cursor resides are corrupted. | +----------+---------------+---------------------------------------+ | PN48816 | TELNETPM.EXE | Fixed: The cursor in telnetpm is hard | | | | to see when it reaches the bottom of | | | | the window. Need block cursor. | +----------+---------------+---------------------------------------+ | PN48902 | SENDMAIL.EXE | Fixed: SENDMAIL receives stack over- | | | | flow msg and stops. | +----------+---------------+---------------------------------------+ | PN49057 | TELNETPM.EXE | Fixed: TELNETPM V2.0 not processing | | | TELNET.EXE | screen size parameters properly | +----------+---------------+---------------------------------------+ | PN49117 | FTPPM.EXE | Fixed: FTPPM to an MVS machine results| | | | in truncated host dataset names. Any | | | | attempt to put/get local/remote files | | | | fails. | +----------+---------------+---------------------------------------+ | PN49266 | SENDMAIL.EXE | Fixed: SENDMAIL 1.2.12 makes only one | | | | attempt to deliver mail to hosts that | | | | are inactive. | +----------+---------------+---------------------------------------+ | PN49360 | PMANT.EXE | Fixed: Not all applications that | | | TN3270.EXE | contain APL box characters work | | | 3278XLT.TBL | properly. | +----------+---------------+---------------------------------------+ | PN49775 | CNBNFS.DLL | Fixed: When using the Configuration | | | | Notebook to set up the exports, unable| | | | to make a directory R/O if no hostname| | | | is specified. | +----------+---------------+---------------------------------------+ | PN50202 | SNMPD.EXE | Fixed: SNMPD loses requests when | | | | connection is still open. | +----------+---------------+---------------------------------------+ | PN50269 | FTPPM.EXE | Fixed: FTPPM, when left idle, will | | | FTPAPI.DLL | restore files to home directory. | +----------+---------------+---------------------------------------+ | PN50287 | FTP.EXE | Fixed: 'Codepage none' is not the | | | | default when starting the FTP client. | +----------+---------------+---------------------------------------+ | PN50505 | TELNET.EXE | Fixed: TELNET VT220 connection to | | | TELNETPM.EXE | Silicon Graphics machine hangs. | +----------+---------------+---------------------------------------+ | PN50783 | SENDMAIL.EXE | Fixed: SENDMAIL for TCP/IP 2.0 cannot | | | | handle MAILGATE.COMPUSERVE.COM | +----------+---------------+---------------------------------------+ | PN50873 | WPTELNET.DLL | Fixed: Telnet template doesn't save | | | | port number properly. | +----------+---------------+---------------------------------------+ | PN50874 | SLIP.EXE | Fixed: When NFS is running over SLIP | | | SLCFG.EXE | to an OS/2 NFS server, if a DIR | | | SLIPHOLD.EXE | command results in more than 74 files,| | | SLIPTERM.EXE | NFS hangs. | | | SLIPWAIT.EXE | | | | ATANS.CMD | | | | ATDIAL.CMD | | +----------+---------------+---------------------------------------+ | PN51021 | FTPWPS.DLL | Fixed: FTPPM doesn't save user ID and | | | | password settings properly. | +----------+---------------+---------------------------------------+ | PN51031 | FTPPM.EXE | Fixed: Drag-and-drop to FTPPM results | | | FTPPM.HLP | in local directory name being lost. | | | FTPAPI.DLL | | +----------+---------------+---------------------------------------+ | PN51032 | FTPPM.EXE | Fixed: Trailing colon on remote | | | | directory names (for Sun and Unix | | | | workstations) prevents doubleclicking | | | | on a directory name to change to it. | +----------+---------------+---------------------------------------+ | PN51218 | TELNET.EXE | Fixed: Telnet displays File Not Found | | | TELNETPM.EXE | message when scrolling up or down in | | | | UniSPF editor window. | +----------+---------------+---------------------------------------+ | PN51436 | SO32DLL.DLL | Fixed: Select() call fails when more | | | | than 256 connections are attempted. | +----------+---------------+---------------------------------------+ | PN51569 | RSH.EXE | Fixed: RSH doesn't accept commands of | | | REXEC.EXE | the form: /USR/BIN/LS | +----------+---------------+---------------------------------------+ | PN51617 | TCPINST.EXE | Fixed: Installing TCPIP from a | | | TCPINST2.EXE | diskette drive other than drive A:, | | | | TCPINST continues to ask for | | | | information from drive A: and not B:. | +----------+---------------+---------------------------------------+ | PN51626 | PMANT.EXE | Fixed: TN3270 PMANT reset modified | | | TN3270.EXE | data tag fields using -ext option not | | | | being returned properly; protected | | | | versus unprotected field attributes | | | | not being saved properly. | +----------+---------------+---------------------------------------+ | PN51641 | IFNDIS.SYS | Fixed: ARP -s does not work. | +----------+---------------+---------------------------------------+ | PN51643 | BOX.EX | Fixed: LaMail doesn't place notes in | | | DRAW.EX | correct folder. | | | E3EMUL.EX | | | | EPM.EX | | | | EPMLEX.EX | | | | EPMLIST.EX | | | | EXTRA.EX | | | | FIC.DLL | | | | GET.EX | | | | HELP.EX | | | | LAM55LE.DLL | | | | LAM55LR.DLL | | | | LAMAIL.EXE | | | | LAMAIL.HLP | | | | LAMAILER.DLL | | | | LAMEXTRA.EX | | | | LAMFILES.OUT | | | | LAMRES.DLL | | | | MAILLIST.EX | | | | MATHLIB.EX | | | | PUT.EX | | +----------+---------------+---------------------------------------+ | PN51923 | WPTELNET.DLL | Fixed: TelnetPM settings notebook does| | | | not properly process the Logfile and | | | | Tracefile fields. | +----------+---------------+---------------------------------------+ | PN52008 | RSH.EXE | Fixed: RSH and REXEC are only able to | | | REXEC.EXE | transmit first part of large files. | +----------+---------------+---------------------------------------+ | PN52479 | TELNETPM.EXE | Fixed: TELNETPM copy to clipboard | | | | replaces some lines with CRLF. | +----------+---------------+---------------------------------------+ | PN52743 | TELNET.EXE | Fixed: VI editor problems occur when | | | TELNETPM.EXE | using VT100 emulator. | +----------+---------------+---------------------------------------+ | PN52769 | CNBSTK.DLL | Fixed: When using the TCP/IP | | | CNBAPP.DLL | Configuration utility to edit the | | | | sendmail.cf file, the Mlocal | | | | definition is deleted or commented | | | | out. | +----------+---------------+---------------------------------------+ | PN52828 | TELNET.EXE | Fixed: Unable to emulate VT100 in | | | TELNETPM.EXE | Telnet. | +----------+---------------+---------------------------------------+ | PN52910 | CNBAPP.DLL | Fixed: TCPIPCFG doesn't build the | | | | RHOSTS file properly. | +----------+---------------+---------------------------------------+ | PN53112 | FTP.EXE | Fixed: When using the FTP client to go| | | | to an MVS or VM machine, some ASCII | | | | characters are translated to EBCDIC | | | | incorrectly. | +----------+---------------+---------------------------------------+ | PN53370 | IFCONFIG.EXE | Fixed: Unable to enter all 1's as a | | | | broadcast address. | +----------+---------------+---------------------------------------+ | PN53515 | LAMAIL.EXE | Fixed: First note sent after startup | | | LAM551R.DLL | is lost. | +----------+---------------+---------------------------------------+ | PN53518 | LAMAIL.EXE | Fixed: LaMail Nickname file is lost | | | | from notes options panel. | +----------+---------------+---------------------------------------+ | PN53573 | RPCGEN.EXE | Fixed: RPCGEN.EXE doesn't produce the | | | | proper TYPEDEF statement. | +----------+---------------+---------------------------------------+ | PN53626 | FTPD.EXE | Fixed: Using DOS FTP or WFTP to OS/2 | | | FTPDC.EXE | produces: 550 GETCWD ERROR: ERROR 13 | | | FTPDTRC.EXE | | +----------+---------------+---------------------------------------+ | PN53938 | TELNET.EXE | Fixed: Lines do not wrap properly in | | | | an ansi telnet VI session. | +----------+---------------+---------------------------------------+ | PN54025 | FTPPM.EXE | Fixed: FTPPM settings does not allow | | | FTPAPI.DLL | creation of a new window. | | | FTPWPS.DLL | | +----------+---------------+---------------------------------------+ | PN54041 | SNMPD.EXE | Fixed: SNMPD.EXE crashes if an OID | | | | has more than 50 sub-IDs | +----------+---------------+---------------------------------------+ | PN54640 | INET.SYS | Fixed: TCPIP stops after running | | | IFNDIS.SYS | customer application on offload. | | | IFNDISNL.SYS | | +----------+---------------+---------------------------------------+ | PN54999 | FTPWPS.DLL | Fixed: FTPPM icon disappears from | | | | TCP/IP and templates folders. | +----------+---------------+---------------------------------------+ | PN55603 | TCPIPCFG.EXE | Fixed: When SLIP changes for | | | CNBSTK.DLL | SETUP.CMD are saved the lan adapter | | | CNBAPP.DLL | settings get lost. | +----------+---------------+---------------------------------------+ | PN55942 | INET.SYS | Fixed: A 3172 Mod 3 running OS/2 | | | IFNDIS.SYS | TCPIP Version 2.0 and Offload code | | | | received a TRAP 000D. | +----------+---------------+---------------------------------------+ | PN56391 | FTPPM.EXE | Fixed: FTPPM does not send a proper | | | FTPAPI.DLL | character for a space when it is part | | | | of the password. | +----------+---------------+---------------------------------------+ -------------------------- 1.2 The following APARs were fixed in CSD UN50382. These changes are included in this CSD as well. +==========+===============+=======================================+ | | MODULES | | | APAR | AFFECTED | DESCRIPTION | +==========+===============+=======================================+ | PN44375 | INET.SYS | Code enhancement for Broadcast Mode. | | | IFNDIS.SYS | | | | BOOTPD.EXE | | +----------+---------------+---------------------------------------+ | PN45987 | TN3270.EXE | Fixed: TN3270 resizes the OS/2 window | | | | when finished. | +----------+---------------+---------------------------------------+ | PN46145 | TN3270.EXE | Fixed: TN3270 and PMANT blank fill | | | PMANT.EXE | invisible input fields. | +----------+---------------+---------------------------------------+ | PN46718 | TCPINST.EXE | Fixed: TCP/IP for OS/2 is not CID | | | | enabled. | +----------+---------------+---------------------------------------+ | PN47132 | TN5250.EXE | Fixed: Unable to remap some TN5250 | | | | functions to different keys. | +----------+---------------+---------------------------------------+ | PN47313 | TELNETD.EXE | Fixed: TELNET to OS/2 cannot Ctl-c | | | TELNETDC.EXE | out. | +----------+---------------+---------------------------------------+ | PN47508 | TELNET.EXE | Fixed: VT220 keyboard mapping of | | | | SETTERM facility produces X'0D'. | +----------+---------------+---------------------------------------+ | PN47666 | CNBAPP.DLL | Fixed: TCPIPCFG does not read SLIP | | | | script ATTACH parms with comma. | +----------+---------------+---------------------------------------+ | PN47668 | CNBAPP.DLL | Fixed: TCPIPCFG SLIP doesn't check | | | | ATTACH parms for double quotes. | +----------+---------------+---------------------------------------+ | PN47906 | TELNET.EXE | Fixed: Unable to Telnet to HP | | | TELNETPM.EXE | machine. | +----------+---------------+---------------------------------------+ | PN49178 | WPTELNET.EXE | Fixed: Terminal type on TEMPLATE is | | | | not being passed. | +----------+---------------+---------------------------------------+ | PN49356 | RSH.EXE | Fixed: RSH causes CPU utilization to | | | | go to 100%. | +----------+---------------+---------------------------------------+