TCP/IP Configuration and Management Manual Abstract This manual describes how to set up and manage the HP NonStop™ Transmission Control Protocol/Internet Protocol (TCP/IP) subsystem on an HP NonStop S-series server and on an HP Integrity NonStop NS-series server. This manual is written for system managers, operators, and others who configure and manage NonStop TCP/IP. Product Version NonStop TCP/IP G06 NonStop TCP/IP H01 Supported Release Version Updates (RVUs) This manual supports G06.
Document History Part Number Product Version Published 137045 Tandem NonStop TCP/IP G05 February 1998 140771 Tandem NonStop TCP/IP G06 May 1998 427132-001 NonStop TCP/IP G06 August 2002 427132-002 NonStop TCP/IP G06 August 2004 427132-003 NonStop TCP/IP G06 NonStop TCP/IP H01 July 2005 427132-004 NonStop TCP/IP G06 NonStop TCP/IP H01 March 2014 New editions incorporate any updates issued since the previous edition.
Legal Notices Copyright 1998, 2014 Hewlett-Packard Development Company, L.P. Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor's standard commercial license. The information contained herein is subject to change without notice.
TCP/IP Configuration and Management Manual Glossary Index Examples Figures Legal Notices What’s New in This Manual xi Manual Information xi New and Changed Information xi About This Manual xiii Who Should Use This Manual xiii How This Manual Is Organized xvii Notation Conventions xviii 1.
2.
3. Configuring the NonStop TCP/IP Subsystem Contents ATM Term Definitions 2-7 RFC Compliance 2-7 Hardware and Software Requirements 2-8 Viewing ARP and ATMARP Tables 2-8 3.
4.
4.
A. Configuration Reference Contents Socket Command Records 4-139 UDP User Request Records 4-143 A. Configuration Reference Other NonStop TCP/IP Services A-2 Domain Name Server (DNS) A-2 Domain Name Resolver A-5 Domain Name Server Files A-5 B.
C. Well-Known Port Numbers for TCP and UDP Contents C. Well-Known Port Numbers for TCP and UDP D. SCF Command Summary E. SCF Error Messages F. NonStop Systems Used as Internet Gateways Startup Files for HOST1 F-5 Startup Files for GTWY1 F-10 Glossary Index Examples Example 1-1. Example 1-2. Example 1-3. Example 1-4. Example 1-5. Example 1-6. Example 1-7. Example 3-1. Example 3-2. Example 3-3. Example 3-4. Example 3-5. Example 3-6. Example 3-7. Example 3-8. Example 3-9. Example 3-10. Example 3-11.
Figures Contents Example 3-19. Example 3-20. Example 3-21. Example 3-22. Example B-1. Example F-1. Example F-2. Example F-3. Example F-4. Example F-5. Example F-6. SERVICES File 3-40 PORTCONF File 3-40 Setting the OSS DEFINE to Specify the TCP/IP Process Preventing Port Collisions by Modifying inet.
Tables Contents Figure F-1. Figure F-2. Figure F-3. Figure F-4. Figure F-5. Figure F-6. Hosts and a Gateway in an Internet F-1 Subnets F-2 NonStop Systems in an Internet F-3 A NonStop System as a Gateway Between Subnets F-3 A NonStop System as Two Logical Hosts Connected to a Subnet Three Gateways in the NonStop TCP/IP Environment F-5 Tables Table 4-1. Table 4-2. Table 4-3. Table 4-4. Table 4-5. Table 4-6. Table 4-7. Table B-1. Table C-1. Table C-2. Table C-3.
Contents TCP/IP Configuration and Management Manual—427132-004 x
What’s New in This Manual Manual Information TCP/IP Configuration and Management Manual Abstract This manual describes how to set up and manage the HP NonStop™ Transmission Control Protocol/Internet Protocol (TCP/IP) subsystem on an HP NonStop S-series server and on an HP Integrity NonStop NS-series server. This manual is written for system managers, operators, and others who configure and manage NonStop TCP/IP.
What’s New in This Manual New and Changed Information Updated TCP Layer Stats with new attributes in the section Description of Statistics for the TCP Layer on page 4-69. The earlier version of the TCP/IP Configuration and Management Manual contained the following changes: Manuals for that support the Integrity NonStop server have been added to About This Manual. Wherever SUBNET refers to the NonStop TCP/IP subsystem object, it is capitalized.
About This Manual This manual describes G-series and H-series Release Version Updates (RVUs) of HP NonStop TCP/IP. Who Should Use This Manual This manual describes the installation, configuration, and management of the NonStop TCP/IP subsystem. It is for system managers, operators, and others who configure and manage the NonStop TCP/IP subsystem.
Background Manuals About This Manual RFC 819 “Domain Naming Convention for Internet User Applications” RFC 821 “Simple Mail Transfer Protocol” RFC 826 “Ethernet Address Resolution Protocol” RFC 894 “Standard for the Transmission of IP Datagrams Over Ethernet Networks” RFC 973 (1034:1982, 1876, 1101; 1035: 1348, 1995, 1996 RFC 974 “Mail Routing and Domain System” RFC 1034 (Now 1101, 1982 and 1876) RFC 1042 “Standard for the Transmission of IP Datagrams Over IEEE 802 Networks” RFC 1495 “M
About This Manual Integrity NonStop Server Configuration Manuals The NonStop S-Series Planning and Configuration Guide describes how to plan and configure a NonStop S-series server and provides a guide to other manuals for the NonStop S-series server. The SCF Reference Manual for G-Series RVUs describes the operation of SCF and the commands used to configure, control, and inquire about supported data communications subsystems.
X25AM Manuals About This Manual The TCP/IP TELNET Management Programming Manual describes the command/response interface and the EMS interface available to an application program for communication with the TCP/IP TELNET process. The QIO Configuration and Management Manual describes how to install and manage a QIO data communications subsystem.
How This Manual Is Organized About This Manual How This Manual Is Organized The following table summarizes the contents of this manual: Section and Title Description Section 1, Configuration Quick Start This section provides concise examples of setting up the HP NonStop TCP/IP environment.
Notation Conventions About This Manual Section and Title Description Appendix C, Well-Known Port Numbers for TCP and UDP This appendix lists the port numbers preassigned to specific services when accessed from TCP or UDP. Appendix D, SCF Command Summary This appendix summarizes the SCF command syntax. Appendix E, SCF Error Messages This appendix contains a description of the TCP/IP subsystem SCF error messages.
General Syntax Notation About This Manual italic computer type. Italic computer type letters within text indicate C and Open System Services (OSS) variable items that you supply. Items not enclosed in brackets are required. For example: pathname [ ] Brackets. Brackets enclose optional syntax items. For example: TERM [\system-name.]$terminal-name INT[ERRUPTS] A group of items enclosed in brackets is a list from which you can choose one item or none.
Notation for Messages About This Manual Quotation marks around a symbol such as a bracket or brace indicate the symbol is a required character that you must enter as shown. For example: "[" repetition-constant-list "]" Item Spacing. Spaces shown between items are required unless one of the items is a punctuation symbol such as a parenthesis or a comma. For example: CALL STEPMOM ( process-id ) ; If there is no space between two items, spaces are not permitted.
Notation for Messages About This Manual Bold Text. Bold text in an example indicates user input entered at the terminal. For example: ENTER RUN CODE ?123 CODE RECEIVED: 123.00 The user must press the Return key after typing the input. Nonitalic text. Nonitalic letters, numbers, and punctuation indicate text that is displayed or returned exactly as shown. For example: Backup Up. lowercase italic letters. Lowercase italic letters indicate variable items whose values are displayed or returned.
Notation for Management Programming Interfaces About This Manual % Percent Sign. A percent sign precedes a number that is not in decimal notation. The % notation precedes an octal number. The %B notation precedes a binary number. The %H notation precedes a hexadecimal number.
Change Bar Notation About This Manual Change Bar Notation Change bars are used to indicate substantive differences between this manual and its preceding version. Change bars are vertical rules placed in the right margin of changed portions of text, figures, tables, examples, and so on. Change bars highlight new or revised information. For example: The message types specified in the REPORT clause are different in the COBOL85 environment and the Common Run-Time Environment (CRE).
About This Manual HP Encourages Your Comments TCP/IP Configuration and Management Manual—427132-004 xxiv
1 Configuration Quick Start This section provides concise examples of setting up the HP NonStop TCP/IP environment. The services are not described in detail in this section; for more detailed information, see Networking Services Provided in the Guardian Environment on page 3-28, and Appendix B, NonStop TCP/IP Processes and Protocols. For more complex configuration examples, see Section 3, Configuring the NonStop TCP/IP Subsystem.
Configuration Quick Start Name of a SLSA LIF Name of a SLSA LIF Follow the instructions below to obtain the name of a SLSA LIF. 1. Obtain a list of all LIFs by entering the following command at the SCF prompt: STATUS LIF $ZZLAN.* 2. Using one of the LIF names displayed, determine if the LIF is of TYPE ETHERNET by entering toe following command: INFO LIF $ZZLAN.lifname 3.
Configuration Quick Start Name of Second Host Name of Second Host You’ll need this name when adding a second TCP/IP host to your system. This name is arbitrary, you assign it. Example 1-4 uses the host name gust. This is item 8 in the configuration form. Name of Second SLSA LIF Follow the instructions for choosing a SLSA LIF under the subheading “Name of SLSA LIF,” above, and select a second SLSA LIF. Example 1-4 uses a SLSA LIF name LAN02. This is item 9 in the configuration form.
Configuration Quick Start IP Address of the SUBNET IP Address of the SUBNET Get the SUBNET IP address from the network administrator. Example 1-6 uses the address 172.16.192.203. This is item 13 in the configuration form. DEVICENAME for the Adapter To choose a DEVICENAME for an adapter, issue the following command at the SCF prompt: NAMES ADAPTER $ZZATM.* This command results in a listing of the operational ATM adapters on the system. Choose an adapter and use its name as the DEVICENAME for the adapter.
ATM Address of the Entry Configuration Quick Start ATM Address of the Entry Get entry ATM address from the network administrator. Example 1-6 uses the address %H47000580FFE1000000F21A29EB000000000A100. Figure 1-1. TCP/IP Configuration Form TCP/IP Configuration Form Date: __________________ 1. Host name ____________________________________________ 2. IP address of host being set up ______________________ 3. Name of SLSA LIF _____________________________________ 4.
Configuration Quick Start Assumptions Task 2: Create a command file Task 3: Start the NonStop TCP/IP Environment Assumptions The SCF environment is operational (that is, SCP is running) QIOMON is running The SLSA subsystem is configured (see the LAN Configuration and Management Manual to configure the SLSA subsystem) You need to substitute real values for the following values; (these variables are indicated by italics in the example): IP Address of Host being set up (the example uses
Tasks Configuration Quick Start Tasks Hint. Before you add your NonStop TCP/IP process pair, issue an SCF STATUS LIF $ZZLAN.* command. Select a running LIF and issue a STATUS LIF $ZZLAN.lif-name on that LIF. The “CPUs with Data Path” field lists the available CPUs. Select two CPUs in the “CPUs with Data Path” list. Use the CPUs you selected when specifying the primary and backup CPU numbers for the TCP/IP process. In addition, use the same CPUs when adding TELNET and LISTNER.
Configuration Quick Start Task Summary: Shutting Down TCP/IP Example 1-1. TCPIPUP Command File TCPIP/NAME $ZTC0,TERM $TRM0.#A,OUT $TRM0.#A,CPU 0,NOWAIT/1 SCF/INLINE/ INLPREFIX + + ASSUME PROCESS $ZTC0 + ALTER,HOSTNAME "breeze" ==breeze is the Host Name + ALTER,HOSTID 150.20.30.1 ==150.20.30.1 is the IP Address of Host + ALTER SUBNET #LOOP0,ADDRESS 127.1 + ADD SUBNET #SN0,TYPE ETHERNET,DEVICENAME LAN21, & IPADDRESS 150.20.30.
Configuration Quick Start Tasks Tasks 1. Create a TACL command file as shown in Example 1-2. Example 1-2. TCPIPDN Command File SCF/INLINE/ INLPREFIX + + ABORT WINDOW $ZTN0.* INLEOF STOP $ZTN0 STOP $LSN0 SCF ABORT PROCESS $ZTC0 2. Issue the OBEY command on the TCPIPDN command file while running as user SUPER.SUPER.
Configuration Quick Start Task Summary: Adding a Second TCP/IP Host Using a Second TCP/IP Process 3. Change the TCPIPUP command file to include the correct hostname for this system (including the domain suffix). Example 1-3 shows the new TCPIPUP command file: Example 1-3. TCPIPUP Command File TCPIP/NAME $ZTC0,TERM $TRM0.#A,OUT $TRM0.#A,CPU 0,NOWAIT/1 SCF/INLINE/ INLPREFIX + + ASSUME PROCESS $ZTC0 + ALTER,HOSTNAME "breeze.mydomain.com" ==breeze is the Host Name + ALTER,HOSTID 150.20.30.1 ==150.20.30.
Configuration Quick Start Tasks Name of Second SLSA LIF (the example uses LAN02). See Hint on page 1-7 for choosing an appropriate LIF. Tasks 1. Create a new TACL command file called TCPIPUP1 to start the second NonStop TCP/IP environment using the contents of Example 1-4. The define for TCPIP^PROCESS^NAME is needed when starting applications which don't use the default NonStop TCP/IP process. The default NonStop TCP/IP process name is $ZTC0, so you do not need to use the define when using $ZTC0.
Configuration Quick Start Assumptions Task 2: Issue an OBEY command on the file Assumptions A second TCP/IP process is running on the system. You need to substitute real values for the opener process names. (List all openers of the TCP/IP home terminal process provider.) Tasks 1. Create a TACL command file with the contents shown in Example 1-5. Example 1-5. TCPIPDN1 Command File SCF/INLINE/ INLPREFIX + + ABORT WINDOW $ZTN1.
Configuration Quick Start Assumptions Process name (the example uses $ZTC1) Hint. Before you add your NonStop TCP/IP process pair, issue an SCF NAMES ADAPTER ZZATM.* command. Select a running ATM adapter and issue an INFO ADAPTER $adaptername. The “Access List” field lists the available CPUs. Select two CPUs from the “CPUs with Data Path” list. Use the CPUs you selected when specifying the primary and backup CPU numbers for the TCP/IP process.
Configuration Quick Start Tasks Tasks 1. Create a TACL command file with the contents shown in Example 1-6. Example 1-6. ATMUP Command File TCPIP/NAME $ZTC1,TERM $TRM0.#A,OUT $TRM0.#A,CPU 1,NOWAIT/2 SCF/INLINE/ INLPREFIX+ +ALLOW 100 ERRORS +ADD SUBNET #EN1,TYPE ATM,IPADDRESS 172.16.192.203, & DEVICENAME $AM2 ==#EN1 is the SUBNET Name ==172.16.192.
Configuration Quick Start Assumptions Assumptions The SCF environment is operational (that is, SCP is running) QIOMON is running The ATM subsystem is configured (see the ATM Configuration and Management Manual to configure the ATM subsystem) You must have the correct ATM address (see your network manager) You need to substitute real values for the following variables (indicated by italics in the example): TCP/IP Process Name (the example uses $ZTC1).
Configuration Quick Start Tasks Example 1-7. ATMUP2 Command File TCPIP/NAME $ZTC1,TERM $TRM0.#A,OUT $TRM0.#A,CPU 1,NOWAIT/2 ==$ZTC1 is the TCP/IP Process Name SCF/INLINE/ INLPREFIX + +ALLOW 100 ERRORS +ADD SUBNET #EN1,TYPE ATM,IPADDRESS 172.16.192.203, & DEVICENAME $AM2, ARPSERVER ON ==#EN1 is the SUBNET Name ==172.16.192.
2 Overview of NonStop TCP/IP This section describes the NonStop TCP/IP subsystem in relation to other NonStop data communications subsystems. Note. Only Ethernet communications through the G4SA are available on the Integrity NonStop server; to use other protocols and adapters, the Integrity NonStop server must be connected to an IOMF2 customer replaceable unit (CRU) in a NonStop S-series I/O enclosure. For more information, see the Introduction to Networking for HP Integrity NS-Series Servers.
Overview of NonStop TCP/IP Figure 2-1. NonStop TCP/IP Subsystem Within the NonStop System SCF Commands and Responses DSM Management Applications WAN SCF SPI-formatted messages X25AM SCP QIO Shared Memory Segment TCP/IP ATM SLSA LAN Drivers/Interrupt Handlers ATM Drivers/Interrupt Handlers ServerNet Fabrics ServerNet Fabrics GESA Adapter G4SA Adapter E4SA Adapter TRSA Adapter FESA Adapter ATM3SA Adapter LAN X.
Overview of NonStop TCP/IP Management Interfaces and Network Components Shown in Figure 2-1 Management Interfaces and Network Components Shown in Figure 2-1 The following subsections describe the management interfaces and network components shown in Figure 2-1. The NonStop TCP/IP product for NonStop servers provides transparent connections to IEEE 802.3 Ethernet LANs and IEEE 802.5 token-ring LANs through the SLSA subsystem. NonStop TCP/IP uses the WAN subsystem to interface to X.
Overview of NonStop TCP/IP GESA GESA A single-port ServerNet adapter that provides 1000 Mbps data transfer rates between NonStop S-series systems and Ethernet LANs. SCF and SCP SCF is an interactive interface that allows operators and system managers to configure, control, and monitor the NonStop TCP/IP subsystem. SCF is part of DSM. The Subsystem Control Point (SCP) provides an interface to the I/O processes of the various subsystems.
NonStop TCP/IP Interface to LAN, WAN, and ATM networks Overview of NonStop TCP/IP through the rest of the SLSA subsystem, which consists of ServerNet Addressable Adapters (SACs) and PIFs, then out to a LAN, WAN, or X.25 connection. A brief description of the SLSA components relevant to this discussion follows Figure 2-2. The NonStop TCP/IP process interfaces to ATM adapters (ATM3SAs) through the ATM subsystem which is not shown in Figure 2-2 (see Figure 2-1). Figure 2-2.
Overview of NonStop TCP/IP Management Interfaces and Network Components Shown in Figure 2-2 Management Interfaces and Network Components Shown in Figure 2-2 Filters Filters provide a logical mechanism whereby frames received from the LAN can be sorted, then delivered, to a particular client such as NonStop TCP/IP. Filters replace the PORT objects used in systems prior to the ServerNet architecture in the sense that filters are the final destination for data received from the LAN.
Overview of NonStop TCP/IP ATM Term Definitions ATM Term Definitions The following terms are used throughout this manual in reference to ATM. ATMARP Table The TCP/IP process maintains a table of ATM IP addresses in an ATMARP table. Similar to the ARP table TCP/IP uses to map an IP address to a MAC address in Ethernet and token-ring networks, TCP/IP uses the ATMARP table to map IP addresses to an ATM address in ATM networks. You can view and change the ATMARP table.
Overview of NonStop TCP/IP Hardware and Software Requirements Path MTU Discovery is not supported on UDP-type sockets. PVCs must be added specifying the IP address of the other endpoint. MIB is not supported. ARP table entries associated with PVCs are not aged. The retry-timeout value for connection attempts to an ATMARP server is 10 seconds. Maximum retry count is 5 before attempting another server if more than one is configured. Only 20 octet (40 digit) NSAP addressing is supported.
3 Configuring the NonStop TCP/IP Subsystem This section discusses a set of command files that you can use to start and configure your NonStop TCP/IP environment over a network connected to a NonStop system. This section also provides information about networking services provided by NonStop TCP/IP and explains how to configure the OSS environment to use NonStop TCP/IP. See Appendix A, Configuration Reference, for reference information about the DNS server and the standard resource record format.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment The sample environment shown in Figure 3-1 consists of a backbone communications link with three industry-standard routers connected to three different networks. These networks are made up of other NonStop systems.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment HOST1 does not use a Domain Name server for resolving the names of other hosts into their corresponding IP addresses. Thus, a HOSTS file is part of this sample configuration. The TCPIPUP2 File The following TACL command file starts the processes, adds and starts subsystem objects through SCF, and sets appropriate parameters.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment Type Syntax Description Line ADD DEFINE =TCPIP^HOST^FILE, FILE $SYSTEM.ZTCPIP.HOST S Sets the =TCPIP^HOST^FILE define to point to the desired HOSTS file. When set, this define tells the DNR to use the HOSTS file to translate host names to IP addresses. For information about the RESOLVER, see RESCONF Details on page 3-37 and the TCP/IP Programming Manual.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment Type Syntax Description Line TELSERV/NAME $ZTN0,CPU 2,NOWAIT,IN $ZHOME, & Starts the Telserv process. When an application (such as Outside View) connects to the IP address of the TCP/IP process, Telserv presents a window with a Welcome Banner and Service Menu. You can start TACL using the Service Menu.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment For the ADD SUBNET command, you can assign the SUBNET name to be anything you like, provided the name is no more than seven alphanumeric characters long and begins with an alphabetic character. The DEVICENAME attribute, which specifies the logical interface (LIF) name associated with the adapter that the NonStop TCP/IP process will access, is required.
Configuring the NonStop TCP/IP Subsystem Configuration 1: Startup Files for a Host in a Basic NonStop TCP/IP Environment Subnet Mask The subnet mask is not altered in this file because the default mask of %hFFFF0000 is adequate for a class B IP address on a network without subnets. The first two octets of the IP address are adequate for determining the proper network. Be aware that adding SUBNETs and routes is not the same as implementing subnetting.
Configuring the NonStop TCP/IP Subsystem Configuration 2: Startup Files for a Host in a Subnet Addressing Environment your own ECHO service. If you send an ECHO datagram to lan01 or lan02, you also test the actual physical network connection for your HOST1 NonStop TCP/IP environment. Configuration 2: Startup Files for a Host in a Subnet Addressing Environment This environment consists of a backbone communications link that has three routers leading to three subnets.
Configuring the NonStop TCP/IP Subsystem Configuration 2: Startup Files for a Host in a Subnet Addressing Environment Figure 3-2. NonStop TCP/IP Environment Using Subnet Addressing Backbone 128.30.128.1 RTR1 128.30.192.1 128.30.192.2 128.30.192.3 128.30.192.4 128.30.128.2 RTR2 HOST1 HOST2 128.30.64.5 128.30.64.6 HOST3 128.30.64.7 HOST4 128.30.128.3 RTR3 128.30.32.8 128.30.32.9 HOST5 128.30.32.10 HOST6 VST 008.
Configuring the NonStop TCP/IP Subsystem Configuration 2: Startup Files for a Host in a Subnet Addressing Environment The specifications for Logical Interface objects, which provide an interface between TCP/IP and the E4SA adapter, appear in boldface. Example 3-4.
Configuring the NonStop TCP/IP Subsystem Configuration 2: Startup Files for a Host in a Subnet Addressing Environment The configuration in SCFBNT allows HOST1 to communicate with the three subnets shown in Figure 3-2: 128.30.192.0 (the local subnet for HOST1 and HOST2) 128.30.64.0 (a remote subnet for HOST3 and HOST4) 128.30.32.0 (a remote subnet for HOST5 and HOST6) When you add SUBNET objects to the NonStop TCP/IP process, the process needs to know what mask is being applied to messages on that SUBNET.
Configuration 2: Startup Files for a Host in a Subnet Addressing Environment Configuring the NonStop TCP/IP Subsystem This plan will be simple to implement if you assign the subnet addresses beginning with the high-order bits and host IDs beginning with the low-order bits. In such a case, the first few subnet addresses would be 128, 64, 192, and so on. In this case, you might need to change the subnet number and the subnet mask. In this sample environment, suppose that you need to combine subnets 128.30.
Configuring the NonStop TCP/IP Subsystem WAN-Based Connections WAN-Based Connections This subsection describes startup files for configuring X.25 connections. Note. As for LAN-based connections, the SLSA subsystem must be operational before you can configure SUBNETs of type X25. Configuring an X25AM I/O Process on NonStop S-Series Systems and Integrity NonStop NS-series Systems If your application requires you to configure a TCP/IP SUBNET of type X25, you must have the X25AM I/O process configured first.
Configuring an X25AM I/O Process on NonStop S-Series Systems and Integrity NonStop Configuring the NonStop TCP/IP Subsystem Figure 3-3. Configuring a Subnet of Type X25 Processor 0 Legend $X25P1 1a 1b One possible path Alternative path t Da a t Pa h X25 SUBNET 1a Ethernet SUBNET 1b IP1 X25 IP0 $ZTC0 SUBNET X IP2 $ZTC1 Y ServerNet SAN G4SA SAC LAN0 PIF SAC LAN1 LAN2 PIF LAN3 PIF PIF IP0 Data Path Ethernet Port A SWAN Concentrator $ZZWAN.
Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection This subsection describes how to configure NonStop TCP/IP hosts that connect through an X.25-based network. Figure 3-4 illustrates an X.25 public data network (PDN) that connects a local NonStop TCP/IP host (\LA) to a remote NonStop TCP/IP host (\NY) across the country in New York.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Each host has an X25AM I/O process to control its SWAN concentrator. Each I/O process to has six subdevices (SUs). A maximum of 128 SUs per line is possible. The X25AM I/O process uses these SUs to provide multiple simultaneous connections over the X.25 line. SCF will be used to configure and start these SUs.
Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Configuring the NonStop TCP/IP Subsystem Configuring an X.25 Connection for \LA Consider the organization of the set of command files that you will be examining. There are four files to look at: UP1 is the main command file. It invokes X25SUBU1, TCPUP1, and STATUS1. X25SUBU1 configures the X.25 line and SUs, then starts the line. TCPUP1 sets the appropriate ADD DEFINEs and starts the TCP/IP process.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Command Description OBEY TCPUP1 invokes the command file to start and configure TCPIP, LISTNER, and TELSERV. TCISCF /IN STATUS1/ invokes SCF against the command file STATUS1 to display status information. SCP /NAME $ZNET, NOWAIT, CPU 1/0; AUTOSTOP -1 starts SCP and keeps it running. X25SUBU1 The command file X25SUBU1 adds and starts X.25 subdevices.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection SCF is invoked to run the commands contained in the command file X25SUBU1. This file contains the commands needed to alter line attributes and add SUs. Command Description ALLOW ALL instructs SCF to permit all errors to occur without exiting the command file. ASSUME LINE $X250 sets the assumed object type to LINE (specifically $X250 is the device name of the X.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection The following commands add five more SUs to this line. These SUs allow the NonStop TCP/IP process to communicate with the X25AM I/O process to establish an X.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Command Description ADD DEFINE =TCPIP^HOST^FILE, FILE MJHOSTS sets the ADD DEFINE =TCPIP^HOST^FILE. Whenever a TCP/IP process must convert from a host name to an IP address, the TCP/IP process either uses the DNS or a HOSTS file like MJHOSTS. When the ADD DEFINE =T=TCPIP^HOST^FILE is set, the TCP/IP process uses the file as indicated.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Command Description LISTNER /NAME $ZPORT, NOWAIT, PRI 160, CPU 2/3 PORTCONF invokes the LISTNER process using a process name of $ZPORT. $ZPORT uses $ZTC1 (TCP/IP process) and listens for connection requests from the processes listed in the file PORTCONF. CPU 2 starts the LISTNER in CPU 2. The LISTNER can run as a NonStop process pair; therefore, CPU 3 is specified as backup.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Command Description ALLOW ALL ERRORS permits SCF to continue running commands in the command file regardless of errors. ASSUME PROCESS $ZTC1 sets an SCF assumed object. All subsequent commands apply to this object unless explicitly entered otherwise. ALTER SUBNET #LOOP0, IPADDRESS 127.1 sets the IP address for the loopback SUBNET.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection STATUS1 The command file STATUS1 displays status information about the $X250 SUs, the NonStop TCP/IP process $ZTC1, and the $ZTC1 SUBNET and ROUTE objects. Example 3-10 shows a listing of the command file STATUS1: Example 3-10. Command File for Status Information on X.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection The TELSERV process name is $ZTN2. In TCPSUBU2: The SUBNET added specifies the device name $X25A and IP address 128.30.224.12 The LISTNER process name is $LSN2. The ADDRMAP entry (#A2) has an IPADDRESS specification of 128.30.224.11 and an X121ADDR specification of 00000011300000.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection Example 3-12. Command File for Adding X.25 Subdevices == SCF command file to add and start X25 subdevices for TCPIP on == \NY. allow all errors allow all warnings == SWAN concentrator: Line $X25A available.
Configuring the NonStop TCP/IP Subsystem Configuration 3: Startup Files for Two Hosts Using an X25AM-Based Connection IP address of \LA to its X121 source address, Example 3-14 shows a listing of the command file TCPSUBU2 (see Note on page 3-23 for information about obtaining the devicename in TCPSUBU2): Example 3-14. Command File for Adding an X.25 Subdevice == SCF command file to add a SUBNET to a TCP/IP process for an == X25 line allow all errors assume process $ztc2 alter subnet #loop0,ipaddress 127.
Configuring the NonStop TCP/IP Subsystem Networking Services Provided in the Guardian Environment Networking Services Provided in the Guardian Environment Section 1, Configuration Quick Start and Example 3-1 through Example 3-16 provided configuration files for starting NonStop TCP/IP in the Guardian environment. This subsection focuses on the LISTNER process, the services that LISTNER provides, and the Domain Name Server (DNS).
Configuring the NonStop TCP/IP Subsystem Configuring Subsystem Processes as that of the routes automatically added by the ADD SUBNET command. TCP/IP creates dynamic routes when it receives ICMP redirect packets. Dynamic routes have names of the form #DRT*. (When adding a route, note that the DSM object-naming conventions reserve #RT and #DRT; so select names that start with something else (such as #RO).) For a more detailed discussion of routing see TCP/IP Illustrated by W. Richard Stevens.
Configuring the NonStop TCP/IP Subsystem Configuring Subsystem Processes Starting the Domain Name Server (DNS) Normally, you start the domain name server (NAMED) when you start NonStop TCP/IP. To start NAMED, enter the following TACL command: TACL TACL TACL TACL 3> 3> 3> 3> RUN $SYSTEM.ZTCPIP.NAMED /NAME $ZNAME, PRI 140../ RUN $SYSTEM.ZTCPIP.NAMED /NAME $ZNAME, PRI 140../ [-p port-num] & [[-b] boot-file-name] & & NAMED listens to TCP and UDP port 53 for incoming requests.
Configuring the NonStop TCP/IP Subsystem Configuring Subsystem Processes valid. The SERVICES file contains information on the known services available on the Internet. Once the accuracy of the PORTCONF file content is verified against the SERVICES file, the LISTNER process listens to the configured ports, waiting for incoming connection requests from the remote client. LISTNER continues to wait for new connections on that port and other well-known ports.
Configuring the NonStop TCP/IP Subsystem Configuring Subsystem Processes Start the LISTNER process from a static or continuously available hometerm. If this condition is not possible, (as is the case when a dynamic terminal such as a Multilan dynamic window is used for startup operations), specify a local hometerm and IN and OUT files or specify $ZHOME (a terminal simulator) or use the virtual hometerm system, usually named $VHS.
Configuring the NonStop TCP/IP Subsystem Configuring Subsystem Processes FTP includes a standard set of commands and responses and, if applicable, reports corresponding state changes. For a detailed description of the FTP server and client commands, refer to the TCP/IP Applications and Utilities User Guide For more detailed technical information on FTP, refer to RFC 959 (DDN Protocol Handbook, Volume 2, DDN Network Information Center, December, 1985, pp. 2-739 through 2-808).
Configuring the NonStop TCP/IP Subsystem Configuration Files in the Guardian Environment Configuration Files in the Guardian Environment This section describes the files you can customize to configure your Internet environment. Configuration Files for the Internet Environment There are several files that come on your system update tape (SUT) that are used for configuring your internet environment. You need to customize some of these files to match your own NonStop TCP/IP environment.
Configuring the NonStop TCP/IP Subsystem Configuration Files for the Internet Environment HOSTS file NonStop TCP/IP hosts communicate through their IP addresses. However, IP addresses are not easy to remember. As a result, it is common practice to assign host names to IP addresses. Hosts then can be referred to by name. To provide for the translation, a Domain Name Resolver (DNR) is provided. A DNR, in turn, uses either a Domain Name Server (DNS) or a HOSTS file; configure the DNR to use one or the other.
Configuration Files for the Internet Environment Configuring the NonStop TCP/IP Subsystem Remember, you must issue such an ADD DEFINE command to indicate that a HOSTS file is to be used, as well as the name of the desired HOSTS file. Otherwise, the DNR assumes it must use a DNS and consults a RESCONF file. Also, note that you must set the TCPIP^HOST^FILE parameter at each terminal that uses the TCP/IP network.
Configuring the NonStop TCP/IP Subsystem Configuration Files for the Internet Environment Center (NIC) has authority over the root and the first level domains (for example, MIL, COM, GOV, EDU, NET, ORG). NonStop received permission from the NIC to add the subdomain NonStop. The network administrator at HP, however, now takes responsibility for all the subdomains that are created at NonStop. Domain names are written with the most specific name first and the most general last, for example: eeb.SWEd.nonstop.
Configuring the NonStop TCP/IP Subsystem Configuration Files for the Internet Environment comes with your NonStop TCP/IP software is a prototype; you should customize this file. To override the default RESCONF file, use the environment variable TCPIP^RESOLVER^NAME. This variable provides flexibility in selecting the RESCONF file accessed for name resolution. The default RESCONF file is located on $SYSTEM.ZTCPIP.
Configuring the NonStop TCP/IP Subsystem Configuration Files for the Internet Environment PROTOCOL File The PROTOCOL file contains the names of the protocols currently supported by the NonStop TCP/IP software as well as some not currently supported. Applications use the PROTOCOL file to get protocol names and Internet protocol numbers. When an application calls the functions getprotobyname or getprotobynumber, the PROTOCOL file provides this information. You do not need to alter this file.
Configuring the NonStop TCP/IP Subsystem Configuration Files for the Internet Environment Example 3-19.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Simple Mail Transfer Protocol (SMTP) The Simple Mail Transfer Protocol (SMTP, RFC 821 and RFC 1495) gateway implements the SMTP on NonStop systems running Guardian. The SMTP gateway uses the sockets interface to establish connection with Internet hosts. It provides an interface between TRANSFER mail systems and the TCPIP process.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) 3. Run the file $SYSTEM.ZTCPIP.SMTPRCV including the runtime option -bt, and specify the configuration file you created in step 1 using the run-time option -C configuration-file. RUN $SYSTEM.ZTCPIP.SMTPRCV -bt -Cconfiguration-file Note. The -C syntax for SMTPRCV differs from the -C syntax for SMTPSND; do not enter a space between -C and configuration-file for SMTPRCV.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Required Edits To set up SMTP for your organization, you only need to make a few changes to the SMTPCONF file. The editing required includes the following: Change the Class W (Cw) specification (Internet host-name) to your host-name. Change (DN) to the hostname that you want other hosts to know you as. Alter the macro A (DA) to add in the ARPANET gateway, if you have one.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) # Cwexpand localhost me # # This host-name is specified in the format by which you # want the outside world to know your host. Sometimes # there is confusion in the way the $w.$D macro is # represented, If your host-name already contains your # domain name. Host-names are obtained from the $ZTC0 # process. Check to see if the domain name is a # part of it. In any case, define the host-name here # in full format.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) # Enter the Transfer correspondent password here. Replace # with your password. Note that the password is # case sensitive, so be careful about mixing uppercase and # lowercase in the password. # OP # # Replace with the name of the System # Administrator to whom undeliverable mail is forwarded. # The default name is SYS-ADMIN.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Rule for rewriting—R The core of address parsing is the rewriting rules, which you specify in lines starting with R. These rules are ordered production systems. Each rule has a left-hand side and a right-hand side. When an address is to be rewritten, the address parser scans through the rule sets comparing the left-hand side against the specified address.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Tokens beginning with a dollar sign are expanded according to the following metasymbols: $n Substitute indefinite token n from the left-hand side $[name$] Canonicalize name $>n Call rule set n $@host Specify host $:user Specify user $#mailer Specify mailer The $n syntax substitutes the corresponding value from a $+, $*, $=, or $~ match on the left-hand side. You may use $n syntax anywhere.
Simple Mail Transfer Protocol (SMTP) Configuring the NonStop TCP/IP Subsystem Substitution occurs in the order described; parameters from the left-hand side are substituted, hostnames are canonicalized, subroutines are called, and, finally, $#, $@, and $: are processed. Rule set number—S The rules (R) are combined into rule sets, as specified by the S command. The syntax of this command is as follows: Sn The S command sets the number of the current rule set being collected to n.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Macros are interpolated using the construct $x where x is the name of the macro to be interpolated. In particular, lower-case letters are reserved for specifying special semantics which are passed information in or out of smtpgate. Some special characters are reserved for specifying conditionals. Specify conditionals by using the following syntax: $?x text1 $| text2 $.
Configuring the NonStop TCP/IP Subsystem v The version number of SMTPSND w The hostname of this site x The full name of the sender Simple Mail Transfer Protocol (SMTP) The $e macro is displayed when SMTP starts up, that is, when connection is first established with the mail RECEIVER. The first word must be the $j macro. The $j macro should be in RFC 821 format. The $n macro can be considered constant.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Option processing—O You can set many options from a configuration file. Options are represented by single characters. The syntax of this line is as follows: Oxstring This syntax sets option x to the value specified in string. The options are: C—Mail Correspondent Name. The syntax of this line is as follows: OCSMTP_MAILER This option sets the mail correspondent name to SMTP_MAILER.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Continuation lines are reflected directly into the outgoing message. The header-name specifies the name of the header. You can use (but are not limited to) the following header names: From Date Subject Received ResentFrom ResentDate Return-Path Message-Id The header-template is macro-expanded (processed by macros) before insertion into the message.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) Define Mailer—M Each mailer must have an internal name. This name can be arbitrary, except that the names local must be defined. The format of a define mailer line is: M mailer-name field=value [,field=value ...] where mailer-name is the name of the mailer (used internally only) and the field=value pairs define attributes of the mailer. Fields are: P - (Path) The mailer path. F - (Flags) Special flags for this mailer.
Configuring the NonStop TCP/IP Subsystem Simple Mail Transfer Protocol (SMTP) RFC 819 describes the specifics of domain-based addressing. These specifics are briefly described in RFC 822 as well. Essentially, each host is given a name that is a right-to-left, dot-qualified pseudo-path from a distinguished root. The elements of the path need not be physical hosts; the domain is logical. For example, at BigCityU one legal host may be a.CC.BigCityU.
Configuring the NonStop TCP/IP Subsystem rule rule rule rule rule rule rule rule rule rule rule Simple Mail Transfer Protocol (SMTP) set 3 returns: "user" "<" "@" "sun" "." "kci" ">" set 0 input: "user" "<" "@" "sun" "." "kci" ">" set 3 input: "user" "@" "sun" "." "kci" set 8 input: "user" "@" "sun" "." "kci" set 8 returns: "user" "@" "sun" "." "kci" set 6 input: "user" "<" "@" "sun" "." "kci" ">" set 6 returns: "user" "<" "@" "sun" "." "kci" ">" set 3 returns: "user" "<" "@" "sun" ".
Configuring the NonStop TCP/IP Subsystem Configuring the OSS Environment to Use NonStop TCP/IP rule set 24 returns: "^V" "tcp" "^W" "kciwrl" "." "kci" "." "com" "^X" "user" "%" "sun" "." "KCI" "<" "@" "kciwrl" "." "kci" "." "com" ">" Finally, rule set 4 is applied. rewrite: rule set 4 input: "^V" "tcp" "^W" "kciwrl" "." "kci" "." "com" "^X" "user" "%" "sun" "." "KCI" "<" "@" "kciwrl" "." "kci" "." "com" ">" rewrite: rule set 4 returns: "^V" "tcp" "^W" "kciwrl" "." "kci" "." "com" "^X" "user" "%" "sun" ".
Configuring the NonStop TCP/IP Subsystem OSS Sockets Support Files Note. You may need to be logged on as SUPER.SUPER to write to the /etc directory. To establish these files in the /etc directory, you can either copy their equivalents from the Guardian subvolume ZTCPIP or establish a symbolic link to the files so that any changes made to the Guardian files also apply to OSS.
Installing inetd Configuring the NonStop TCP/IP Subsystem Installing inetd In the OSS environment, the inetd daemon provides services similar to and in extension to services provided by the LISTNER process in the Guardian environment. Installing inetd requires that its object file and corresponding configuration file be copied from a Guardian subvolume to OSS directories.
Configuring the NonStop TCP/IP Subsystem Installing inetd Starting inetd To start inetd, enter the following command at the shell prompt: /usr/ucb/inetd If inetd is already running, you need to kill the first one before starting a new one. First, get the OSS process identification (OSS PID) of the running inetd daemon by using the ps command.
Installing inetd Configuring the NonStop TCP/IP Subsystem Split the service between the LISTNER process and the inetd daemon, (taking advantage of the fact that LISTNER supports only TCP ports), by commenting out the TCP port for the service in the inetd.conf file. In this case, the LISTNER process provides the service (on the TCP port) from the Guardian environment and the inetd daemon provides the service (on the UDP port) from the OSS environment. Example 3-22 shows part of an inetd.
4 SCF Reference This section provides information about: The Subsystem Control Facility (SCF) SCF commands available for TCP/IP (to find a command quickly, see Table 4-5 on page 4-9.) The PTrace facility SCF for NonStop TCP/IP SCF provides an operator interface to an intermediate process, the Subsystem Control Point (SCP), which in turn provides the interface to the I/O processes of the various subsystems.
ADDRMAP Object Type SCF Reference PROCESS ROUTE SERVER SUBNET Figure 4-1 shows that the ROUTE, SUBNET, ENTRY, and ADDRMAP object types are peers. The ROUTE, SUBNET, ENTRY, and ADDRMAP object types are subordinate to the PROCESS object. The SERVER is subordinate to the SUBNET (and can only be added or deleted when the SUBNET is in the STOPPED state.) This hierarchy is important when issuing commands to the NonStop TCP/IP subsystem for processing.
null Object Type SCF Reference The ENTRY object also allows you to view the ATM address table which shows the current ATM, SVC, and PVC connections on the TCP/IP process. Each ATM address entry is displayed with the name set to ATMentry. The ATM address entries are used to track existing ATM connections. You cannot add or delete these type of entries with SCF.
ROUTE Object Type SCF Reference ROUTE Object Type The ROUTE object establishes the path a data packet travels to reach its destination. Instead of specifying a full path, a route specifies the packet’s first host address and the packet’s destination. The first host then routes the packet to the next appropriate address in-route to the destination. This sequence repeats until the packet reaches the destination.
SUBNET Object Type SCF Reference The SERVER object name must be preceded by a pound sign (#) and can have at most seven alphanumeric characters. The first character following the pound sign must be a letter. SUBNET Object Type When you assign a name to a SUBNET object, you must assign a unique subnet-name to each SUBNET associated with a given process. The name must be preceded by a pound sign (#) and can have at most seven alphanumeric characters.
Naming Convention Summary SCF Reference Naming Convention Summary Table 4-3 summarizes the reserved names for each object type and the naming convention rules. Table 4-3.
Summary States SCF Reference The following example deletes all routes subordinate to $ZTC0 that start with #R and end with 5: -> DELETE ROUTE $ZTC0.#R*5 ? Use the question mark to represent a single unknown character in a specific position. For example, $ZTC0.#S?1 selects all object names subordinate to $ZTC0 that begin with #S, end with 1, and contain exactly one character between the #S and the 1. You can use wildcard characters in any combination.
NonStop TCP/IP SCF Commands SCF Reference SUBNET X X REGISTERED SERVER REGISTERING X CONNECTING X FREE STARTED ROUTE STARTING Object STOPPED Table 4-4. Valid Object Summary States X X X X X In the STARTED summary state, the object is available for data transfer. In the STOPPED summary state, the object is defined (that is, the object exists) but it is not available for data transfer. The STOPPED summary state is not applicable to the PROCESS object.
Supported Commands and Object Types SCF Reference The object specification, which shows the supported object types and object names. Descriptions, by object type, of the attribute. Considerations you should be aware of before using the command. Command examples. Note. The terms CPU and processor are used interchangeably in this section. Supported Commands and Object Types This section describes the SCF commands that are interpreted specifically for the NonStop TCP/IP subsystem.
Supported Commands and Object Types SCF Reference Table 4-5.
Entering SCF Commands SCF Reference Entering SCF Commands You start SCF interactively with the following TACL command: 1>SCF It is seldom necessary to specify SCF RUN parameters because the default values are appropriate for most situations. For a more detailed description of the TACL RUN command parameters that apply to SCF, refer to the SCF Reference Manual for GSeries RVUs or the SCF Reference Manual for H-Series RVUs.
ABORT Command SCF Reference You must not enter more than 2048 characters for any input command. Note. SCF accepts input from either a terminal or a disk (OBEY) file and directs output to either a terminal, disk file, or printer. However, in this manual, all examples assume that a terminal is being used for both input and output. The rest of this section describes each SCF command for the NonStop TCP/IP subsystem.
ABORT ROUTE Command SCF Reference Example The following command aborts and deletes a process named $ZTC0: 1-> ABORT PROCESS $ZTC0 ABORT ROUTE Command The ABORT ROUTE command terminates the operation of a route as quickly as possible; only enough processing is done to ensure the integrity of the subsystem. The object is left in the STOPPED summary state.
ADD Command SCF Reference Command Syntax ABORT [ / OUT file-spec / ] [ SUBNET subnet-spec ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. SUBNET subnet-spec names the point of connection between the NonStop TCP/IP process and an I/O device. The fully-qualified subnet-spec has the form: $process-name .
ADD ADDRMAP Command SCF Reference ADD ADDRMAP Command The ADD ADDRMAP command defines the name for the entry in the NonStop TCP/IP X.25 address table (an internal table that maps IP addresses to X.121 addresses). An ADD ADDRMAP command must be entered for each X.121 address. Command Syntax ADD [ / OUT file-spec / ] [ ADDRMAP addrmap-spec ] , IPADDRESS ip-address , X121ADDR x25-address OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
ADD ENTRY Command SCF Reference ADD ENTRY Command The ADD ENTRY command creates an entry in an ARP or ATMARP table. This is a sensitive command. Command Syntax ADD [ /OUT file-spec/ ] [ ENTRY entry-spec] , TYPE { ARP | ATMARP } , IPADDRESS ip-addr [, MACADDR mac-address ] [, ATMADDR atm-address ] [, PVCNAME “name” ] [, SUBNET “name” ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. ENTRY entry-spec specifies the name of the ENTRY object.
ADD ENTRY Command SCF Reference connections. For PVC connections, the PVCNAME and SUBNET attributes, not the ATMADDR attribute, are required. IPADDRESS ip-addr specifies the internet address for the entry and is specified in dotted decimal notation. MACADDR mac-address specifies the Ethernet or token-ring address for the ENTRY. It is entered as a string of twelve hexadecimal digits preceded by a “%”. The MACADDR attribute is only valid for entries of type ARP.
ADD ROUTE Command SCF Reference ADD ROUTE Command The ADD ROUTE command creates a route. Command Syntax ADD [ / OUT file-spec / ] [ ROUTE route-spec] , , [ , [ , [ , DESTINATION destination-ip-address GATEWAY gateway-ip-address DESTTYPE { HOST | BROADCAST } ] NETMASK mask-val ] METRIC metric-val ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. ROUTE route-spec specifies the path on which data is sent in order to reach a destination.
ADD ROUTE Command SCF Reference GATEWAY gateway-ip-address specifies the Internet address of the router or gateway host through which the network or host addressed in DESTINATION can be reached. This is a required attribute. Default: The subsystem supplies no default value for it unless the route is being added automatically as a part of the ADD SUBNET operation. The default value is the IP address of the SUBNET's host (the value for the SUBNET's IPADDRESS attribute).
ADD SERVER Command SCF Reference ADD SERVER Command The ADD SERVER command adds to a SUBNET the ATM addresses of ATMARP servers to use when trying to resolve the ATM address associated with an IP address. The ADD SERVER command requires both the SUBNET and ATMADDR attributes. The SUBNET name is entered as a quoted ASCII character string. The SUBNET must be in the STOPPED state before the ADD SERVER command is issued. This is a sensitive command.
ADD SUBNET Command SCF Reference The SUBNET name is entered as a quoted ASCII character string. The SUBNET must be in the STOPPED state before ADD SERVER can be issued to the SUBNET. ADD SUBNET Command The ADD SUBNET command creates an SCF SUBNET object. The SUBNET object associates a PROCESS object with a LIF, thereby connecting the NonStop TCP/IP process to a port on a communications adapter or X25AM line.
ADD SUBNET Command SCF Reference DEVICENAME device-name is the name of the SLSA LIF that provides access to the Ethernet or token-ring LAN, is the name of the ATM adapter, or is the name of the X25AM line that provides access to the X.25 network. For information on how to choose a SLSA device name, see the Hint on page 1-7. For information on how to choose the ATM device name, see the Hint on page 1-13.
ADD SUBNET Command SCF Reference ATMSEL byte-number is the ATM selector byte to use on an ATM type SUBNET. The default is 0 and valid value s are 0 to 255. GATEWAY ON | OFF ON enables the SUBNET to add a default IP address filter to Ethernet LAN adapters. For the NonStop TCPIP process to operate as a Gateway, you must configure it with more than one SUBNET, and you must add each SUBNET configured on an Ethernet LAN adapter with this option enabled.
SCF Reference ADD SUBNET Command The NonStop TCP/IP X.25 interface can open a maximum of 128 subdevices on a specified X25AM device (line). The specified X25AM device must be active and the subdevices must be added and in the STARTED summary state before you add an X.25 SUBNET. Refer to the X25AM Configuration and Management Manual for information on the methods for adding X.25 subdevices.
ALTER Command SCF Reference ALTER Command The ALTER command changes attribute values associated with the specified TCP/IP object. This is a sensitive command. ALTER ADDRMAP Command The ALTER ADDRMAP command is used to change the attribute values associated with a particular address map name.
ALTER PROCESS Command SCF Reference ALTER PROCESS Command The ALTER PROCESS command is used to change the attribute values of a process.
ALTER PROCESS Command SCF Reference TCPRECVSPACE window-size specifies the default window size used to receive data for the TCP protocol. You can specify values in the range 512 bytes through 262144 bytes. The value you specify affects NonStop TCP/IP subsystem performance. Although some applications may use values in the lower range, you should not set the value below 2K bytes. Default: 8K bytes. UDPSENDSPACE window-size specifies the default window size used to send data for the UDP protocol.
ALTER PROCESS Command SCF Reference HOSTNAME host-name specifies the name by which the NonStop system running the TCP/IP process is known on the Internet. It is a character string of 49 or fewer characters, enclosed in quotation marks. This value only supports the gethostname() socket call. Default: None. HOSTID host-id specifies the 32-bit number assigned to the NonStop system running the TCP/IP process. This is usually the host number portion of the IP address.
ALTER PROCESS Command SCF Reference together with the TCP/IP process stack. OFF declares that only the TCP/IP process stack is saved. Default: ON. ALLNETSARELOCAL ON | OFF The default is ON. ON causes TCP to use the interface MTU as a base for determining the TCP Maximum Segment Size (MSS) for each non-local TCP connection. A non-local TCP connection is one that goes to another network (not just another subnetwork).
ALTER PROCESS Command SCF Reference RFC1323-ENABLE ON | OFF is ON to cause TCP to support TCP Large Windows as documented in RFC 1323. When this option is enabled the TCP/IP process will use the TCP Window Scale and Timestamp options as described in RFC 1323. The largest TCP window supported is 262144 bytes when this option is enabled, and 65535 when the option is disabled. The default for this option is ON.
ALTER SUBNET Command SCF Reference ALTER SUBNET Command The ALTER SUBNET command is used to change the attribute values of a SUBNET. Command Syntax ALTER [ / OUT file-spec / ] [ SUBNET subnet-spec ] [ [ [ [ [ [ [ [ , , , , , , , IPADDRESS ip-address ] SUBNETMASK subnet-mask ] IRDP { ON | OFF } ] ADDALIAS ip-addr, SUBNETMASK %H0..FFFFFFFF ] DELETEALIAS ip-addr ] MTU size ] X25IDLETIME time ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
ALTER SUBNET Command SCF Reference Since non-contiguous mask values are not supported, netmask values such as 0.255.255.0, 255.0.255.0, 0.0.255.0 are considered invalid. IRDP enables (ON) or disables (OFF) the ICMP Router Discovery Protocol on the SUBNET interface. IRDP is a mechanism for locating default routers and is specified in RFC 1256. IRDP must also be enabled on any local LAN routers.
DELETE Command SCF Reference Considerations See Table 4-3 on page 4-6 for naming conventions and reserved object names. The object must be in the STOPPED summary state when the ALTER command is issued. When the ALTER command is completed, the object remains in the same summary state that existed before you issued the command. Use the INFO command to view the current attribute values.
DELETE ENTRY Command SCF Reference DELETE ENTRY Command The DELETE ENTRY command removes entries from the ARP or ATMARP tables. Entries can be deleted by specifying the entry name or by specifying the IP address which is the only way dynamically added entries can be deleted. This is a sensitive command. Command Syntax DELETE [ /OUT file-spec/ ] [ ENTRY entry-spec ] [ , IPADDRESS ip-addr ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
DELETE ROUTE Command SCF Reference DELETE ROUTE Command The DELETE ROUTE command removes a route from the NonStop TCP/IP subsystem. Only routes in the STOPPED summary state may be deleted. Command Syntax DELETE [ / OUT file-spec / ] [ ROUTE route-spec] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. ROUTE route-spec specifies the path on which data is sent in order to reach a destination. The fullyqualified route-spec has the form: $process-name .
DELETE SUBNET Command SCF Reference SERVER server-spec is the name of the server. The fully-qualified server-spec has the form: $process-name . #server-name If you specify the SCF object type (SERVER) or any portion of the object name (server-spec) in a prior ASSUME command, you can omit it in this command. For information about the ASSUME command, see the SCF Reference Manual for G-Series RVUs or the SCF Reference Manual for H-Series RVUs.
INFO Command SCF Reference Examples The following command deletes all SUBNETs under the process $ZTC0: 1-> DELETE SUBNET $ZTC0.* Considerations The object-name template (wildcard notation) is supported. When the DELETE operation is completed, the definition of the SUBNET you specified for deletion is removed from the subsystem. All routes defined through the SUBNET are deleted with the SUBNET. INFO Command The INFO command displays the current attribute values for the specified TCP/IP object.
INFO ENTRY Command SCF Reference Example The following command requests information about the specified ADDRMAP: 1->INFO ADDRMAP $ZTC0.#A27 INFO ADDRMAP Display Format The format of the display for the INFO ADDRMAP command is: SCF> INFO ADDRMAP $ZTC0.* TCPIP Info ADDRMAP \SYSA.$ZTC0.* Name #A1 Name #A2 IPADDRESS 123.456.78.9 X121ADDR 12345678901 Last Time Used 21 Sep 1994, 9:18:23.984 Last Dev/Subdev Used \SYS.$X25A.#TCPIP1 IPADDRESS 123.456.789.
INFO ENTRY Command SCF Reference OUT file-spec causes any SCF output generated for this command to be directed to the specified file. ENTRY entry-spec is the name of the entry. The fully-qualified entry-spec has the form: $process-name . #entry-name If you specify the SCF object type (ENTRY) or any portion of the object name (entry-spec ) in a prior ASSUME command, you can omit it in this command.
INFO ENTRY Command SCF Reference INFO ENTRY DISPLAY Format The format of the display for the INFO ENTRY command table is: TCPIP Info ENTRY \SAMCAT.$ZTCX.* Name: #A2 (ATMARP) IPADDRESS.... 172.16.192.1 AtmAddr... (NSAP) 47:00:05:80:FF:E1:00:00:00:F2:1A:29:EB:00:00:00:00:01:A1:00 Name: #PVC (ATMARP PVC) IPADDRESS.... 172.16.192.200 PVC Name..... #IP.PVC001 Name: (ATMARP) IPADDRESS.... 172.16.192.210 AtmAddr... (NSAP) 47:00:05:80:FF:E1:00:00:00:F2:1A:29:EB:00:00:00:00:01:B2:00 Name: (ARP) IPADDRESS.... 172.
INFO PROCESS Command SCF Reference INFO PROCESS Command The INFO PROCESS command displays the current attribute values for the TCP/IP process object. Command Syntax INFO [ / OUT file-spec / ] [ PROCESS process-name ] [ , DETAIL ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. DETAIL specifies that the display is to include additional detailed information on the object. PROCESS process-name is a valid process name indicating the desired TCP/IP process.
INFO PROCESS Command SCF Reference TCPSendSpace is the amount of data (in bytes) that can be buffered in the TCP layer when sending data to a remote site. TCPReceiveSpace is the amount of data (in bytes) that can be buffered in the TCP layer when receiving data from a remote site. UDPSendSpace is the amount of data (in bytes) that can be buffered in the UDP layer when sending data to a remote site.
INFO PROCESS Command SCF Reference TCP Receive Space is the amount of data (in bytes) that can be buffered in the TCP layer when receiving data from a remote site. UDP Send Space is the amount of data (in bytes) that can be buffered in the UDP layer when sending data to a remote site. UDP Receive Space is the amount of data (in bytes) that can be buffered in the UDP layer when receiving data from a remote site. Delay Ack Time is the amount of time (in .01-second units) that acknowledgments are delayed.
INFO PROCESS Command SCF Reference Program Filename is the name of the file being executed for this process. Debug is the current setting (ON or OFF) of the DEBUG attribute. Full Dump is the current setting (ON or OFF) of the FULLDUMP attribute.
INFO ROUTE Command SCF Reference RFC1323 Enable indicates whether or not TCP supports the TCP Large Windows feature, documented in RFC 1323. When you enable this option, the TCP/IP process uses the TCP Window Scale and Timestamp options described in RFC 1323. TCP Init Rexmit Timout is the initial retransmit timer value to use on a TCP connection. TCP Min Rexmit Timeout is the minimum value allowed for the TCP retransmission timeout.
INFO ROUTE Command SCF Reference EXAMPLES The following command displays the attribute values for a route object named #MR2: 1-> INFO ROUTE $ZTCO.#MR2 The following command displays the attribute values for all routes subordinate to the PROCESS object $ZTC0: 1-> INFO ROUTE $ZTC0.* INFO ROUTE Display Format The format of the display for the INFO ROUTE command is: SCF> INFO ROUTE * TCPIP Info ROUTE \SYSA.$ZTC0.* Name Subnet Netmask Destination Gateway Type #ROU11 #SN1 255.255.255.0 130.252.153.0 130.
INFO SERVER Command SCF Reference Type indicates one of three types of routes: blank routes to a network. G routes to a network through a gateway. H routes to a host. H,G routes to a host through a gateway. Metric indicates the number of hops to the destination. INFO SERVER Command The INFO SERVER command displays information about the server. This is a nonsensitive command.
INFO SUBNET Command SCF Reference INFO SERVER Display Format The display format for INFO SERVER is: TCPIP Info SERVER \SAMCAT.$ZTCY.* Name: #SRV2 Subnet #EN1 AtmAddr... (NSAP) 47:00:05:80:FF:E1:00:00:00:F2:1A:29:EB:00:00:00:00:01:A2:00 Name is the name of the server. AtmAddr is the ATM address of the server. Subnet is the name of the SUBNET to which the server is associated. INFO SUBNET Command The INFO SUBNET command displays the current attribute values for the specified SUBNETs.
INFO SUBNET Command SCF Reference Examples The first example returns information about all running SUBNETs on the system and the second example returns detailed information about a specific SUBNET: 1->ASSUME PROCESS $ZTC0 2->INFO SUBNET * 1->INFO SUBNET $ZTC0.#EN3, DETAIL INFO SUBNET Display Format The format of the display for the INFO SUBNET command is: SCF> INFO SUBNET * TCPIP Info SUBNET \SYSA.$ZTC0.* Name Devicename #LOOP0 \NOSYS.NOIOP N #SN1 \SYSA.LAN03 N #SN1 \SYSA.LAN03 N #SN2 \SYSA.
INFO SUBNET Command SCF Reference SuName is the name of the X.25 subdevice used by the SUBNET. QIO shows whether or not the SUBNET is currently using the QIO interface. QIO is always on for Ethernet, ATM and SNAP type SUBNETs. (This attribute is not applicable for X25AM SUBNETs.) ON indicates that the interface is currently using QIO mode. OFF indicates that the interface is not currently using QIO mode. R shows whether or not the ICMP Router Discovery Protocol (IRDP) has been enabled on the SUBNET.
INFO SUBNET Command SCF Reference IPADDRESS is the Internet address of this SUBNET and all the IP addresses of the aliases associated with the SUBNET. TYPE is the SUBNET type. It can be LOOPBACK, ATM, ETHERNET, SNAP or X.25. SUBNETMASK is a 32-bit integer that specifies which portion of the network number and the IP host address is to be masked to define a SUBNET. SuName is the name of the X.25 subdevice used by the SUBNET. QIO shows whether or not the SUBNET is currently using the QIO interface.
INFO SUBNET Command SCF Reference ATMARP Server Mode indicates whether the SUBNET has been configured to act as an ATMARP server or client. ATM SELECTOR BYTE is the value of the selector byte the SUBNET uses. INFO SUBNET Display With Detail for Ethernet The format of the INFO SUBNET with the detail option selected for an Ethernet SUBNET is: TCPIP Detailed Info SUBNET \NEWYORK.$ZTC4.* Name Devicename *IPAddRESS TYPE *SUBNETMASK SuName QIO *R #EN1\NY.$LAN03 172.17.214.
INFO SUBNET Command SCF Reference SUBNETMASK is a 32-bit integer that specifies which portion of the network number and the IP host address is to be masked to define a subnet. SuName is the name of the X.25 subdevice used by the SUBNET. QIO shows whether or not the SUBNET is currently using the QIO interface. QIO is always on for Ethernet, ATM and SNAP type SUBNETs. (This attribute is not applicable for X25AM SUBNETs.) ON indicates that the interface is currently using QIO mode.
LISTOPENS Command SCF Reference STARTING indicates that the multicast group is transitioning to the STARTED (and operational) state but is not yet fully operational. STOPPED indicates that multicast is not operational for the group. Considerations The object-name template (wildcard notation) is supported. The STATUS and STATS commands provide information on the summary state and the statistics of TCP/IP objects.
LISTOPENS PROCESS Command SCF Reference Examples The following command requests non-detailed and detailed information about the openers of the specified process: 1->LISTOPENS PROCESS $ZTC0 1->LISTOPENS PROCESS $ZTC0,DETAIL LISTOPENS PROCESS Display Format The format of the display for the LISTOPENS command without the DETAIL option is: SCF> LISTOPENS PROCESS $ZTC0 TCPIP Listopens PROCESS \SYSA.$ZTC0.
LISTOPENS PROCESS Command SCF Reference Lport is the local port number for either TCP or UDP, depending on the value of Protocol. The more common port values are displayed in text form; others are displayed as four-decimal octets. LISTOPENS Display Format With DETAIL The format of the display for the LISTOPENS command with the DETAIL option is a combination of the displays for the LISTOPENS and STATUS PROCESS commands: SCF> LISTOPENS PROCESS $ZTC0,DETAIL TCPIP Detailed Listopens PROCESS \SYSA.$ZTC0.
LISTOPENS PROCESS Command SCF Reference State is the current state of the socket; it applies only to sockets whose Proto value is TCP. The possible values are: CLOSING if waiting for a terminate connection request acknowledgment from the remote site. CLOSE-WAIT if waiting for a terminate connection request from the local user. ESTAB if the connection is open and the user can send and receive data. This is the normal state for data transfer.
NAMES Command SCF Reference Laddr is the local Internet address associated with the socket, displayed as four-decimal octets. Lport is the local port number for either TCP or UDP, depending on the value of Proto. The more common port values are displayed in text form; others are displayed as four-decimal octets. Faddr is the foreign (remote) Internet address associated with the socket, displayed in four-decimal octets.
NAMES ROUTE Command SCF Reference Command Syntax NAMES [ / OUT file-spec / ] [ ADDRMAP process-name ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. ADDRMAP process-name specifies the name of the TCP/IP process. If you omit the object name, SCF uses the assumed object name. For information about the ASSUME command, see the SCF Reference Manual for G-Series RVUs or the SCF Reference Manual for HSeries RVUs.
NAMES SUBNET Command SCF Reference ROUTE route-spec specifies the path on which data is sent in order to reach a destination. The fullyqualified route-spec has the form: $process-name . #route-name If you specify the SCF object type (ROUTE) or any portion of the object name (route-spec) in a prior ASSUME command, you can omit it in this command. For information about the ASSUME command, see the SCF Reference Manual for GSeries RVUs or the SCF Reference Manual for H-Series RVUs.
PRIMARY Command SCF Reference Example The following command requests the names of the SUBNETs associated with $ZTC0: SCF> NAMES SUBNET $ZTC0.* NAMES SUBNET Display Format The format of the display for the NAMES SUBNET command is: TCPIP Names SUBNET \SYSA.$ZTC0.* SUBNET #LOOP0 #EN1 #EN2 Considerations The object-name template (wildcard notation) is supported. PRIMARY Command The PRIMARY command can be used when the NonStop TCP/IP subsystem is running as a NonStop process pair.
START Command SCF Reference Examples The following command causes the $ZTC0 primary process to become the backup process (the attribute processor 6 identifies the processor where the former backup process resided): 1-> PRIMARY PROCESS $ZTC0, CPU 6 Considerations Run the NonStop TCP/IP process as a NonStop process pair.
START SUBNET Command SCF Reference Example The following commands start all routes under the assumed process: 1-> ASSUME PROCESS $ZTC0 2-> START ROUTE * START SUBNET Command The START SUBNET command creates implicit connections to and from a SUBNET. The SUBNET transitions through the STARTING state and, upon successful completion, ends in the STARTED summary state.
STATS Command SCF Reference STATS Command The STATS command returns statistics for a specified TCP/IP object. All statistics are 32-bit numbers. The letter D in the display values indicates that the value is a doubleword. Whenever a RESET option is included, the counters associated with the specified objects are displayed and reset to 0, and the timestamp for the reset is recorded.
STATS PROCESS Command SCF Reference STATS ADDRMAP Display Format The format of the display for the STATS ADDRMAP command is: SCF> STATS ADDRMAP $ZTC0.* TCPIP Stats ADDRMAP \SYS.$ZTC0.* Sample Time.. Reset Time... 08 SEP 1996, 10:15:04.541 08 SEP 1996, 09:10:03.123 Name IPADDRESS X121ADDR #DTE 123.456.78.9 12345678901234 Call-Out 3 Call-In 1 Last Time Used 08 SEP 1996, 09:12:16.123 Sample Time is the date and time when the statistics were sampled.
STATS PROCESS Command SCF Reference Command Syntax STATS [ / OUT file-spec / ] [ PROCESS process-name ] [ , RESET ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. PROCESS process-name is a valid process name indicating the desired TCP/IP process. If you omit the object name, SCF uses the assumed object name. For information about the ASSUME command, see the SCF Reference Manual for G-Series RVUs or the SCF Reference Manual for H-Series RVUs.
STATS PROCESS Command SCF Reference STATS PROCESS Display Format The format of the display for the STATS PROCESS command is: TCPIP Stats PROCESS \SYSA.$ZTC0 Sample Time ... 17 Oct 1996, 17:17:41.169 Reset Time .... 28 Sep 1996, 17:08:04.488 Bad Checksum.......... 0D Invalid Header Size... 0D Retransmitted Packets. 608D Total Packets Input... 103579D Incoming Connections.. 804D No Ports For Packets.. 9D Packets Unacknowledged 0D Connections Dropped... 0D Connections Closed.... 0D RTT Updated...........
STATS PROCESS Command SCF Reference The STATS PROCESS command display (continued): Bad Checksum.......... Invalid Header Size... Reflect Packets....... Bad ICMP Code......... In Echo Reply......... In Dest Unreachable... In Source Quench...... In Redirect........... In Echo............... In Time Exceeded...... In Parameter Problem.. In Timestamp.......... In Timestamp Reply.... In Info Request....... In Info Reply......... Bad Router Adv Subcode Bad Router Words/Addr. Router Advertisement..
STATS PROCESS Command SCF Reference Description of Statistics for the TCP Layer ACK Bytes Received is the number of ACK bytes acknowledged by received ACKs. ACK Packets Received is the number of ACK packets received. ACK Packets Sent is the number of ACK packets sent. ACK Predictions OK is the number of times the header predictions were correct for ACKs. Bad Checksum is the number of packets received with invalid checksum values.
STATS PROCESS Command SCF Reference network congestion. The retransmission timer expires when a packet is not acknowledged within a certain time. Packet retransmission can be caused by any of the following conditions: the network is overloaded; the other end of the connection is overloaded (so that appropriate acknowledgments cannot be received and/or sent); or a corrupted packet (that is, a packet with an invalid checksum) is received.
STATS PROCESS Command SCF Reference Duplicate PKTs Recv is the number of duplicate packets received. Embryonic Conn Dropped is the number of embryonic connections dropped. Established Connects is the number of connections established. Incoming Connections is the number of incoming connection requests. Invalid Header Size is the number of packets received with an invalid header size. This error usually indicates a problem between IP and TCP.
STATS PROCESS Command SCF Reference Partial Duplicate Byte is the number of duplicate bytes received in partially duplicate packets. Partial Duplicate PKTs is the number of packets received with some duplicate data. Persist Timeouts is the number of persistent timeouts. PKTs Recv After Close is the number of packets received after close. PKTs Recv After Window is the number of packets received exceeding the window boundary. Retransmitted Bytes is the number of bytes retransmitted.
STATS PROCESS Command SCF Reference SYN-C dropped, Overflow is the number of SYN cache entries dropped because of overflow. SYN-C dropped, RST is the number of SYN cache entries dropped because of RST. SYN-C dropped, Unreach is the number of SYN cache entries dropped because ICMP is unreachable. SYN-C dropped, Buktflow is the number of SYN cache entries dropped because of bucket overflow. SYN Cache Aborted is the number of SYN cache aborted (no memory).
STATS PROCESS Command SCF Reference Window Probes Sent is the number of window probes sent. Window Update PKT Sent is the number of window update packets sent. Window Probe PKTs Recv is the number of window-probes packets received. Window Update Pkts is the number of window update packets received. Description of Statistics for the UDP Layer Bad Checksum is the number of packets received with invalid checksum values. An invalid checksum is usually caused by a noisy link.
STATS PROCESS Command SCF Reference Description of Statistics for the IP Layer Bad Checksum is the number of packets received with invalid checksum values. An invalid checksum is usually caused by a noisy link. Bad Packet Size is the number of packets received with a packet length shorter than expected. This error is very similar to the Invalid Header Size and is usually caused by similar conditions. Fragments Dropped is the number of packet fragments dropped.
STATS PROCESS Command SCF Reference Packets Forwarded is the number of packets destined for another host that were forwarded. Packets Too Small is the number of packets that contained less data than was expected when the packet was read into the local buffers. This error usually indicates a problem with the local machine's buffering scheme. Short Packets is the number of packets that contained less data than specified in their header.
STATS PROCESS Command SCF Reference Description of Statistics for the ICMP Layer Bad Checksum is the number of packets received with invalid checksum values. An invalid checksum is usually caused by a noisy link. Bad ICMP Code is the number of packets received that contain invalid ICMP packet-type codes in the header.
STATS PROCESS Command SCF Reference Good Routes Recorded is the number of valid routes discovered by IRDP messages that have been entered in the TCP/IP route table. In Dest Unreachable is the number of Destination Unreachable (type 3) messages received. A Destination Unreachable message is sent to the NonStop TCP/IP subsystem when another host, gateway or router determines that a destination host or port is unreachable.
STATS PROCESS Command SCF Reference (functioning as a source host) that one of its datagrams has been discarded because the header parameters are incorrect. In Redirect is the number of Redirect (type 5) messages received. A gateway sends this message to the NonStop TCP/IP subsystem (functioning as a source host) to indicate that there is a shorter path to the destination through another router or gateway.
STATS PROCESS Command SCF Reference In Timestamp Reply is the number of Timestamp Reply (type 14) messages received. A host, router, or or gateway sends this message in reply to a Timestamp message. This message indicates the time in the original Timestamp message and the time at which the Timestamp message was received by the destination. The Timestamp facility is used to obtain the network time. Special applications can be written to use this facility.
STATS PROCESS Command SCF Reference Out Redirect is the number of Redirect messages sent. Out Source Quench is the number of Source Quench messages sent. Out Time Exceeded is the number of Time Exceeded messages sent. Out Timestamp is the number of Timestamp messages sent. Out Timestamp Reply is the number of Timestamp Reply messages sent. Router Advertisement is the number of IRDP discovery messages detected by the NonStop TCP/IP subsystem.
STATS PROCESS Command SCF Reference Maximum Data MDs Used is the maximum number of data MDs that have been in use. Maximum Dup MDs Used is the maximum number of duplicate MDs not assigned to inbound driver MDs that have been in use by the process. Max Dup Driv MDs Used is the maximum number of duplicate MDs assigned to inbound driver MDs in use by the process. Maximum MBUFs Used is the maximum number of MBUFs to be used. Maximum Pool Allocation is the maximum pool space used.
STATS PROCESS Command SCF Reference QIO Limit Warnings is the number of times the process received an event signifying a pool or an MD shortage from the QIO monitor. Total MBUFs Allocated is the current number of MBUFs allocated. Description of Statistics for Socket Send Size Histogram Size 1-128 is the count of socket sends between 1 and 128 bytes. Size 129-256 is the count of socket sends between 129 and 256 bytes. Size 257-512 is the count of socket sends between 257 and 512 bytes.
STATS PROCESS Command SCF Reference Description of Statistics for the ARP STATS In ARP Requests is the number of ARP requests received. Out ARP Requests is the number of ARP requests sent. In ARP Replys is the number of ARP replies received. Out ARP Replys is the number of ARP replies sent. In InARP Requests is the number of inverse ARP requests received. Out InARP Requests is the number of inverse ARP requests sent. In InARP Replys is the number of inverse ARP replies received.
STATS ROUTE Command SCF Reference Bad Checksum is the total number of IGMP packets received that had an incorrect checksum. Total Queries Input is the total number of IGMP query packets received. Bad Queries is the total number of IGMP query packets received with the IP destination address not equal to the all hosts group. Total Reports Input is the total number of IGMP membership reports received. Bad Reports is the total number of bad IGMP membership reports received.
STATS ROUTE Command SCF Reference information about the ASSUME command, see the SCF Reference Manual for GSeries RVUs or the SCF Reference Manual for H-Series RVUs. RESET resets the statistical counters to zero. (This option is sensitive.) Example The following commands request statistics about all running routes: 1-> ASSUME PROCESS $ZTCO 2-> STATS ROUTE.* STATS ROUTE Display Format The format of the display for a ROUTE object is: TCPIP Stats ROUTE \SYSA.$ZTC0.* Sample Time ... 23 March 1996, 17:18:25.
STATS SUBNET Command SCF Reference STATS SUBNET Command The STATS SUBNET command displays the NonStop TCP/IP subsystem statistics for the specified SUBNETs. This is a nonsensitive command unless the RESET option is specified. Command Syntax STATS [ / OUT file-spec / ] SUBNET subnet-spec [ , RESET ] [ , DETAIL ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
STATS SUBNET Command SCF Reference STATS SUBNET Display Format The format of the display for the STATS SUBNET command is: TCPIP Stats SUBNET \SYSTEM.$ZTC8.* Sample Time ... 19 Feb 1998, 9:00:56:.054 Reset time ... 18 Feb 1998, 21:09:10.986 Name Output Packets Timeouts #LOOP0 0D Input Output Packets Errors 0D Input Errors 0D 0D Filter Errors Filter 0D OD SAMPLE TIME ... 19 FEB 1998, 9:00:56:055 RESET TIME .... 19 FEB 1998, 7:36:15.
STATS SUBNET Command SCF Reference Input Errors is the number of errors detected when packets were received by the SUBNET. Each input error also generates one of the following operator messages: DEVICE READ ERROR error ON IOP iopname DEVICE WRITE ERROR error ON IOP iopname ERROR error ON IOP iopname Output Errors is the number of errors that occurred when packets were sent by the SUBNET.
STATUS Command SCF Reference STATUS Command The STATUS command reports the status of the specified TCP/IP object. The STATUS command with the detail option is only supported for the PROCESS object. STATUS ENTRY Command The STATUS ENTRY command displays the dynamic status of the specified entry.
STATUS ENTRY Command SCF Reference STATUS ENTRY Response Display The format of the STATUS ENTRY display is: TCPIP Status ENTRY \SAMCAT.$ZTC0.* Name: (ARP) IPADDRESS........ 172.16.119.1 Arp Timer........... 19 (Min) Arp Flags........ (INUSE,COM) MacAddress.......... %H00 000C 3920CE Name: #A2 (ATMARP) IPADDRESS........ 172.16.192.1 ATM Arp Timer....... 0 (Min) ATM Arp Flags...(INUSE,COM,PERM) Endpoint............ 0D ATM State Flags.. (None) AtmAddr...
STATUS ENTRY Command SCF Reference Name is the name of the entry. The entry type is indicated in parentheses to the right of the name. IPADDRESS is the IP address of the entry. Arp Timer is the time in minutes left to expire. Arp Flags is the state of the ARP table entry; possible values are: completed, permanent, or in use. AtmAddr is the ATM address of the destination. MacAddress is the MAC address of the entry in hexadecimal format.
STATUS PROCESS Command SCF Reference PVC Name shows the name of the PVC being used on the ATM adapter. Note that the name does not include the name of the ATM adapter and needs to be added in this manner. Subnet is the name of the SUBNET to which the PVC is associated. Considerations If the ENDPOINT parameter is supplied, you must issue the command specifying all ENTRY objects, for example: STATUS ENTRY *, ENDPOINT 1234567.
STATUS PROCESS Command SCF Reference Examples The following examples return status information without and with detail: 1->STATUS PROCESS $ZTC0 2->STATUS PROCESS \SYSA.$ZTC0, DETAIL STATUS PROCESS Display Format The format of the display for the STATUS PROCESS command is: TCPIP Status PROCESS \SYSA.$ZTC0 Status: STARTED PPID.................. ( 0,107) Proto TCP TCP TCP TCP State TIME-WAIT TIME-WAIT ESTAB TIME-WAIT Laddr 130.252.12.3 130.252.12.3 130.252.12.3 130.252.12.3 BPID..................
STATUS PROCESS Command SCF Reference ESTAB if the connection is open and the user can send and receive data. This is the normal state for data transfer. FIN-WAIT-1 if waiting for a terminate connection request from the remote TCP site or if waiting for acknowledgment of the terminate connection request that the process has sent previously. FIN-WAIT-2 if waiting for a termination of data to be received after having sent a FIN (termination of data being sent).
STATUS PROCESS Command SCF Reference Fport is the foreign port number for either TCP or UDP, depending on the value of Proto. The more common port values are displayed in text form; others are displayed as four-decimal octets. SendQ is the number of bytes of data in the send queue of the socket. RecvQ is the number of bytes of data in the receive queue of the socket.
STATUS PROCESS Command SCF Reference BPID is the process ID of the TCP/IP backup process. If TCP/IP is running without a backup process, this field shows ( 0, 0). Proto is the protocol associated with the socket, which can be UDP (for a UDP socket), TCP (for a TCP socket), or a protocol number (for a raw IP socket). State is the current state of the socket; it applies only to sockets whose Proto value is TCP.
STATUS PROCESS Command SCF Reference SYN-SENT if waiting for a SYN-ACK after having sent a SYN. TIME-WAIT if waiting for sufficient time to pass (about two round trips) to be sure that stray packets are flushed from the network. Laddr is the local Internet address associated with the socket, displayed as four-decimal octets. Lport is the local port number for either TCP or UDP, depending on the value of Proto.
STATUS ROUTE Command SCF Reference STARTING indicates that the multicast group is transitioning to the STARTED (and operational) state but is not yet fully operational. STOPPED indicates that multicast is not operational for the group. STATUS ROUTE Command The STATUS ROUTE command displays the dynamic status of the specified TCP/IP routes. Command Syntax STATUS [ / OUT file spec / ] [ ROUTE route-spec ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
STATUS SERVER Command SCF Reference STATUS ROUTE Display Format The format of the display for the STATUS ROUTE command is: TCPIP Status ROUTE \SYSA.$ZTC0.* Name Status #ROU11 #ROU9 #ROU12 #ROU8 #ROU3 STARTED STARTED STARTED STARTED STARTED RefCnt 0 0 0 1 1 Name is the name of the route. Status is the summary state of the route. RefCnt specifies the number of users currently using the specific route. If the value is greater than zero, an application is currently using the specified route.
STATUS SERVER Command SCF Reference Example The following command requests status information on all servers running on the system: 1>ASSUME PROCESS $ZTC0 2->STATUS SERVER STATUS SERVER Display Format The format of the display for the STATUS SERVER command is: TCPIP Status SERVER \SAMCAT.$ZTC0.* Name: #SRV2 ATM Endpoint Timer.. 487 (Sec) ATM State Flags.. (OUT,SRVR) Endpoint............ 805582288D State............ REGISTERED Subnet.............. #EN1 AtmAddr...
STATUS SUBNET Command SCF Reference State indicates the current state for the SERVER object. Possible values are: FREE Not is use. CONNECTING waiting for an ATM API Connect command to complete. REGISTERING Waiting for an ATMARP reply to an ATMARP request to register the SUBNET’s IP address with the ATMARP server. REGISTERED The SUBNET has successfully registered with the ATMARP server. Endpoint is the handle being used by the ATM address entry.
STOP Command SCF Reference Example The following commands request status information about the specified SUBNET: 1-> ASSUME PROCESS $ZTC0 2-> STATUS SUBNET #SN2 STATUS SUBNET Display Format The format of the display for the STATUS SUBNET command is: TCPIP Status SUBNET \SYSA.$ZTC0.* Name #LOOP0 #EN1 Status STARTED STARTED Name is the name of the SUBNET. Status is the summary state of the SUBNET. Considerations The object-name template (wildcard notation) is supported.
STOP ROUTE Command SCF Reference Command Syntax STOP [ / OUT file-spec / ] [ PROCESS process-name ] OUT file-spec causes any SCF output generated for this command to be directed to the specified file. PROCESS process-name is a valid process name indicating the desired TCP/IP process. If you omit the object name, SCF uses the assumed object name. For information about the ASSUME command, see the SCF Reference Manual for G-Series RVUs or the SCF Reference Manual for H-Series RVUs.
STOP SUBNET Command SCF Reference Example The following command terminates the operation of all routes under the assumed process: 1-> ASSUME PROCESS $ZTC0 2-> STOP ROUTE * STOP SUBNET Command The STOP SUBNET command terminates the activity of the specified SUBNETs in a normal manner. Command Syntax STOP [ / OUT file-spec / ] [ SUBNET subnet-spec] OUT file-spec causes any SCF output generated for this command to be directed to the specified file.
TRACE Command SCF Reference TRACE Command The TRACE command allows you to capture and store records that you can then display using the PTrace utility. The TRACE command can request the capture of data items, alter trace attributes that were set by a previous use of the command, or stop a previously requested trace operation. This is a sensitive command. Caution. The trace operation can significantly increase processor use by the TCP/IP process.
TRACE PROCESS Command SCF Reference SELECT selects the operations to be traced. For the PROCESS object, you can specify the following for select-spec: ALL All records. SOCKCMD Socket requests (bind, listen, accept, connect, send). MSGSYS Message system interface. MALLOC Resource allocation and deallocation events. ROUTING Requests for route changes. UDP IDP interface layer. TCP Transmission Control Protocol message layer. IP IP layer.
TRACE SUBNET Command SCF Reference Examples The following command traces the assumed process, writes results into the file named $DATA1.TRC.TRACE, allows the trace data to be overwritten when the EOF is reached, and selects tracing of all TCP/IP process activity: SCF> TRACE PROCESS, TO $DATA1.TRC.TRACE, WRAP, RECSIZE 300, & SELECT ALL TRACE SUBNET Command The TRACE SUBNET command traces a SUBNET.
TRACE SUBNET Command SCF Reference SELECT selects the operations to be traced.
VERSION Command SCF Reference Examples The following command traces the $ZTC0.#SN2 SUBNET, writes the results into the file $SYSA.TRACES.TCPSUB, allows the trace data to be overwritten when the EOF is reached, and traces all TCP/IP process activity on the SUBNET: SCF> TRACE SUBNET $ZTC0.#SN2, TO $SYSA.TRACES.TCPSUB, WRAP RECSIZE 300 & VERSION Command The VERSION command displays the NonStop TCP/IP subsystem version number, product name, product number, and RVU date.
NonStop TCP/IP Trace Facility SCF Reference Examples The second example shows the null command. ->VERSION PROCESS $ZTC0 ->VERSION ->VERSION PROCESS $ZTC0, DETAIL VERSION Command Display Format The format of the display of the VERSION command without the DETAIL option is: SCF> VERSION PROCESS $ZTC0 VERSION PROCESS \SYSA.
Introduction to PTrace SCF Reference cannot be printed or displayed directly. You use PTrace to display and examine the trace files. The PTrace program formats the data stored in these unstructured trace files for output to terminals, printers, or disk files. Figure 4-2 shows the four general steps involved in recording and formatting trace data. Figure 4-2. Recording and Displaying Trace Data Start the trace interactively with the SCF TRACE command or programmatically through SPI. Collect trace data.
PTrace Commands SCF Reference PTrace Commands The PTrace commands provide options for selecting trace records for display, so you can suppress those records that do not relate to the problem you are investigating. The PTrace commands also provide options for specifying the way in which the trace records are formatted. Although PTrace provides a common set of commands for displaying trace records, not all of the PTrace commands are supported by each subsystem.
PTrace Commands SCF Reference Table 4-7.
DETAIL Command SCF Reference DETAIL Command The DETAIL command controls the detailed display option. When DETAIL is set to ON, PTrace displays extended formatted versions of some records (for example, ARP traffic). Command Syntax DETAIL [ ON | OFF ] ON turns on detailed display mode. OFF turns off detailed display mode. Considerations The NonStop TCP/IP DETAIL command is implemented in the standards defined in the PTrace Reference Manual.
LABEL Command SCF Reference Considerations The NonStop TCP/IP HEX command is implemented in the standards defined in the PTrace Reference Manual. If the HEX command is not used, the OFF attribute is assumed. If HEX is specified without the ON or OFF attribute, the ON attribute is assumed. The RESET and FROM commands set the HEX command to OFF. LABEL Command The LABEL command controls the formatted display of trace records.
SELECT Command SCF Reference OFF turns off octal display mode. Considerations If the OCTAL command is not used, the OFF attribute is assumed. If OCTAL is specified without the ON or OFF attribute, the ON attribute is assumed. The RESET and FROM commands set the OCTAL command to OFF. SELECT Command The SELECT command establishes the selection criteria that control which trace records are to be displayed. Command Syntax SELECT [ [ [ mask ] keyword ] ( keyword [ , keyword ] ...
SELECT Command SCF Reference The following keywords apply to the SUBNET object: ALL All records IPI IP input records IPO IP output records ARPI ARP input records ARPO ARP output records LOGIC A combination of all the above records USERDATA Used with IPI and IPO to display user data Considerations If the SELECT command is not entered, the default mask and keyword is ALL. If the SELECT command is specified with no mask or keywords, the ALL keyword is assumed.
TEXT Command SCF Reference TEXT Command The TEXT command controls the text display option. When TEXT is set to ON, PTrace displays an interpreted text of trace-file records (excluding the record header). The textual display appears below labeled data, the HEX display, and the OCTAL display, if they are present. The textual display consists of ASCII characters, with control codes represented by two- or three-character mnemonics. Command Syntax TEXT [ ON | OFF ] ON turns on text display mode.
Trace Record Formats SCF Reference Type Record 11 ROUTE 12 SOCKCMD 13 UDPUREQ Each description includes the time when the record is generated, the record-type code, the text of the record, and the definitions of any values contained in the record.
Socket Creation Records SCF Reference type indicates the record-type code. The record-type code identifies the type of information contained in the record. It is subsystem-dependent. Socket Creation Records This subsection describes the formatted trace records displayed when the SOCKCR keyword is specified for the PTrace SELECT command. Note that all of the socket creation records are preceded by a header containing the record-type code 1.
Socket Creation Records SCF Reference Sofree Record The sofree record is generated each time the SOFREE procedure is called. The SOFREE procedure frees up a socket data structure. header procedure:sofree freeing socket_handle nnnaaa nnnnaaaa indicates the internal ID of the socket being freed. Socket Closing Record The socket closing record is generated when the process initiates the close of a socket.
Memory Buffer Allocation Records SCF Reference nnnnaaaa indicates the internal ID of the socket for which the new control block could not be created. Creating New TCPCB Record The creating new TCPCB record is generated each time a new control block is created for a TCP socket. header socket_handle nnnnaaaa creating new tcpcb for TCP socket nnnnaaaa indicates the internal ID of the socket for which the new control block is being created.
Interprocess Communication Records SCF Reference nnnnnnnnnn indicates the number of bytes allocated. rrrrrrr indicates whether the memory allocation attempt succeeded or not. The value can be succeed or failed. Interprocess Communication Records This subsection describes the formatted trace records displayed when the IPC keyword is specified for the PTrace SELECT command.
TCP Records SCF Reference Data Acked Record The data acked record is generated each time an ACK is received for the local socket. header socket_handle nnnnaaaa: acked ack-bytes, sb_cc unack_bytes nnnnaaaa indicates the internal socket ID. ack-bytes indicates the number of bytes of data acknowledged. unack-bytes indicates the number of bytes of data in the queue waiting to be acknowledged.
TCP Records SCF Reference After Changes Record The after changes record is generated each time data or an ACK is received for a TCP socket. Note that the values reported indicate the values of these variables after they have been updated by the packet. The preliminary values are reported in the send next record. header socket_handle nnnnaaaa: After Changes: snd_nxt snd-nxt, snd_una snd-una, snd_max snd-max nnnnaaaa indicates the internal socket ID. snd-nxt indicates the next sequence number to be sent.
TCP Records SCF Reference ti-ack indicates the sequence number of the data currently being acknowledged. Receive State Change Record The receive send state change record is generated when data is received. header socket_handle nnnnaaaa tcp_handle nnnnn init-state: input (start-no..end-no) @ ack-no, urp=urp [f1,f2,f3,f4,f5,f6] -> fin-state...
TCP Records SCF Reference [f1,f2,f3,f4,f5,f6] indicates the control flags set. The possible flags that can be set are SYN, ACK, FIN, RST, PUSH, and URG. fin-state indicates the final state after the data was received. The possible states are: CLOSE-WAIT LAST-ACK CLOSED LISTEN CLOSING SYN-RECVD ESTABLISHED SYN-SENT FIN-WAIT-1 TIME-WAIT FIN-WAIT-2 rcv-nxt indicates the next sequence number expected to be received. rcv-wnd indicates the receive window.
TCP Records SCF Reference snd-wnd indicates the send window. Send State Change Record The send state change record is generated when a user sends data. header socket_handle nnnnaaaa tcp_handle nnnnn init-state: user req-type -> fin-state... rcv_(nxt,wnd,up) (rcv-nxt, rcv-wnd, rcv-up) snd_(una,nxt,max) (snd-una, snd-nxt, snd-max) snd_(wl1,wl2,wnd) (snd-wl1, snd-wl2, snd-wnd) nnnnaaaa indicates the internal socket ID. nnnnn indicates the internal ID of the TCP packet.
TCP Records SCF Reference CONNECT RCVOOB CONNECT2 SEND CONTROL SENDOOB DETACH SENSE DISCONNECT SHUTDOWN FASTIMO SLOWTIMO LISTEN SOCKADDR fin-state indicates the final state after the data was sent. The possible states are: CLOSE-WAIT LAST-ACK CLOSED LISTEN CLOSING SYN-RECVD ESTABLISHED SYN-SENT FIN-WAIT-1 TIME-WAIT FIN-WAIT-2 rcv-nxt indicates the next sequence number expected to be received. rcv-wnd indicates the receive window.
TCP Records SCF Reference snd-max indicates the maximum sequence number that can be sent. snd-wl1 indicates the sequence number used for the last window update. snd-wl2 indicates the acknowledgment number used for the last window update. snd-wnd indicates the send window. Accepting Connection Record The accepting connection record is generated each time an incoming connection is accepted on a local socket.
UDP Input Records SCF Reference forgn-addr indicates the remote Internet address associated with the incoming connection. forgn-port indicates the remote port number associated with the incoming connection. TCP Socket Request Record The TCP socket request record is generated each time a TCP socket request is made. header socket_handle nnnnaaaa: tcp_usrreq: socket request #nnnnn nnnnaaaa indicates the internal socket ID. nnnnn indicates the internal request number used to manipulate the TCP socket.
Detailed UDP Input Records SCF Reference Sent UDP Packet to User Record The sent UDP packet to user record is generated each time a valid user is identified for an incoming UDP packet and the packet is delivered to the user. This record is preceded by a header containing the record-type code 5. header udp_input: Sent UDP packet to user --> udp_header_handle nnnnaaaa nnnnaaaa indicates the internal ID of the UDP packet.
UDP Output Records SCF Reference lllll indicates the packet's length. nnnnaaaa indicates the internal ID of the UDP packet. Source Address and Port Record The source address and port record is generated each time a UDP packet is received. This record is preceded by a header containing the record-type code 5. header udp_input: src ip-addr, sport portno udp_header_handle nnnnaaaa ip-addr indicates the packet's source Internet address. portno indicates the packet's source UDP port number.
IP Input Records SCF Reference UDP Sending to Record The UDP sending to record is generated each time the NonStop TCP/IP process sends a packet. header udp_output: sending to ip-addr.udp-port ip-addr indicates the destination IP address. udp-port indicates the destination UDP port number. IP Input Records This subsection describes the formatted trace records displayed when the IPI keyword is specified for the PTrace SELECT command.
IP Input Records SCF Reference w Problem (12) Timestamp (13) Timestamp Reply (14) Information Request (15) Information Reply (16) Forwarding to IP Address Record The forwarding to IP address record is generated each time the IP input routines receive a packet destined for another destination. header ipintr: ip_handle nnnnaaaa forwarding to ip address ip-addr nnnnaaaa indicates the internal ID of the IP packet. ip-addr indicates the address to which the packet is forwarded.
IP Output Records SCF Reference nnnnn indicates the IP protocol number (either 6 for TCP or 17 for UDP). Rebuilt Fragment Record The rebuilt fragment record is generated each time the IP input routines rebuild a packet from packet fragments. header ipintr: ip_handle nnnnaaaa rebuilt fragment len lllll nnnnaaaa indicates the internal ID of the IP packet. lllll indicates the rebuilt packet's total length.
Route Records SCF Reference ppppp indicates the IP number associated with the packet sent. For a list of the commonly used IP numbers, refer to the TCP/IP Programming Manual. For a complete list of the IP numbers, refer to Request for Comments document 1010, “Assigned Numbers.” Fragmenting Record The fragmenting record is generated each time the IP must fragment a packet. header ip_output: fragmenting offset bbbbb bbbbb indicates the IP offset of the fragments (in bytes).
Socket Command Records SCF Reference indicates whether the route is to a gateway or router, bit 3 indicates whether the route is to a point-to-point connection, bit 4 indicates whether the route is marked down, and bit 5 indicates whether the route is a dynamic route. Route Addition Record The route addition record is generated each time a route is added. Note that this record does not return any values.
Socket Command Records SCF Reference Accept Record The accept record is generated each time a connection is accepted on the local socket. header accept: socket_handle nnnnaaaa connection on 1234abcd.12345 nnnnaaaa indicates the internal socket ID. 1234abcd.12345 indicates the remote IP address and port number. Address Family Record The address family record is generated each time a connection request is received on the local socket.
Socket Command Records SCF Reference Connection Request Record The connection request record is generated each time a connection request is received on the local socket. header connect: socket_handle nnnnaaaa, to address 1234abcd.12345 nnnnaaaa indicates the internal socket ID. 1234abcd.12345 indicates the remote IP address and port number. Connection Waiting Record The connection waiting record is generated each time the socket has to wait for a connection to complete.
Socket Command Records SCF Reference Waiting for Reply Record The waiting for reply record is generated each time an accept call is not completed immediately (that is, if the socket has to wait for an incoming connection). header listen: socket_handle nnnnaaaa waiting for reply nnnnaaaa indicates the internal socket ID. Send Record The send record is generated each time a send call is made. header send: socket_handle nnnnaaaa bbbbb nnnnaaaa indicates the internal socket ID.
UDP User Request Records SCF Reference Socket Family Record The socket family record is generated each time a socket is created. header sock_reply: family fffff, type ttttt, proto proto fffff indicates the address family specified by the programmer in the socket call. ttttt indicates the socket type specified by the programmer in the socket call. proto indicates the IP number specified by the programmer in the socket call (either 0 for IP, 6 for TCP, or 17 for UDP).
UDP User Request Records SCF Reference nnnnn indicates the internal request number used to manipulate the UDP socket. The possible values that can appear and their meanings are explained in the PROTOSWH INCLUDE file. UDP Socket Request Completed Record The UDP socket request completed record is generated each time a UDP socket request is completed with an error.
A Configuration Reference This appendix provides reference material required for configuring the NonStop TCP/IP subsystem.
Configuration Reference Other NonStop TCP/IP Services DNSREV (Reverse Address-to-Name Mapping) on page A-14 Standard Resource Record Format on page A-14 $INCLUDE on page A-15 $ORIGIN on page A-16 SOA—Start Of Authority on page A-16 NS—Name Server on page A-17 A—Address on page A-17 WKS—Well-Known Services on page A-18 CNAME—Canonical Name on page A-18 HINFO—Host Information on page A-17 MB—Mailbox on page A-18 MR—Mail Rename Name on page A-19 MINFO—Mailbox Information
Configuration Reference Domain Name Server (DNS) boundaries. The DNS allows the authority for this information to be delegated to the organizations on the network that are responsible for it. The DNS, as defined in RFCs 1032, 1033, 1034, and 1035, allows a network to be divided into a hierarchy of domains. The name space is organized as a tree structure according to organizational or administrative boundaries.
Domain Name Server (DNS) Configuration Reference servers: a primary master and one (or more) secondary masters to provide backup service if the primary is unavailable or overloaded. A server may be a master for multiple domains, primary for some domains, and secondary for others. The functions of each type of server are: Primary Server A primary master server loads its data from a disk file. This server also can delegate authority to other servers in its domain.
Configuration Reference Domain Name Resolver which forwards queries and interacts with other name servers to resolve the query before returning the answer. An added benefit of using the forwarding feature is that the central machine develops a much more complete cache of information that all the workstations can use. The use of slave mode and forwarding is discussed in more detail under the description of the domain name server’s startup commands in Domain Name Server Files on page A-5.
Domain Name Server Files Configuration Reference The -b option specifies the boot file. The default boot file is $SYSTEM.ZTCPIP.DNSBOOT. Lines in the files are divided into fields. If a field in the line is optional, the notation [optional-field-value] is used. If the field is mandatory, the notation reqfield is used. If a field must be used as specified, the field is shown as it must be used (for example, IN means that the string IN should appear in the line in the position shown).
Domain Name Server Files Configuration Reference This record indicates that the server is a primary server for zonename, and datafile-name contains data for this zone-name. Primary master also should have a record for the IN-ADDR.ARPA domain of the format: primary internet_address.in-addr.arpa data-file-name For example: primary 32.128.in-addr.arpa $SYSTEM.ZTCPIP.DNSREV For an explanation of IN-ADDR.ARPA domain, refer to Domain Data Files, later in this section.
Domain Name Server Files Configuration Reference All servers must have a line in the boot file to prime the domain name server’s cache: cache . root-cache-file-name . $SYSTEM.ZTCPIP.NAMEDCA For example: cache Note that the period (.) between the keyword cache and the root-cache-filename is required. There is no default cache file. All cache files listed are read in at the NAMED startup (boot) time.
Domain Name Server Files Configuration Reference Remote Server To set up a host that uses a remote server instead of a local server to answer queries, set up the file $SYSTEM.ZTCPIP.RESCONF as specified in RESCONF File on page 3-36. This file designates which domain name servers on the network should be sent queries. Do not create (set up) this file if you have a local server running, since if this file exists, it is read almost every time gethostbyname or gethostbyaddr is called programmatically.
Domain Name Server Files Configuration Reference Example Boot Files The following section contains sample boot files for the different types of servers. A boot file is the startup file that specifies the mode of operation and specifies other information to NAMED. The boot file is named $SYSTEM.ZTCPIP.
Domain Name Server Files Configuration Reference If the primary has a different serial number, new information is loaded from the primary, and the DNSBAK file is updated. This domain name server is the secondary server for the reverse-address mapping of the domain 33.128.IN-ADDR.ARPA. The information required for this server is loaded from the file $SYSTEM.ZTCPIP.DNREVBAK. The version in the SOA record is compared with the primary server specified in the address 128.33.0.11.
Configuration Reference Domain Name Server Files DNSCACHE The DNSCACHE file primes the address of the root-domain name servers, basically giving the domain name server a location to start searches. An example of the DNSCACHE file is: ; Initial cache data for root domain servers. ; . 99999999 IN NS SRI-NIC.ARPA. 99999999 IN NS NS.NASA.GOV. 99999999 IN NS TERP.UMD.EDU. 99999999 IN NS A.ISI.EDU. 99999999 IN NS BRL-AOS.ARPA. 99999999 IN NS GUNTER-ADAM.ARPA. 99999999 IN NS C.NYSER.NET.
Configuration Reference Domain Name Server Files DNSHOSTS The DNSHOSTS file specifies the name-to-address mapping along with other information on the hosts. This file is not the same as the hosts file ($SYSTEM.ZTCPIP.HOSTS). An example of the DNSHOSTS file is as follows: @ IN SOA jiffy.kentcomm.com. joe.jiffy.kentcomm.com. ( 1.1 ; Serial 3600 ; Refresh 300 ; Retry 3600000 ; Expire 3600 ) ; Minimum IN NS pubs.kentcomm.com. IN NS jiffy.kentcomm.com. local host IN A 127.1 jiffy IN A 128.33.4 IN A 10.0.0.
Domain Name Server Files Configuration Reference In this example, the machine curly has two addresses (128.32.0.7 and 128.32.130.6). The machine type is an HP 9000 Model 750, running HPUX that has timed and TFTP services over UDP, and ECHO, TELNET, FTP, FINGER, and SMTP services over TCP. These entries should be refreshed every hour (3600 seconds). They expire in 1000 hours (3600000 seconds). This refresh time and expiration time is given to the other systems (such as secondary domain name servers).
Configuration Reference Domain Name Server Files The third field addr-class is the address class. There are currently two address classes: IN for Internet addresses and ANY for all address classes. The fourth field record-type states the type of the resource record. The record-specific fields record-specific-data are dependent on the type of the resource record. All comparisons and lookups in the domain name server database are case insensitive. The following characters have special meanings: .
Configuration Reference Domain Name Server Files $ORIGIN You can use the origin line to change the origin in a data file. The line starts in column 1 and is followed by a domain origin. Use this line to put more than one domain in a data file. For example: $ORIGIN kentcomm.com SOA—Start Of Authority The format of this record is: name [ttl] addr-class SOA origin person-in-charge ( serial-number refresh-time retry-time expire-time minimum-time ) For example: @ IN SOA pubs.kentcomm.com joe.pubs.kentcomm.
Domain Name Server Files Configuration Reference NS—Name Server The format of this record is: [name] [ttl] addr-class NS name-server-name For example: IN NS jiffy.kentcomm.com In the example, note that the name and the ttl fields are empty. The Name Server record (NS) lists a domain name server responsible for a given domain. The first field name lists the domain that is serviced by the listed domain name server. Each Primary Master server for the domain should have one NS record.
Domain Name Server Files Configuration Reference WKS—Well-Known Services The format of this record is: [name] [ttl] addr-class WKS address protocol services For example: pubs IN IN WKS WKS 128.33.0.11 128.33.0.11 UDP who route timed TCP (echo telnet discard sunrpc sftp uucp-path systat daytime) The Well-Known Service record (WKS) describes the well-known services supported by a particular protocol at a specified address.
Domain Name Server Files Configuration Reference For example: jim IN MB jiffy.kentcomm.com. A Mailbox record (MB) lists the machine where a user wants to receive mail. The name field is the user’s logon password (or correspondent name in KentComm Guardian machines); the machine field denotes the machine to which mail is to be delivered. Mail Box names should be unique to the zone.
Domain Name Server Files Configuration Reference An example for setting up a mail list is as follows: Bind IN IN IN IN IN IN MINFO MG MG MG MG MG Bind-Request kjd.BigCityU.Edu. Ralph.BigCityU.Edu. Zhou.BigCityU.Edu. Painter.BigCityU.Edu. Riggle.BigCityU.Edu. Terry.pa.XYZCorp.Com. MX—Mail Exchanger The format of this record is: name [ttl] addr-class MX preference-value mailer-exchanger For example: Miller.OY.AB. IN *.IN. IN MX MX 0 0 Seismo.ESS.GOV. RELAY.ES.NET.
B NonStop TCP/IP Processes and Protocols This appendix provides additional information about the NonStop implementation of Transmission Control Protocol/ Internet Protocol. It also includes information on the services provided by the NonStop TCP/IP environment not documented in Section 1, Configuration Quick Start and Section 3, Configuring the NonStop TCP/IP Subsystem. Note. This appendix is not intended to replace the reference books listed in About This Manual.
NonStop TCP/IP Processes and Protocols Addressing system appears as more than one logical host, a one-to-one correspondence between a logical host on a network and its Internet address on that network still exists. Multiple NonStop TCP/IP processes in a system on the same IP network must be represented by multiple Internet addresses, so that they appear to the IP network as multiple hosts.
Addressing NonStop TCP/IP Processes and Protocols 333 Ravenswood Avenue Menlo Park, CA 94025 Assigned Internet addresses are Class A, B, or C depending on the value of the network address. If your LAN is standalone or private, your LAN administrator can choose a private set of Internet addresses. In that case, typical use calls for all Class A addresses. As shown in Figure B-1, the class of the Internet address determines how the 32 bits are divided between the network address and the local address.
NonStop TCP/IP Processes and Protocols The Problem of Resolving Addresses networks can exist; however, each Class A network can have as many as 16,777,214 hosts, with addresses from %H000001 through %HFFFFFE (local addresses %H000000 and %HFFFFFF are reserved). Note. The local address 0 is never assigned to an individual host. Internet addresses that have a local address of 0 are always refer to the network itself. The local address of all 1’s is also never assigned to an individual host.
NonStop TCP/IP Processes and Protocols The Problem of Resolving Addresses Encapsulation of Messages Because TCP/IP uses encapsulation, (placing a message of a higher layer into the data portion of the lower layer which then adds its own header), we cannot rely solely upon IP addresses to communicate with other machines on the network.
The Problem of Resolving Addresses NonStop TCP/IP Processes and Protocols To address subnets within your LAN, your LAN administrator can further divide the local address part of your Internet addresses into a subnet number (identifying a particular subnet) and a host number (uniquely identifying a host system within the subnet). To identify what part of the Internet address represents the subnet number, a 32-bit subnet mask is used.
NonStop TCP/IP Processes and Protocols The Problem of Resolving Addresses Subnet Addressing Advantages There are many advantages of using subnet addressing. For one thing, it is transparent to the rest of the Internet. That is, the rest of the Internet does not know or care whether you have implemented subnet addressing on your own network. All routing and delivery mechanisms remain unchanged on the Internet.
The Problem of Resolving Addresses NonStop TCP/IP Processes and Protocols Example: Configuring an Intranet Using Subnetting Your company is headquartered in Omaha and has branch offices in Atlanta, Los Angeles, Seattle, and New York. The headquarters has 100 hosts, Atlanta has 17, Los Angeles has 30 hosts, Seattle has 12 hosts and New York has 65 hosts. The objective is to design the subnet masking scheme to accommodate all the offices and to allow for growth.
The Problem of Resolving Addresses NonStop TCP/IP Processes and Protocols Table B-1. Subnet Addressing Based on Current and Projected Needs Required Room for Growth Required Bits for Networks Required Bits for Hosts # Networks #Hosts Omaha 1 100 1,000 hosts, 25 networks 6 10 Atlanta 1 17 100 hosts, 2 networks 1 7 Los Angeles 1 30 100 hosts, 2 networks 1 7 Seattle 1 12 50 hosts, 2 networks 1 6 New York 1 30 100 hosts, 2 networks 1 7 2.
The Problem of Resolving Addresses NonStop TCP/IP Processes and Protocols Figure B-6. Splitting a Class B IP Address for Subnetting Physical Subnet Address Host Address Local Address 10000000 6 Bits for Networks 2 Bits for Hosts 00000011 All 8 Bits for Hosts VST 025.VSD 1. Allocate the subnets to each of the offices. Omaha has subnets 0-24, Atlanta has 25-26, Los Angeles has 27-28, Seattle has 29-30, and New York has 31-32. Figure B-7 shows Omaha’s subnets.
The Problem of Resolving Addresses NonStop TCP/IP Processes and Protocols Figure B-7. Omaha’s Subnets Subnet 0; Range 128 - 131.xx 1 0 0 0 0 0 Reserved for Subnets 0 0 0 1 128 129 1 1 0 1 130 131 Reserved for Hosts Subnets 1 - 23; Ranges 132 - 223 Subnet 24; Range 224 - 227.xx 1 1 1 0 0 0 0 0 1 1 Reserved for Subnets 0 1 0 1 224 225 226 227 Reserved for Hosts VST 026 .VSD 2.
Networking Technologies NonStop TCP/IP Processes and Protocols 4. Translate the subnet mask to hexadecimal format. Break each byte into two nibbles (four bits) and assign its hexadecimal value (hexadecimal ranges from 0 to F): 1111 1111 1111 1111 1111 1100 0000 0000 F F F F F C 0 0 5. Add the subnets for Omaha. For examples of command files, see Section 1, Configuration Quick Start.
Listing all Running NonStop TCP/IP Processes NonStop TCP/IP Processes and Protocols STATS, STATUS, and VERSION commands all return useful configuration and/or operation data, as follows: The NAMES command displays all the names currently configured for a specific object type (SUBNET, ROUTE, or WINDOW). The INFO SUBNET command displays the subnet name, the name of the SLSA line, the subnet’s Internet address, the subnet type, and the subnet mask.
NonStop TCP/IP Processes and Protocols NonStop TCP/IP Processes and Protocols The socket interface is modeled after the BSD socket interface to facilitate porting of existing UNIX TCP/IP applications to a NonStop system. For a detailed description of the socket library routines and a summary of the differences between the NonStop TCP/IP socket interface and the 4.3BSD UNIX interface, refer to the TCP/IP Programming Manual.
Supported Protocols NonStop TCP/IP Processes and Protocols Figure B-9. NonStop TCP/IP Subsystem Processes FUP FTP Server PATHWAY ECHO Server TACL TEDIT TELNET Subsystem Domain Name Server TFTP Server FINGER Server LISTNER Process SMTP Sockets Interface FTP Client HOST ECHO Client and/or Other Clients Transfer Mail LAN TCP/IP Subsystem TFTP Client TN6530 SLSA Subsystem Sun Workstation Ethernet Adapter SWAN Concentrator TELNET Client FINGER Client ECHO Client X.
NonStop TCP/IP Processes and Protocols NonStop TCP/IP Subsystem Processes LISTNER Process NonStop TCP/IP Protocols Transmission Control Protocol (TCP) User Datagram Protocol (UDP) Internet Protocol (IP) Internet Control Message Protocol (ICMP) Address Resolution Protocol (ARP) Subnetwork Access Protocol (SNAP) File Transfer Protocol (FTP) Trivial File Transfer Protocol (TFTP) Simple Mail Transfer Protocol (SMTP) ECHO Server/Client FINGER Server/Client Other Services
NonStop TCP/IP Process NonStop TCP/IP Processes and Protocols network; it accepts requests, performs the specified services, and returns the results to the clients. All supported servers and clients are described later in this section. Figure B-10.
NonStop TCP/IP Layered Architecture NonStop TCP/IP Processes and Protocols Figure B-11. NonStop TCP/IP Process FTP TFTP SMTP DOMAINNAME SERVER (DNS) TELNET TCP UDP IP ICMP ARP VST 030.VSD NonStop TCP/IP Layered Architecture The NonStop TCP/IP product provides a layered architecture that includes many commonly used TCP/IP applications (as shown in Figure B-12). Domain Name Server uses UDP for queries and TCP for messages such as zone transfers.
NonStop TCP/IP as a NonStop Process Pair NonStop TCP/IP Processes and Protocols Figure B-12. Layered Architecture of NonStop TCP/IP Software Other User Applications TN6530 FTP SMTP ECHO TELNET FINGE R TCP DOMAIN Other NAME TFTP Services SERVER (DNS) UDP IP with ICMP and ARP IP filter ARP filter LLC1 filter SLSA driver/interrupt handler VST 031.VSD NonStop TCP/IP as a NonStop Process Pair Run the NonStop TCP/IP process as a NonStop process pair.
TN6530 Emulation Utility NonStop TCP/IP Processes and Protocols creates a save abend dump file. The save file created only contains the stack area of the process and not the full QIO segment. If the backup abends during its initialization code, no save abend file is created. TN6530 Emulation Utility The multiple-page terminal emulation utility (TN6530) is a UNIX TELNET client that interfaces to NonStop TCP/IP and TELSERV. TN6530 emulates a 6530 terminal, including its block-mode support.
NonStop TCP/IP Processes and Protocols User Datagram Protocol (UDP) TCP and UDP allow IP to run several simultaneous sessions with a given host. Multiple sessions are accommodated by specifying a port number, which identifies the communication path, along with the Internet address. Each end of the communications path is assigned a port number for that session.
NonStop TCP/IP Processes and Protocols Internet Control Message Protocol (ICMP) protocol. Applications can build their own transport-layer protocols directly on this protocol by using raw sockets. Message Routing In the IP, messages called datagrams pass from a source (host) to a destination. An IP datagram may consist of one or several network-layer messages, or packets. Datagrams in the Internet often pass through a dozen different networks before reaching their final destination.
NonStop TCP/IP Processes and Protocols ICMP Router Discovery Protocol ICMP uses the Address Resolution Protocol (ARP) to convert 32-bit internet (IP) addresses into 48-bit Ethernet addresses. NonStop TCP/IP software uses the Internet Control Message Protocol (ICMP, RFC 792) along with IP to report errors and other control information from destination systems or gateways. The ICMP is considered a required part of any IP implementation.
NonStop TCP/IP Processes and Protocols Address Resolution Protocol (ARP) through 2-172). Also, refer to the book TCP/IP Illustrated by W. Richard Stevens, Prentice Hall, 1994. Address Resolution Protocol (ARP) The Address Resolution Protocol (ARP) dynamically binds a high-level Internet Address to a low-level physical hardware address. Two machines on a single physical network must know each other’s physical address to communicate.
NonStop TCP/IP Processes and Protocols Subnetwork Access Protocol (SNAP) For more detailed technical information on the ARP, refer to RFC 826 (DDN Protocol Handbook, Volume 3, DDN Network Information Center, December, 1985, pp. 3-615 through 3-624). Also, refer to the book TCP/IP Illustrated by W. Richard Stevens, Prentice Hall, 1994. Subnetwork Access Protocol (SNAP) The Subnetwork Access Protocol (SNAP) defines an interface between the IP layer and the Logical Link Control (LLC) layer.
Trivial File Transfer Protocol (TFTP) NonStop TCP/IP Processes and Protocols Figure B-15.
NonStop TCP/IP Processes and Protocols Simple Mail Transfer Protocol (SMTP) system. Because these programs are not connection-oriented, and they do not use any user-authentication mechanism, you must place some restrictions on the files that can be sent and received from a remote system. The TFTP server imposes certain restrictions when you retrieve a file from a NonStop system. You cannot retrieve a file unless it is readable to the network (unless its access security is NXXX).
NonStop TCP/IP Processes and Protocols Simple Mail Transfer Protocol (SMTP) The SMTP gateway also can relay SMTP messages between SMTP hosts. Acting as an intermediate node, it can store and forward SMTP messages. Messages received by the gateway from one SMTP host to be relayed to another SMTP host are first sent to TRANSFER. TRANSFER treats the SMTP gateway as a mail correspondent (SMTPGATE).
NonStop TCP/IP Processes and Protocols Simple Mail Transfer Protocol (SMTP) SMTPSND A separate process (called SMTPSND) is started by the system administrator as part of the NonStop TCP/IP startup mechanism. The SMTPSND process starts a session with TRANSFER by logging in as SMTPGATE (the gateway correspondent). The SMTPSND process then scans its INBOX for queued-up mail messages. Each message is examined for identifying recipients.
NonStop TCP/IP Processes and Protocols Simple Mail Transfer Protocol (SMTP) TCP/IP Configuration and Management Manual—427132-004 B-30
C Well-Known Port Numbers for TCP and UDP This appendix lists the port numbers preassigned to specific services when accessed from TCP or UDP. Table C-1, Table C-2, and Table C-3 give the name or names of each service as used in C programs. These port numbers are provided in the file $SYSTEM.ZTCPIP.SERVICES. Table C-1.
Well-Known Port Numbers for TCP and UDP Table C-2. Port Numbers for Host-Specific Functions Port Number Protocol C Name(s) of Service or Function 117 TCP uucp-path 119 TCP untp, usenet 123 TCP ntp 1524 TCP ingreslock Table C-3.
D SCF Command Summary This appendix lists the NonStop TCP/IP SCF commands in alphabetical order. This quick reference summarizes the commands for those already familiar with how the NonStop TCP/IP SCF commands function. For detailed information about the following commands, refer to Section 4, SCF Reference.
SCF Command Summary ADD [ /OUT file-spec/ ] [ SERVER entry-name ] , SUBNET “name” , ATMADDR atm address ADD [ / OUT file-spec / ] , , , [ [ [ [ [ [ SUBNET subnet-name ] TYPE { LOOPBACK | ATM | ETHERNET | SNAP | X25 } DEVICENAME device-name IPADDRESS ip-address , SUNAME subdevice-name ] , FORCEQIO { ON | OFF } ] , IRDP { ON | OFF } ] , ARPSERVER { ON | OFF } , ATMSEL byte-number ] ALTER [ / OUT file-spec / ] [ ADDRMAP addrmap-name ] [ , IPADDRESS ip-address ] [ , X121ADDR x25-address ] ALTER [ / OUT fi
SCF Command Summary ALTER [ / OUT file-spec / ] [ SUBNET subnet-name ] [ [ [ [ [ [ , , , , , IPADDRESS ip-address ] SUBNETMASK subnet-mask ] IRDP { ON | OFF } ] ADDALIAS ip-addr, SUBNETMASK %H0..
SCF Command Summary INFO [ / OUT file-spec / ] [ ROUTE route-name ] INFO [ /OUT file-spec/ ] [ SERVER server-name ] INFO [ / OUT file-spec / ] [ SUBNET subnet-name ] [, DETAIL ] LISTOPENS [ / OUT file-spec / ] [ PROCESS process-name ] [ , DETAIL ] NAMES [ / OUT file-spec / ] [ ADDRMAP process-name ] NAMES [ / OUT file-spec / ] [ ROUTE route-name ] NAMES [ / OUT file-spec / ] [ SUBNET subnet-name ] PRIMARY [ / OUT file-spec / ] [ PROCESS process-name ] , CPU cpu-number START [ / OUT file-spec / ] [
SCF Command Summary STATS [ / OUT file-spec / ] [ ADDRMAP addrmap-name ] [ , RESET ] STATS [ / OUT file-spec / ] [ PROCESS process-name ] [ , RESET ] STATS [ / OUT file-spec / ] [ ROUTE route-name [, RESET ] ] STATS [ / OUT file-spec / ] SUBNET subnet-name [ , RESET ] [ , DETAIL ] STATUS [ / OUT file-spec / ] [ ENTRY entry-name ] [ , IPADDRESS ip-addr ] [ , ENDPOINT endpoint-value ] STATUS [ / OUT file spec / ] [ PROCESS process-name ] [ , DETAIL ] STATUS [ / OUT file spec / ] [ ROUTE route-name ]
SCF Command Summary STOP [ / OUT file-spec / ] [ PROCESS process-name ] STOP [ / OUT file-spec / ] [ ROUTE route-name ] STOP [ / OUT file-spec / ] [ SUBNET subnet-name ] TRACE [ / OUT file-spec / ] [ PROCESS process-name ] { , STOP { , TO file-spec [ [ [ [ [ [ , , , , , , SELECT select-spec COUNT count NOCOLL PAGES pages RECSIZE size WRAP } ] ] ] ] ] ]...
E SCF Error Messages This appendix describes the NonStop TCP/IP subsystem SCF error messages. For the operator display of event messages, see the Operator Messages Manual. TCPIP 00001 TCPIP 00001 Invalid file name. Cause. You specified a file with an invalid format. Effect. The command is not executed. Recovery. Verify the file-name format and retry the command. TCPIP 00002 TCPIP 00002 INTERNAL ERROR: Case value out of range. Cause. An invalid case value was generated, with no associated case label.
SCF Error Messages TCPIP 00005 TCPIP 00005 Attribute value out of range attribute-name. attribute-name is the name of the attribute you specified in an ALTER PROCESS command. Cause. You specified a value for the ALTER PROCESS command that is outside the valid range. Effect. The command is not executed. Recovery. Enter a valid range for the command and retry it. Refer to the ALTER command in Section 4, SCF Reference, for more information on valid ranges. TCPIP 00007 TCPIP 00007 Duplicate address.
SCF Error Messages TCPIP 00010 TCPIP 00010 SNAP MTU not available. Cause. TCP/IP cannot communicate with the manager process to obtain the MTU size. Effect. The command is not executed. Recovery. Check or start the manager process. TCPIP 00011 TCPIP 00011 Invalid IP address. Cause. The IP address is invalid. Effect. The command is not executed. Recovery. Use a correct IP address. TCPIP 00012 TCPIP 00012 Invalid CPU number. Cause. The processor number is invalid. Effect. The command is not executed.
SCF Error Messages Effect. The command is not executed. Recovery. Use the RECSIZE parameter while starting a trace. When a larger trace record size is used, there is less chance of trace records being truncated. TCPIP 00019 TCPIP 00019 Primary not allowed, some subnets still in STARTED state. Cause. This occurs when a Primary command is rejected because at least one subnet is still in the started state and switching to another CPU. This takes any started subnets out of service. Effect.
SCF Error Messages Effect. The command is not executed. Recovery. Select a device that is available or correct the problem. TCPIP 00023 The Device selected for the subnet returned a NULL MAC address. Cause. The device selected for the subnet returned a NULL MAC address. Effect. The command is not executed. Recovery. Try the operation again. TCPIP 00024 Invalid ATM address Cause. ATM address is invalid. Effect. The command is not executed. Recovery. Use a correct ATM address.
SCF Error Messages TCPIP 00027 ATM adapter not configured in same CPU pair as TCPIP. Cause. ATM adapter is not configured in same CPU pair as TCP/IP. Effect. The command is not executed. Recovery. Restart TCP/IP in the same CPU pair as the ATM adapter. TCPIP 00028 ATM selector value already in use. Cause. A subnet is already added using the same ATM adapter and selector value. Effect. The command is not executed. Recovery. Add the subnet with a different ATM selector value.
F NonStop Systems Used as Internet Gateways Although networks in the Internet are normally connected to each other by industry-standard routers, these networks can be connected by devices called gateways. A gateway is a special-purpose, dedicated computer that attaches to two or more networks and routes packets from one to the other. Gateways route packets to other gateways until the packets can be delivered to the final destination directly across one physical network.
NonStop Systems Used as Internet Gateways Figure F-2 shows the same two networks illustrated in Figure F-1, but now part of NETB consists of subnets NETC and NETD. Figure F-2 shows the configuration as seen from within NETB. However, to the hosts in NETA, all hosts in NETB, NETC, and NETD appear simply as part of NETB. Figure F-2. Subnets Network NETB Internet Address = 37.0.0.0 Gateway Network NETA Internet Address = 98.0.0.0 HOST1 98.0.2.9 Gateway HOST2 98.0.3.1 HOST4 98.0.1.6 37.0.4.1 HOST6 37.0.8.
NonStop Systems Used as Internet Gateways Figure F-3. NonStop Systems in an Internet HOST1 NonStop System HOST2 NonStop System HOST3 37.1.0.8 Adapter Gateway Adapter LAN 2 LAN 1 VST 020.VSD A NonStop system connected as a host in an Internet likewise can act as a gateway between two subnets, as shown in Figure F-4. A single NonStop TCP/IP process runs in the NonStop system, which appears as a single host. In this example, the subnets are on different LANs. Figure F-4.
NonStop Systems Used as Internet Gateways Figure F-5. A NonStop System as Two Logical Hosts Connected to a Subnet LAN Subnet1 Internet Address = 37.0.1.0 HOST1-NonStop System Internet Address = 37.1.0.1 HOST2-NonStop System Internet Address = 37.1.0.2 HOST3 37.0.1.3 HOST4 37.0.1.4 Adapter Adapter HOST5 37.0.1.5 HOST6 37.0.1.6 VST 022.
Startup Files for HOST1 NonStop Systems Used as Internet Gateways Figure F-6. Three Gateways in the NonStop TCP/IP Environment Backbone 128.30.128.1 GTWY1 150.50.192.1 150.50.130.2 150.50.130.3 150.50.130.4 128.30.128.2 GTWY2 150.60.64.3 GTWY3 HOST2 150.60.64.1 150.60.64.2 128.30.128.3 HOST1 HOST3 HOST4 150.70.128.1 150.70.128.2 HOST5 150.70.128.3 HOST6 VST 040.VSD Startup Files for HOST1 HOST1 is a NonStop system that has two Ethernet 4 ServerNet Adapters (E4SAs).
NonStop Systems Used as Internet Gateways Startup Files for HOST1 comments, use the word “comment” or a double equal sign (==). Lines which call other files are discussed separately below. Note. The default NonStop TCP/IP process is $ZTC0 so you do not need to specify defines and params for NonStop TCP/IP applications such as LISTNER and TELSERV for the first part of Example F-1. Note. The TCP/IP primary and backup processes must be configured in CPUs that have access to the SAC. See Hint on page 1-7.
NonStop Systems Used as Internet Gateways Startup Files for HOST1 starts the TCP/IP process named $ZTC0 in processors 1 (primary) and 0 (backup). The RUN command: TCPIP/NAME $ZTC1, NOWAIT, CPU 1/0 starts the TCP/IP process named $ZTC1 in processors 1 (primary) and 0 (backup). Note that TCP/IP starts at a priority of 200, regardless of the priority specified in the RUN command. The line: ADD DEFINE =TCPIP^PROCESS^NAME, FILE $ZTC0 sets the =TCPIP^PROCESS^NAME parameter to $ZTC0.
NonStop Systems Used as Internet Gateways Startup Files for HOST1 Example F-2. SCFSBNT File for TCPIPUP2 === SCFSBNT ===== SCFSBNT ==== SCFSBNT ======== == SCF command file to ADD and START SUBNETs == This file is created to support HOST1 (Refer to == Figure F-6) ALLOW ALL ERRORS ALLOW ALL WARNINGS == ADD AND START SUBNET $ZTC0.#SN0 ASSUME PROCESS $ZTC0 ADD SUBNET #SN0,TYPE ETHERNET,DEVICENAME LAN01,IPADDRESS 150.50.130.2 == HELP TCPIP ALTER SUBNET ALTER SUBNET #LOOP0, IPADDRESS 127.
NonStop Systems Used as Internet Gateways Startup Files for HOST1 The address 127.1 or 127.0.0.1 is the standard for loopback operation. ROUTE Objects A ROUTE object is added for each remote subnet destination with which this host will need to communicate. The Route object specifies the destination network IP address and the gateway IP address to which this host is physically connected.
NonStop Systems Used as Internet Gateways Startup Files for GTWY1 Example F-3. HOSTS File for TCPIPUP2 ########## HOSTS FOR HOST1 ########## HOSTS FOR HOST1 ############ # Filename = \CB1.$SYSTEM.ZTCPIP.HOSTS # Date = January 31/93 150.50.192.1 GTWY1B gtwy1b gw1b 128.30.128.1 GTWY1 gtwy1 gw1 128.30.128.2 GTWY2 gtwy2 gw2 128.30.128.3 GTWY3 gtwy3 gw3 127.0.0.1 me loop 150.50.130.2 LAN01 lan01 con1 150.50.130.3 LAN02 lan02 con2 150.50.130.4 HOST2 host2 corp2 150.60.64.1 GTWY2B hgtwy2b gw2b 150.60.64.
NonStop Systems Used as Internet Gateways Startup Files for GTWY1 the files will be discussed: the most obvious differences are that there will be a single NonStop TCP/IP process ($ZTC0) and different routes will be added. The TCPIPUP3 File As the TCPIPUP3 file in Example F-4 shows, the only significant difference from the file for HOST1 (The TCPIPUP2 File on page F-5) is that a second NonStop TCP/IP process is not started. As a result, there is neither a second LISTNER nor a second TELSERV process.
NonStop Systems Used as Internet Gateways Startup Files for GTWY1 Example F-5. SCFSBNT File for TCPIPUP3 == == == == SCFSBNT FOR GTWY1========= SCFSBNT FOR GTWY1======== SCF command file to ADD and START SUBNETs This file is created to support GTWY1 (Refer to Figure F-6) ALLOW ALL ERRORS ALLOW ALL WARNINGS == ADD AND START SUBNET $ZTC0.#SN0 ASSUME PROCESS $ZTC0 == HELP TCPIP ADD SUBNET ADD SUBNET #SN0,TYPE ETHERNET,DEVICENAME LAN01,IPADDRESS 150.50.192.
NonStop Systems Used as Internet Gateways Startup Files for GTWY1 Example F-6. HOSTS File for TCPIPUP3 ########## HOSTS FOR GTWY1 ########## HOSTS FOR GTWY ############ # Filename = \CB1.$SYSTEM.ZTCPIP.HOSTS # Date = January 31/90 150.50.192.1 GTWY1 gtwy1 gw1 128.30.128.1 GTWY1B gtwy1b gw1b 128.30.128.2 GTWY2 gtwy2 gw2 128.30.128.3 GTWY3 gtwy3 gw3 127.0.0.1 me loop 150.50.130.2 HOST1 host1 h1 150.50.130.3 HOST1B host1b h1b 150.50.130.4 HOST2 host2 corp2 150.60.64.1 GTWY2B gtwy2b gw2b 150.60.64.
NonStop Systems Used as Internet Gateways TCP/IP Configuration and Management Manual—427132-004 F-14 Startup Files for GTWY1
Glossary This glossary defines terms used both in this manual and in other NonStop TCP/IP manuals. Both industry-standard terms and NonStop terms are included. Because this is a glossary for NonStop TCP/IP as a whole, not all of the terms listed here appear in this manual. address mask. A bit mask used to select bits from an Internet address for subnet addressing. The mask is 32 bits long and selects the network portion of the Internet address and one or more bits from the local portion. address resolution.
ATM Glossary ATM. See Asynchronous Transfer Mode (ATM). bridge. A router that connects two or more networks and forwards packets among them. Usually, bridges operate at the physical network level. For example, an Ethernet bridge connects two physical Ethernet cables and forwards from one cable to the other exactly those packets that are not local. Bridges differ from repeaters; bridges store and forward complete packets, while repeaters forward electrical signals. broadcast.
connection Glossary connection. The path between two protocol modules that provides reliable stream delivery service. In the Internet, a connection extends from a TCP module on one machine to a TCP module on another machine. connectionless service. Characteristic of the packet delivery service offered by most hardware and by the Internet Protocol (IP). The connectionless service treats each packet or datagram as a separate entity that contains the source and destination address.
Distributed Systems Management Glossary Distributed Systems Management. A set of tools used to manage NonStop systems and EXPAND networks. These tools include the VIEWPOINT console application, the Subsystem Control Facility (SCF) for data communications subsystems, the Subsystem Programmatic Interface (SPI), the Event Management Service (EMS), the Distributed Name Service (DNS), and token-oriented programmatic interfaces to the management processes for various NonStop subsystems. distributor.
Ethernet Glossary Ethernet. A popular local area network technology invented at the Xerox Corporation Palo Alto Research Center. An Ethernet itself is a passive coaxial cable; the interconnections all contain active components. Ethernet is a best-effort delivery system that uses CSMA/CD technology. Xerox Corporation, Digital Equipment Corporation, and Intel Corporation developed and published the standard for 10 Mbps Ethernet. Ethernet 4 ServerNet adapter (E4SA).
fabric Glossary fabric. A simplified way of representing a complex set of interconnections through which there can be multiple and (to the user) unknown paths from point to point. The term fabric is used to refer to the X or Y portion of the ServerNet system area network (ServerNet SAN), for example the X fabric. Fast Ethernet ServerNet adapter (FESA).
Gateway to Gateway Protocol Glossary any machine that transfers information from one network to another, as in “mail gateway.” Gateway to Gateway Protocol. The protocol core gateways used to exchange routing information, GGP implements a distributed shortest path routing computation. Under normal circumstances, all GGP participants reach a steady state in which the routing information at all gateways agrees. G4SA. See Gigabit Ethernet 4-Port ServerNet adapter (G4SA). GESA.
hop count Glossary of header tokens by passing appropriate token codes to SSGET and can change the values of some header tokens by passing their token codes to SSPUT. Examples of header tokens for commands are the command number, the object type, the maximum-response token, the server-version token, the maximum-field-version token, and the checksum token. Command and response messages contain a specified set of header tokens; event messages, a different set with some overlap. See also SPI message.
Internet address Glossary Internet address. The 32-bit address assigned to hosts that want to participate in the Internet using TCP/IP. Internet addresses are the abstraction of physical hardware addresses, just as the Internet is an abstraction of physical networks. Actually assigned to the interconnection of a host to a physical network, an Internet address consists of a network portion and a host portion. The partition makes routing efficient. Internet Control Message Protocol (ICMP).
LANMON Glossary LANMON. See LAN monitor (LANMON) process. LAN monitor (LANMON) process. The process provided as part of the ServerNet local area network (LAN) systems access (SLSA) subsystem that has ownership of the Ethernet adapters controlled by the SLSA subsystem. LAPB (Link Access Protocol —Balanced). ITU-T standards that define in the Data Link Layer the requirements for X.25 connections over wide area networks (WANs). Level 2.
management process Glossary management process. In DSM, a process through which an application issues commands to a subsystem. A management process can be part of a subsystem, or it can be associated with more than one subsystem; in the latter case, the management process is logically part of each of the subsystems. SCP is the management process for all NonStop data communications subsystems that support DSM. See also subsystem. manager process.
object Glossary before returning control to the caller. In order to make the called procedure wait for the completion of the operation, the application calls a separate procedure. Compare wait mode. object. (1) In general HP NonStop use, one or more of the devices, lines, processes, and files in a NonStop subsystem; any entity subject to independent reference or control by one or more subsystems.
PDN Glossary PDN. See Public Data Network (PDN). physical interface (PIF). The hardware components that connect a system node to a network. physical layer. Layer 1 in the OSI Reference Model. This layer establishes the actual physical connection between the network and the computer equipment. Protocols at the Physical Layer include rules for the transmission of bits across the physical medium and rules for connectors and wiring. PIF. See “physical interface (PIF).” PING. See PING. predefined value.
return token Glossary also contain error lists that include error tokens. A response can consist of multiple response records, spread across one or more response messages. A response record cannot be split between two response messages. If multiple response records are in a response message, each response record is enclosed in a data list. See also data list. Each response record is required to contain a return token. See also return token. return token.
session Glossary layers: the lower layer is the service provider, and the upper layer is the service user. Compare protocol. session. For a management application, the period during which an application can issue commands to a subsystem. Simple Mail Transfer Protocol (SMTP). The Internet standard protocol for transferring electronic mail messages from one machine to another.
subnetwork Glossary addressing, routing continues as usual by dividing the destination address into an Internet portion and local portion. Gateways and hosts inside a site using subnet addressing interpret the local portion of the address by dividing it into a physical network portion and host portion. subnetwork. One or more intermediate systems that provide relaying and through which end open systems may establish network connections. Subnetwork Access Protocol (SNAP).
summary state Glossary summary state. In DSM interfaces to data communications subsystems, one of the generally defined possible conditions of an object, with respect to the management of that object. A summary state differs from a state in two ways. First, a summary state pertains to the management of an object, whereas a state may convey other kinds of information about the object.
token number Glossary token number. In DSM programmatic interfaces, the number used by a subsystem to identify each DSM token that it defines. The token type and the token number together form the token code. token ring. 1)þþThe token access procedure used on a network with a sequential or ring topology. (2) A data link level protocol designed to transfer data over ring-oriented LANs.
wide area network (WAN) Glossary assigned to echo servers, time servers, remote login (TELNET) servers, and file transfer (FTP) servers. wide area network (WAN). A network that operates over a larger geographical area than a local area network (LAN)—typically, an area with a radius greater than one kilometer. The elements of a WAN may be separated by distances great enough to require telephone communications. Contrast with local area network (LAN). wide area network (WAN) subsystem.
Glossary See WAN manager process.
Index Numbers 6530 terminal B-20 A ABORT command PROCESS 4-12 example 4-13 ROUTE 4-13 example 4-13 SUBNET 4-13 examples 4-14 ACK Predictions OK 4-69 Adapter ATM3SA 2-3 E4SA 2-3 FESA 2-3 G4SA 2-3 GESA 2-4 TRSA 2-4 ADD Subsystem Control Facility command 3-28 ADD command ADDRMAP example 4-15 IPADDRESS attribute 4-15 X121ADDR attribute 4-15 ENTRY IPADDRESS attribute 4-17 MACADDR attribute 4-17 PVCNAME attribute 4-17 SUBNET attribute 4-17 TYPE attribute 4-16 ROUTE DESTINATION attribute 4-18 DESTTYPE attribute 4
A Index Address parsing (continued) rule line syntax 3-45 rule set number 3-48 Address Resolution Protocol B-24 converting IP addresses into Ethernet addresses B-22 description B-24 Internet-to-physical address translation B-24 using Ethernet frames B-24 ADDRMAP ALTER command parameter 4-25 ADDRMAP attribute 4-33 ADDRMAP names 4-2 ADDRMAP object 4-1 ADDRMAP object type 4-2 ADDRMAP, object type in ADD command 4-15 ALTER command 4-25 DELETE command 4-33 INFO command 4-37 NAMES command 4-58 STATS command 4-6
B Index ATM and ENTRY object 4-3 examples 1-12/1-16 hardware and software requirements 2-8 interface 2-4 overview 2-6/2-8 relationship to system 2-1 RFC compliance 2-7 term definitions 2-7 ATM address displaying 4-92 endpoint 4-91 handle 4-91 PVC 4-91 ATM address entry 4-91 ATM Address Table definition 2-7 viewing 4-3 ATM address table viewing 4-40 ATM Arp Flags 4-91 ATM Arp Timer 4-91 ATM Endpoint Timer 4-100 ATM SELECTOR BYTE attribute 4-52 ATM State Flags 4-91, 4-100 ATM3SA 2-3 AtmAddr 4-91, 4-101 ATMA
C Index C D Caching-Only Server record in example boot files A-11 Call-In attribute 4-65 Call-Out attribute 4-65 Canonical Name A-18 CIDR See Classless Inter-Domain Routing Class A Internet address B-3 Class B Internet address B-4 Class C Internet address B-4 Classless Inter-Domain Routing 4-19 Client TCP/IP definition B-16 CLOSE-WAIT socket state 4-93 CLOSE-WAIT state 4-57 CLOSING socket state 4-93 CLOSING state 4-57 CNAME See Canonical Name Comer, Douglas 3-33, B-21, B-22, B-23 Commands 4-9 PTrace 4-1
D Index Destination Service Access Point B-25 DESTYPE attribute 4-19 DETAIL command 4-114 DETAIL command parameter 4-41 Detailed UDP input records 4-132 Device type and subtype 4-111 DEVICENAME G-series considerations 4-24 DEVICENAME attribute 4-22 Devicename attribute 4-50, 4-52 Display example INFO ADDRMAP 4-38 INFO PROCESS 4-37, 4-41 INFO PROCESS, DETAIL 4-42 INFO ROUTE 4-46 INFO SUBNET 4-49 LISTOPENS PROCESS 4-55 with DETAIL option 4-56 NAMES ADDRMAP 4-59 NAMES ROUTE 4-60 NAMES SUBNET 4-61 STATS ADDRM
E Index Domain Name Server (continued) format (continued) Well-Known Services A-18 $INCLUDE A-15 $ORIGIN A-16 hardware/software information A-3 machine information A-3 mailbox information A-3 Master Server A-3 Primary Server A-3 Secondary Server A-3 name-to-address mapping A-3 Remote Server A-4 resource records A-14 Slave Server A-4 types A-3 using A-5 Domain Name System 3-36 DSAP See Destination Service Access Point DSM See Distributed Systems Management Dup Driver MDs In Use attribute 4-80 Dup MDs in Us
G Index Files (continued) DNSLOCAL A-9 DNSREV A-9 HOSTS A-5 MBOX default name A-15 PORTCONF 3-30, 3-32 PORTCONF default name 3-41 PROTOCOLS B-22 RCVSMTP default name 3-41 RESCONF A-5 RESCONF default name A-11 SMTPCONF 3-41 SMTPCONF default name 3-42 Standard Resource Record Format file A-14 Filter Errors statistic 4-87 Filter Timeout statistic 4-87 filters definition of 2-6 FINGER application layer 3-33 Client/Server 3-33 FIN-WAIT-1 socket state 4-94 FIN-WAIT-1 state 4-57 FIN-WAIT-2 socket state 4-94 FIN-
I Index I ICMP Router Discovery Protocol 4-22, 4-32 IEN See Internet Engineering Note In ARP Requests 4-83 In Dest Unreachable attribute 4-77 In Echo attribute 4-77 In Echo Reply attribute 4-77 In Info Reply attribute 4-77 In Info Request attribute 4-77 In Parameter Problem attribute 4-77 In Redirect attribute 4-78 In Source Quench attribute 4-78 In Time Exceeded attribute 4-78 In Timestamp attribute 4-78 In Timestamp Reply attribute 4-79 In Use Count 4-101 INFO retrieving TCP/IP configuration information
K Index IPADDRESS attribute 4-17, 4-22, 4-25, 4-31, 4-40, 4-51, 4-52 IRDP attribute 4-22, 4-32 IRDP Protocol B-23 K Keep Alive Idle attribute 4-43 Keep Alive Interval attribute 4-43 Keep Alive Retry Cnt attribute 4-43 L LABEL command 4-115 Laddr attribute 4-58, 4-94 Last Time Used attribute 4-65 LAST-ACK socket state 4-94 LAST-ACK state 4-57 LISTEN socket state 4-94 LISTEN state 4-57 LISTNER process description 3-30 restart 3-42 starting 3-31 super server 3-30 waiting for incoming connection requests 3-
N Index MINFO See Mailbox Information MR See Mail Rename Name MTU 4-32 Multicast Groups attribute 4-53, 4-97 MX See Mail Exchanger N Name 4-91 Name attribute 4-40, 4-46, 4-50, 4-52, 4-65 Name Server record A-17 NAMED process files needed and their formats A-5 NAMES retrieving TCP/IP configuration information B-12 Names ADDRMAP 4-2 PROCESS 4-3 ROUTE 4-4 SUBNET 4-5 NAMES command ADDRMAP 4-58 ROUTE 4-59 SUBNET 4-60 NAMES PROCESS B-13 NETMASK attribute 4-19 Network name space A-3 order B-2 Network Informatio
P Index Object types (continued) null 4-3 PROCESS 4-3 ROUTE 4-4 SUBNET 4-5 Object-name templates, definition 4-6 Object-name templates, support for ABORT 4-12 DELETE 4-33 INFO 4-37 LISTOPENS 4-54 NAMES 4-58 PRIMARY 4-61 START 4-62 STATS 4-64 STATUS 4-89 STOP 4-102 OCTAL command 4-115 Opener attribute 4-56 Operator messages E-1 Order host B-2 network B-2 Out Dest Unreachable attribute 4-79 Out Echo attribute 4-79 Out Echo Reply attribute 4-79 Out Info Reply attribute 4-79 Out Info Request attribute 4-79 Ou
Q Index PROCESS, object type in (continued) STATUS command 4-92 STOP command 4-102 TRACE command 4-105 VERSION command 4-109 Program Filename attribute 4-44 Proto attribute 4-56 Protocol selecting a socket B-20 Protocol attribute 4-93 PROTOCOL file 3-39 PROTOCOLS file B-22 PS Text Edit 3-34 PSMAIL 3-42 PTR See Domain Name Pointer PTrace commands 4-112/4-118 PTrace overview 4-110 PVC ATM state flags 4-91 attribute 4-17 definition 2-7 ENTRY type 4-3, 4-16, 4-40 PVC Name 4-92 PVC Name attribute 4-40 PVCNAME
S Index Retrieving TCP/IP configuration information (continued) STATUS B-12 VERSION B-12 RFC See Request for Comment RFC Compliance for ATM 2-7 RFC1323-ENABLE attribute 4-30 ROUTE attribute 4-18, 4-35 ROUTE names 4-4 ROUTE object 4-1 ROUTE object type 4-4 Route records 4-137 Router Advertisement attribute 4-80 Router Discovery Protocol (ICMP) 4-22, 4-32 Router Solicitation attribute 4-80 Routes adding B-22 deleting B-22 transmitting datagrams B-22 ROUTE, object type in ABORT command 4-13 ADD command 4-18
S Index Sensitive commands (continued) STATS command with RESET option 4-64 STOP 4-102 TRACE 4-105 Server TCP/IP definition B-16 SERVER attribute 4-20 Service Access Point B-25 SERVICES file 3-39 Session Layer Socket interface B-13 Short IP Packets attribute 4-79 Simple Mail Transfer Protocol application layer B-27 description 3-41, B-27 gateway B-27 hosts B-27 Size 1025-2048 socket statistic 4-82 Size 12289-16384 socket statistic 4-82 Size 129-256 socket statistic 4-82 Size 16385-32768 socket statistic 4
S Index Standard Resource Record Format file A-14 START command 4-62 ROUTE 4-62 examples 4-63 SUBNET 4-63 examples 4-63 STARTED, multicast state 4-54, 4-97 Starting LISTNER process 3-31 TCP/IP process 3-28 TELSERV process 3-32 STARTING, multicast state 4-54, 4-98 Startup files for X25AM 3-15 State 4-53 State attribute 4-57 States defined 4-7 State, SERVER object 4-101 State, socket status 4-93 STATS retrieving TCP/IP configuration information B-12 STATS command ADDRMAP 4-64 display example 4-65 PROCESS 4-
T Index SUBNET, object type defined DELETE command 4-36 INFO command 4-48 NAMES command 4-60 START command 4-63 STATS command 4-86 SUBNET, object type in ABORT command 4-13 ADD command 4-21 ALTER command 4-31 STATUS command 4-101 STOP command 4-104 TRACE command 4-107 Subsystem Control Facility commands ADD 3-28 Subsystem Control Point Subsystem I/O process interface 2-1 Summary states 4-7 SUNAME 4-22 SUNAME attribute 4-22 SuName attribute 4-51, 4-53 SVC definition 2-7 ENTRY type 4-16 Switched Virtual Cir
U Index TELNET client B-20 TELSERV process, configuring 3-32 Templates for object names 4-6 TEXT command 4-118 TIME-WAIT socket state 4-94 TIME-WAIT state 4-57 TISERV process name 3-41 TMF 3-42 TN6530 Emulation Utility described B-20 TO attribute for TRACE 4-105 Token ring 2-3, 4-5 Total MBUFs Allocated attribute 4-82 TPCOMPAT42 attribute 4-29 TRACE command PROCESS 4-105 examples 4-107 SUBNET 4-107 examples 4-109 Trace Filename attribute 4-44, 4-51, 4-53 Trace record formats detailed UDP input records 4-1
V Index User Datagram Protocol (continued) packet integrity B-21 transport layer B-21 V VERSION retrieving TCP/IP configuration information B-12 VERSION command PROCESS detailed display example 4-110 display example 4-110 W Well-known port number See Port number Well-Known Services A-18 Wild-card characters 4-6 WKS See Well-Known Services WRAP attribute 4-106 $SYSTEM.ZTCPIP.HOSTS See HOSTS file $SYSTEM.ZTCPIP.MBOX See MBOX file $SYSTEM.ZTCPIP.PORTCONF See PORTCONF file $SYSTEM.ZTCPIP.