splk product guide - rxn communicationsuffda/archive/internet/sprtlink.doc · web viewtitle: splk...

SprintLinkâ

Customer Handbook 2.1

Released

October 11, 1995

Sprint Document #5953-2

2

Copyright © Sprint United Management Co. 1993-1995All Rights Reserved

3

Table of Contents

1. Introduction...........................................................................

2. SprintLink Description...........................................................2.1 Evolution of the Internet...................................................................................2.2 SprintLink Topology..........................................................................................2.3 Customer Routing Configuration.......................................................................2.4 Customer Premise Equipment (CPE).................................................................

2.4.1 Sprint Provided Routers...........................................................................2.4.2 Customer Provided Routers.....................................................................2.4.3 Sprint Provided CSU/DSU........................................................................2.4.4 Customer Provided CSU/DSU..................................................................2.4.5 CPE Examples..........................................................................................

2.5 Internet Registration Services...........................................................................2.6 SprintLink Value-Added Services......................................................................2.7 SprintLink Security Service..............................................................................2.8 SprintLink Pricing.............................................................................................

3. Internet Registrations............................................................3.1 Classless Interdomain Routing (CIDR) and IP Address Registration.................3.2 Domain Name...................................................................................................3.3 Autonomous System (AS) Number....................................................................3.4 Network Address Change Request (NACR).......................................................3.5 Domain Name Service (DNS)............................................................................

3.5.1 Primary DNS............................................................................................3.5.2 Secondary DNS........................................................................................

3.6 IN-ADDR.ARPA.................................................................................................3.7 Network News Feed..........................................................................................

4. Change Management.............................................................4.1 IP Address Requests..........................................................................................4.2 Routing Changes...............................................................................................4.3 DNS Changes....................................................................................................

4.3.1 Primary DNS............................................................................................4.3.2 Secondary DNS........................................................................................

4

4.4 Network News Feed..........................................................................................

5. Trouble Management.............................................................5.1 Trouble Reporting Procedure............................................................................5.2 Escalation Guidelines........................................................................................5.3 Status Guidelines..............................................................................................

6. Reseller Policy........................................................................6.1 Internet Service Providers (ISPs)......................................................................6.2 Customer Premise Equipment (CPE).................................................................6.3 Routing Configuration.......................................................................................6.4 Internet Registrations.......................................................................................6.5 Domain Name Service (DNS)............................................................................6.6 Trouble Management........................................................................................6.7 Commercial Internet Exchange (CIX)................................................................

7. Appendix - Sprint Templates.................................................7.1 IP Address Template (for ISPs).........................................................................7.2 IP Address Template (for End-Users)...............................................................7.3 IN-ADDR.ARPA Template..................................................................................7.4 IN-ADDR.ARPA Express Template.....................................................................

8. Appendix - InterNIC Templates.............................................8.1 Domain Name Registration...............................................................................

8.1.1 Registering A New Domain Name............................................................8.1.2 Modifying A Domain Name Record..........................................................8.1.3 Deleting A Domain Name Record.............................................................

8.2 AS Number Registration...................................................................................

9. Appendix - InterNIC Bulletin: Fee for Registration of Domain Names...........................................................................

10. Appendix - Glossary..............................................................

October 11. 1995 Page 1

SprintLink Customer Handbook 2.1 Sprint Business

1Introduction

This handbook contains the service description for Sprint’s SprintLink data communications service. This handbook also provides the information you need to successfully establish Internet connectivity using SprintLink services, submit inquiries and changes, and resolve troubles.

All Sprint tariffs apply to services provided and shall take precedence over any other material. All information, timeframes, charges, Sprint obligations and responsiblities are subject to change or modification by Sprint without prior notice.

The service and process descriptions contained in this Handbook apply to all orders submitted after March 27, 1995.

2

2SprintLink Description

Sprint's SprintLink service is a wide-area network (WAN) service for customers requiring Transmission Control Protocol/Internet Protocol (TCP/IP) networking, support for TCP/IP applications (i.e., Telnet, FTP and SMTP) and access to the global Internet. Sprint's commercial SprintLink service was conceived and planned after Sprint was awarded a cooperative agreement with the National Science Foundation (NSF) in January 1991. Through this agreement, Sprint was identified as the NSF International Connections Manager (ICM), responsible for implementing and operating a TCP/IP-based router network service that connects the domestic NSF network (NSFnet) with research and education networks overseas.

Initially, Sprint provided connectivity to research networks in Stockholm, Sweden, and Sophia Antipolles, France. Today, this network extends to London, Paris, Tokyo, China, Bonn, Malaysia, South Africa and the Middle East. Sprint remains the ICM in the aftermath of the NSFnet de-commissioning in April 1995.

Sprint launched SprintLink in July 1992, making Sprint the first carrier to offer a commercial Internet service. SprintLink is available nationwide from all of Sprint’s more than 320 points of presence (POPs). Sprint's network ubiquity keeps SprintLink access costs to a minimum. Furthermore, SprintLink pricing is not distance-sensitive like private line service. SprintLink is available at various circuit speeds from 56/64 kbps to T1. Speeds up to T3 (45 Mbps) are available on a case-by-case basis.

SprintLink was the first TCP/IP-based WAN service offered by an interexchange or long-distance carrier. As a carrier, Sprint is responsible for operating and maintaining both network routing and the physical network that makes Sprint the "one-stop" shop for network connectivity. Most other public TCP/IP network service providers rely on a long-distance network provider for their physical network.

In an effort to reduce access costs, Sprint integrates access to SprintLink on T1 facilities used to access other Sprint services like WATS or 800 service. In these instances, access costs are shared between SprintLink and the other Sprint data and voice services.

2.1Evolution of the InternetThe potential for internetworking among organizations was first realized in the government and education community. The National Science Foundation built

3

a network backbone to connect existing government and education networks. The NSF backbone network became known as the NSFnet. The collection of networks interconnected by the NSFnet backbone quickly became known as the Internet. The original Internet was a research tool governed by an Acceptable Use Policy (AUP) that stated the network was not to be used for commercial purposes. Since its inception, the tremendous growth and potential of the Internet presented an opportunity to commercial and consumer sectors. The migration of commercial traffic onto the Internet removed the need and justification for government subsidizing.

As a result of this diminished need for government funding, the NSFnet was decommissioned in April 1995. The AUP that once governed Internet usage was no longer appropriate. The NSFnet shutdown recognizes the new Internet landscape while positioning the Internet to evolve and touch greater portions of the population. The increasingly commercial infrastructure allows the Internet to grow to new levels of utility and connectivity through competing, for-profit Internet Service Providers (ISPs).

The Internet is a collection of interconnected networks based on Transmission Control Protocol / Internet Protocol (TCP/IP). While the ISPs that provide these networks certainly compete, the cooperative relationship between ISPs is essential to the continued success of the Internet.

The rapid evolution of the Internet from a research and education experiment to a nationwide information superhighway requires elegant design and world-class technology from ISPs. Sprint is the current leader in Internet architecture and technology and positioned to lead the Internet into the twenty first century.

2.2SprintLink TopologyThe SprintLink network has been engineered with functionality, availability, reliability, serviceability, scalability and performance as the drivers of its design. The overall design objective included creating a highly connected network with the goal of offering customers the optimal path to anywhere on the Internet.

The topology is two tiered with an access layer for connecting dedicated customer links and a backbone layer for interconnecting SprintLink node sites. Fiber Distributed Data Interface (FDDI) LANs are used to connect all routers at SprintLink node sites. All routers are dual homed to two independent FDDI hubs, creating a highly available central office design [see the network diagram on the next page]. Backbone routers are independent of customer terminations with the exceptions of customer access DS3 links. The network design includes a two-tiered logical architecture to match the physical design. SprintLink maintains a backbone autonomous system (AS) number of 1239 while regional autonomous systems are used within SprintLink access sites. Within the

4

SprintLink network, the backbone AS and each regional AS operate as if they were the same AS to preserve next-hop and routing metrics.

The backbone links are predominantly DS3 transmission facilities with physical-path diversity built into the topology design. Each SprintLink site has at least two physical backbone paths to ensure survivability in the unlikely event of a brief outage. Each SprintLink site has at least two backbone routers deployed to complement the physical DS3 link diversity. SprintLink routers are deployed in environmentally protected Sprint central office switch sites to provide high circuit and equipment availability.

5

SprintLink Backbone Node Site

7

Interconnect points are gateways that connect networks like SprintLink to other National Service Providers (NSPs). More interconnect points means better performance and improved availability. SprintLink continues to maintain as many Internet Service Provider (ISP) interconnect points as reasonably possible. This network of internetworked NSPs through high-speed interconnect points forms the backbone of the Internet.

SprintLink Interconnect Points and Speeds

SprintLink Interconnect point

Interconnect speed

CIX T1 (1.5 Mbps)

FIX-East DS3 (45 Mbps)

FIX-West DS3 (45 Mbps)

MAE-East 10 Mbps

MAE-East+ DS3 (45 Mbps)

New York Network Access Point (NAP)

FDDI (100 Mbps)

8

Use Word 6.0c or later to

view Macintosh picture.

9

In addition to being the ICM, Sprint’s role as an industry leader was affirmed by our selection as manager of the New York Network Access Point (NAP). SprintLink is also fully operational at the Chicago NAP (managed by Ameritech) and the Stockton NAP (managed by PACBell). Additional interconnect points are planned as traffic requirements and service provider consensus are achieved.

SprintLink's overall design goals include minimizing delay via the shortest path and minimizing the number of router hops. With the current network design, customer's traffic will traverse an average of only two SprintLink nodes sites before reaching potential interconnect sites. Average network delay depends largely on the source and destination address location with terrestrial delay and the size of the access channel likely to be the most significant contributor.

A note on network delay:Delay in the Internet is a complex topic that can be affected by numerous events, many of which are not directly associated with network services. A traffic path may traverse multiple network providers who may have different levels of bandwidth and congestion.

The "ping" utility is often used to measure network delay with misleading results. Ping (simply an ICMP echo request) is a low-priority process in most all IP devices. If a device has other IP data to send and receive, the delayed ping request reports exaggerated round trip delay. This is why many times pings have typical statistical delay performance but may encounter some delays much larger than the statistical average.

Complexities such as these make it difficult to measure network delay. Recognizing customer desire for such measures, Sprint is cautiously investigating options for providing network performance results to SprintLink customers. Product management and sales support will notify the field when such measurements are available.

2.3Customer Routing ConfigurationThere are two general routing configurations for SprintLink customers: static routing and dynamic routing.

Those customers who have a single network connection to the Internet are considered single-homed and are configured with static routes. Statically routed customers may order one of SprintLink's pre-defined router packages or provide their own SprintLink-certified router.

Multi-homed customers, those with two or more diverse connections to SprintLink or another ISP, should use an exterior routing protocol (preferably BGP-4) to announce their routes to SprintLink and other ISPs.

10

The pre-defined router packages described below may not satisfy multi-homed customers' requirements.

2.4Customer Premise Equipment (CPE)As part of the SprintLink service, additional equipment is required at the customer site. A router is required to perform routing functions to and from the SprintLink network. A Channel Service Unit/Data Service Unit (CSU/DSU) or channel bank is also required to terminate the private line at the customer site. Customers may obtain this CPE from Sprint or provide their own CPE in accordance with the guidelines given below.

Sprint provided CPE is a natural extension of the SprintLink service. SprintLink CPE includes transport CPE (i.e., CSU/DSUs and channel banks) and bundled router packages. Bundled router packages are available for non-ISP customers only. The bundled router packages include the router software and hardware and a dial-in modem for remote configuration and management. SprintLink CPE is available on a rental basis only with terms of one, two, and three years. The month-to-month rental option is no longer available.

Sprint and customer responsibilities for the SprintLink service and its associated CPE vary according to which party is providing the various CPE components. Specific scenarios are described later in this section; however, general guidelines are provided below:

∑ Sprint is responsible for managing and maintaining all Sprint provided CPE

∑ The customer is responsible for managing and maintaining all customer provided CPE

∑ Sprint is responsible for providing 24 x 7 in band proactive monitoring of the customer’s access link; however, the demarcation point to which Sprint monitors varies according to which party is providing the various CPE components

∑ Customers will be charged a “dispatch charge” for services rendered by Sprint CPE technicians in support of any customer provided CPE

∑ Customer will be contacted before any planned service interruption outside of routine maintenance

∑ Sprint Management Center (SMC) will initiate SMC SLA during unplanned service outages

2.4.1Sprint Provided Routers

Non-ISP (i.e., reseller) customers may obtain a bundled router package from Sprint only if:

∑ they are statically routed∑ they are not using Customer Provided Access (CPA)

11

∑ they have a dedicated business line (1 MB or better) installed at the site∑ Sprint manages and maintains the customer’s service including the

router

For additional information on CPE for ISPs, please see the “Reseller Policy” section later in this document.

All maintenance for routers provided by Sprint is on a next day basis only. The Sprint CPE Technician Group's Service Level Agreement (SLA) for next day maintenance requires that a technician be on-site at the customer's location no later than 5 PM the next day after receiving the trouble ticket. Please note that on-site by 5 PM the day after receiving the trouble ticket is a worst case scenario. The objective of the CPE Technician Group is to correct the customer's trouble as soon as possible providing they have the necessary equipment available and access to the customer's premise.

Sprint currently offers two bundled router packages. These packages vary only by the type of LAN protocol they support (Ethernet vs. token ring). Both packages include a dial-in modem to allow remote configuration and fault management of the router by Sprint personnel. The router packages are:

SprintLink CPE Router Packages:

Sprint Package: SPC014ND Sprint Package: SPC024ND

Cisco 2501 (4 Mb flash memory) Cisco 2502 (4 Mb flash memory)

Supports Ethernet LAN connection Supports token ring LAN connection

Cisco 2500 IP router software set Cisco 2500 IP router software set

10-foot male DTE V.35 cable 10-foot male DTE V.35 cable

14.4 kbps modem and modem cable for remote management

14.4 kbps modem and modem cable for remote management

Next-day maintenance Next-day maintenance

Router configuration and management

Router configuration and management

2.4.2Customer Provided Routers

Customers may provide their own SprintLink router only if the router:

∑ is a Cisco router running Cisco revision level 10.2 or greater

12

∑ has at least 4MB flash memory for customers with static routing, or 8MB flash memory for customers with dynamic routing

∑ is configured, maintained, and managed by the customer

Customer provided routers can be single-homed or multi-homed. Single-homed customers have one path to the Internet from a single customer IP address. Multi-homed customers have multiple paths to the Internet from a single customer IP address. Multi-homed customer routers must also meet the following requirements:

∑ are Cisco 7000 or 7010 models∑ run BGP-4

When providing the router, the customer will also be responsible for furnishing the necessary ancillary equipment (cables, routing software, etc.) to insure interoperability with the SprintLink network router. Sprint is not obligated to accept trouble calls related to the customer provided router. In the event a customer requests assistance from Sprint to correct a trouble relating to the customer provided router, Sprint will inform the customer that a "dispatch charge" will be applied to the customer's next bill for the services provided by the Sprint CPE Technician.

Sprint is in the process of certifying Bay Networks (formerly Wellfleet) and 3COM routers for use with SprintLink services. Once certified, customers desiring to use either of these routers may do so without an Special Customer Arrangement (SCA) if the router:

∑ meets SprintLink configuration guidelines∑ is managed and maintained by the customer∑ is statically routed

2.4.3Sprint Provided CSU/DSU

CSU/DSUs terminate SprintLink access lines at the customer site. All customers may obtain their CSU/DSU equipment from Sprint. Customers using a Sprint provided router package must use a Sprint provided CSU/DSU.

Maintenance for CSU/DSU's provided by Sprint can be either same day or next day. The Sprint CPE Technician Group's SLA for same day maintenance is as follows:

∑ Sprint CPE Technician to arrive at the customer's location within 4 hours after the Sprint CPE Technician Group receives the trouble ticket.

∑ The 4 hour on-site SLA only applies if the customer location is within a 100 mile radius of a Sprint CPE Technician site.

∑ Any customer site outside a 100 mile radius will receive service as soon as possible.

13

The CSU/DSU packages Sprint offers are:

SprintLink CSU/DSU Packages:

Unitsupported

Notes Speedssupported

Sprint package number

CSU/DSU Stand-alone 232 V.35 56 kbps SP1510-35A

Stand-alone 232 V.35 56 kbps SP1530

Single-port CSU 64-kbps - T1 SP3110

Two-port CSU 64 kbps- T1 SP3120-V35

Two-port CSU 64 kbps- T1 SP3120-TSU

Four-port CSU 64 kbps - T1 SP3140

Channel banks Unintelligent, interfaces with analog PBX

56 kbps- T1 SP3210

Intelligent, interfaces with analog PBX

56 kbps- T1 SP3220

Unintelligent, interfaces with digital PBX

56 kbps- T1 SP3230NDDI

Intelligent, interfaces with digital PBX

56 kbps- T1 SP3230IDDI

Channel bank cards

Used with SP3230 NDDI or SP3210

56 kbps- T1 DECUVV02AC

Used with SP3230 IDDI or SP3220

56 kbps- T1 DEPIG8E3AC

2.4.4Customer Provided CSU/DSU

Customers who are providing and managing their own routers may also furnish their own Sprint certified CSU/DSU. To obtain a listing of all the Sprint

14

certified CSU/DSU's, please refer to the Sprint CPE Product Engineering Catalog of Customer Premises Equipment. Customers obtaining a router package from Sprint must obtain the CSU/DSU from Sprint as well. Customers using CPA must obtain, manage, and maintain their own CSU/DSU.

2.4.5CPE Examples

Scenario 1: Sprint provides the router, CSU/DSU, and access

Since Sprint is providing all of the CPE in this scenario, Sprint must manage, monitor, and maintain the router, modem, and CSU/DSU without exception. At this time, Sprint will only provide the CPE that is listed in the tables above.

Sprint will require all customers choosing this scenario to provide a business line for emergency monitoring of the CPE router. In the event that the dedicated customer link goes down, Sprint will have the ability to access the customer's router via the business line.

As mentioned earlier, Sprint will be responsible for managing the customer's access link and CPE from the SprintLink network to the LAN port on the CPE router. The LAN port on the CPE router is the demarcation point for this scenario.

Scenario 2: Customer provides the router, Sprint provides the CSU/DSU and access

Sprint will not be responsible for managing or maintaining any customer provided router. The customer must be sure to meet the router requirements set forth above. Multi-homed customers have additional requirements that must be addressed. The customer must also provide ancillary equipment (cables, routing software, etc.) and will pay “dispatch charges” for services rendered by Sprint CPE technicians in support of the customer provided router. The CSU/DSU and access responsibilities are the same as for scenario 1.

Sprint will be responsible for managing the customer access link up to the V.35 interface on the CSU/DSU. If the Sprint PLSC is able to perform loop back testing to the demarcation point, the V.35 interface on the CSU/DSU, then the trouble is classified as "Customer CPE" and becomes the customer's responsibility to correct.

Scenario 3: Customer provided router and CSU/DSU, Sprint provided access

In this scenario, the customer’s responsibility for the router remains the same as in scenario 2. Now, however, the customer is responsible for providing a Sprint certified CSU/DSU. To obtain a listing of all the

15

Sprint certified CSU/DSU's, please refer to the Sprint CPE Product Engineering Catalog of Customer Premises Equipment.

The customer will need to select one Sprint certified CSU/DSU or channel bank for each access line provided by Sprint. Sprint will not be responsible for managing, monitoring or maintaining the customer provided CSU/DSU. In the event a customer requests assistance from Sprint to correct a trouble relating to the customer provided CSU/DSU, Sprint will inform the customer that a "dispatch charge" will be applied to the customer's next bill for the services provided by the Sprint CPE Technician.

Sprint will be responsible for managing and monitoring the customer access link up to one of the following demarcation points: a LEC provided Smartjack or the LEC Central Office. If the Sprint PLSC is able to perform loop back testing out to these demarcation points, the trouble is classified as "Customer CPE" and becomes the customer's responsibility to correct.

Scenario 4: Customer provided router, CSU/DSU, and access

In this scenario, the customer has elected to provide their own local access. Therefore, the customer must also provide, manage, and maintain both their own router and CSU/DSU.

Sprint is responsible for managing the customer’s SprintLink connection up to the point of interconnection (i.e. the servicing wire center or the Sprint POP) with the means of local access chosen by the customer.

Scenario 5: Sprint Frame Relay customers connecting to SprintLink via the Frame Relay Gateway

Existing Sprint Frame Relay customers wanting a connection to the SprintLink network can do so via the frame relay gateway. Frame relay customers can connect to the SprintLink network using any of the following Cisco Systems routers that have been certified by Sprint to connect to the Sprint Frame Relay network. Multi-homed customers will not be allowed to connect to SprintLink via the frame relay gateway.

If a frame relay certified Cisco router requires an upgrade in order to connect to the SprintLink network, the customer should order the necessary equipment (i.e. hardware, software, etc.) via the Sprint Frame Relay order process.

All customer frame relay CPE troubles should be reported to Sprint as before and according to the frame relay trouble reporting policy.

16

2.5Internet Registration ServicesSprint offers the following Internet registration services:

∑ IP address registration: Sprint assigns customers Class C IP addresses from its Classless Inter-Domain Routing (CIDR) address block in accordance with Internet Activities Board (IAB) guidance. Generally, Sprint requires that 80% of the requested address space be utilized immediately to encourage its efficient use. Once 90% of the requested addresses are exhausted a customer can request additional Class C IP addresses by having their account team submit an order for additional addresses and by filling out the IP Address Request templates included in the Sprint Templates appendix.

∑ Domain name registration: Sprint strongly recommends that customers register a single top-level domain (e.g., xyz.com) since InterNIC policies discourage registration of multiple top-level domains. Sprint registers a single domain for customers during SprintLink implementation at no charge. Sprint will not process new domain name requests if a customer has an existing registered domain name.

∑ Customers select the name they wish to have registered and provide this to Sprint when SprintLink is ordered. The top three selections, in order of preference, for each domain name must be provided at the time of service ordering. Should the first selection already be in use, Sprint will attempt to register the second selection and then the third selection.

∑Note: There may be a delay of up to 30 days from the time the order is placed to the day a domain name is registered. SprintLink Implementations will call the customer to confirm the domain names chosen and obtain additional information required to register that domain. The InterNIC processes domain name requests on a first-come, first-serve basis. Sprint cannot assume responsibility for and has no control over domain names already in use. Customers may submit their own domain name request directly to the InterNIC. See the Appendix InterNIC Templates for directions and a sample template.

2.6SprintLink Value-Added ServicesThe following value-added services are optional and may carry a monthly recurring charge as outlined in SprintLink pricing:

∑ Domain Name Services: DNS is a distributed hierarchical system for resolving host names (e.g. mnemonic names) into IP addresses (e.g., numeric addresses). Each enterprise that registers an Internet domain name must provide multiple name servers for the domain: a primary server and at least one secondary server for backup purposes. Ideally,

17

the primary and secondary name servers should run on independent and redundant systems so that availability of name service for the domain isn't affected by a single point of failure.

∑ - Primary Domain Name Service (DNS): Sprint provides primary DNS for one or more top-level or subdomains for all hosts on one Class C network address or up to 64 hosts across multiple, Sprint-allocated Class C network addresses.

∑ - Secondary DNS: Sprint can provide secondary DNS for up to 10 domains (or 20 zones assuming one forward and reverse lookup zone per domain).

∑ Customers can get either primary DNS, secondary DNS, or both from Sprint. Please see the Reseller Policy section of this guide for DNS restrictions that apply to ISP customers.

∑ Network Time Protocol (NTP): The NTP is a protocol for synchronizing a set of network clocks using a set of distributed clients and servers. NTP provides the protocol mechanisms to synchronize time in principle to precision on the order of nanoseconds while preserving a non-ambiguous date.

∑ SprintLink provides NTP services to customer routers via SprintLink access routers. If customers would like to get time from SprintLink via NTP, they can do so at any time by pointing their NTPD to their interface on the SprintLink access router. If they don't know what this address is, they can use traceroute to determine the first hop after they leave their local customer premise router. The second hop is the appropriate Sprint access router. Questions or comments about SprintLink NTP Services should be sent via e-mail to [email protected].

∑ Network News feed: Often referred to as Usenet (which is actually a subsection of Network News), Network News is the Internet equivalent to a bulletin board system. Network News is organized into a hierarchy of over 10,000 newsgroups beginning with names such as “alt” for alternative, “comp” for computer science, and “rec” for recreation. Sprint provides an optional full or partial news feed to SprintLink customers.

∑ Customers may also receive Clarinet news feeds from Sprint included in their news service after they order the service directly from Clarinet.

∑ A note on Network News feeds:∑ Full news feeds currently exceed 450 Mb per day. For that

reason, customers desiring a full news feed are encouraged to order 256 kbps or greater SprintLink connections.

∑ Access Management Services:∑ Access lists: Optionally, Sprint will configure access lists on all customer

premise routers managed by Sprint. Access lists can be used to filter unwanted traffic from entering or leaving a customer's network. Access lists provide basic levels of protection.

18

2.7SprintLink Security ServiceIn response to concerns about unauthorized computer access of both public and private networks, Sprint lobbied Congress for passage of computer crime legislation during the mid-1980s. These efforts were instrumental in the passage of the Computer Security Fraud and Abuse Act of 1987. As businesses expand globally, the transport of data over networks becomes more and more widespread. Ironically, the very technology that lets information be shared by hundreds or thousands of employees also provides pathways through which unauthorized users can enter unprotected computer systems.

SprintLink’s Security Service helps to ensure the confidentiality, integrity, and availability of customer data. SprintLink customers receive, at no additional charge, ongoing assistance that minimizes the vulnerability of their network.

The SprintLink Security Service consists of the following:

∑ Internet security presentations. Sprint's data network security staff, experienced in the technical and investigative aspects of the Internet, are available to discuss security issues regarding connections to the Internet.

∑ 24-Hour investigative support. Security support is available 24 hours a day, 7 days a week, to address reported security breaches and other security concerns.

∑ Liaison, law enforcement, and prosecution assistance. Sprint will serve as a liaison between customers and other data network providers, when appropriate. At the customer’s request, Sprint will also provide assistance in working with legal authorities around the world.

∑ 24-hour dedicated security support. One of Sprint's experienced security support managers will be assigned to assist with security issues. Customers can also call Sprint's corporate security department at 703-689-5489 for assistance at any time, including weekends and holidays.

2.8SprintLink PricingSprintLink pricing is easy to understand and straightforward. Components of the pricing include access, port charge and CPE. Each of these components have both installation and monthly recurring fees associated with them.

∑ Access: The access fees are detailed in Sprint's Tariff 8 pricing. Appropriate access charges are determined by the type of access required between the customer site and the Sprint DCS network.

∑ SprintLink Port: The port charge depends upon the speed desired and has both a flat monthly fee and a one-time installation charge. Effective March 27, 1995, SprintLink requires a minimum one-year term commitment on all orders.

∑

19

SprintLink Port Charges

Speed/Feature Install Charge

MRC w/ One-year term

MRC w/ Two-year term

MRC w/ Three-year term

56/64 kbps $750 $960 $940 $920

112/128 kbps 1,000 1,248 1,222 1,196

224/256 kbps 1,000 1,440 1,410 1,380

336/384 kbps 1,000 1,632 1,598 1,564

448/512 kbps 1,000 1,920 1,880 1,840

560/640 kbps 1,000 2,208 2,162 2,116

672/768 kbps 1,000 2,400 2,350 2,300

T1 (1.544 Mbps)

1,000 2,592 2,538 2,484

Upgrade charge

250 NC NC NC

- NC = no charge- Minimum one-year term commitment

∑ SprintLink CPE: SprintLink CPE charges include equipment rental, maintenance, and management. CPE may be rented on a one, two, or three year term. SprintLink CPE may now be discounted in accordance with standard CPE discounting as documented in the CPE product catalog. Leasing of SprintLink CPE is not currently available.

20

SprintLink CPE Packages

Sprint Pkg. Number

Install Charge

MRC w/ One-year

term

MRC w/ Two-yea r

term

MRC w/ Three-year

term

CSU/DSU

SP1510-35A $110 $35 $28 $26

SP1530 110 36 28 25

SP3110 165 113 94 81

SP3120-V35 220 148 122 110

SP3120-TSU 220 148 123 110

SP3140 220 176 149 126

Channel Bank Cards

DECUVV02AC NC 40 33 28

DEPIG8E3AC NC 54 44 39

Channel Bank

SP3210 550 162 135 116

SP3220 550 209 174 150

SP3230NDDI 550 160 132 113

SP3230IDDI 550 226 187 163

Customer Router Packages

SPC014ND 700 365 270 229

SPC024ND 700 425 315 265

21

- NC = no charge- Minimum one year term.- Monthly recurring prices based on term commitments (one, two, and three year).

Further information on discounted prices can be found in the Sprint CPE Product Catalog.

- Rental prices include next-day maintenance for CPE routers and same-day maintenance for CSU/DSUs.

∑ Value-Added SprintLink services: all value-added services require a valid SprintLink order even when there is no charge.

Value-Added Service

Install Charge

MRC w/ One-year

term

MRC w/ Two-year

term

MRC w/ Three-year

term

Primary DNS NC $100 $100 $100

Secondary DNS NC NC NC NC

Network News feed NC 50 50 50

Additional top-level domain names

50 NC NC NC

Access lists NC NC NC NC

- NC = no charge

22

3Internet Registrations

The Internet is one of the most complex and useful data communications systems in the world. Its success is due to its decentralized nature and the universal connectivity it provides. However, there are several requisite coordinating functions. As such, the U.S. government funds the InterNIC that coordinates Internet address and name space management as well as maintenance of Internet data.

3.1Classless Interdomain Routing (CIDR) and IP Address Registration

As a result of more and more networks connecting to the Internet, the size of the routing tables maintained in large ISP backbone routers have grown dramatically. In 1993, many ISP routers reached their capacity in terms of available memory. This is because the backbone routers kept a route to each individual network connected to the Internet.

CIDR is a technology that allows routers to group routes together (called route aggregation) to reduce the number of route table entries in the backbone routers. Under CIDR, large ISPs are assigned blocks of contiguous Class C network numbers by the InterNIC to, in turn, assign to their customers. Here is an example:

∑ Suppose Sprint’s SprintLink service has been assigned a block of Class C addresses, specifically 201.12.0.0 through 201.12.255.0 including 256 contiguous Class C network addresses.

∑ Company XYZ buys its Internet connection from Sprint. Under CIDR, Company XYZ is assigned network addresses by Sprint, instead of getting them directly from the InterNIC.

∑ Sprint assigns Company XYZ two Class C network numbers from its CIDR block -- 201.12.6.0 and 201.12.7.0.

Policies regarding "ownership" of the IP address are still being developed by the Internet community. At odds are the users' need to avoid the cost of changing IP addresses when they change service providers versus the service providers need to keep their CIDR blocks in tact. Allowing customers to take their IP address with them if they leave Sprint undermines the whole purpose of using CIDR blocks. The issue of IP address portability - much like 800 number portability - will most likely be resolved by new technological advances. In the meantime, Sprint's policy is to retain the IP address within our CIDR block if a customer transfers service to another provider. However, customers who are

23

able to retain IP addresses assigned by another carrier may use those addresses with SprintLink.

Before CIDR, other large ISPs' routers had to track individual routes for both of Company XYZ's networks (e.g., 201.12.6.0 and 201.12.7.0) as well as for networks belonging to any other SprintLink customers. In this example, routers outside of SprintLink would have kept as many as 256 routing table entries or one for each individual address between 201.12.0.0 and 201.12.255.0.

Under CIDR, these individual routes are aggregated into an address RANGE maintained by other providers' backbone routers as a SINGLE route table entry. For example, backbone routers know that all addresses in the range of 201.12.x.y belong to SprintLink customers and that IP datagrams destined for those addresses should be sent to Sprint. Sprint, in turn, keeps track of individual routes to network numbers within its own CIDR block and is responsible for delivering datagrams to the correct SprintLink customer.

Sprint assigns customers Class C IP addresses from its CIDR address block as needed. Generally, Sprint requires that 80% of the requested address space be utilized immediately. This is to encourage the efficient use of the IP address space. Please be prepared to supply customers' current IP address(es) and requests for additional IP address space during the SprintLink order process.

3.2Domain NameThe InterNIC announced that beginning 9/13/95, they would begin charging for all registration and maintenance of Domain names. All new registrations would incur a $100.00, two year “Registration Fee” followed by a “Maintenance Fee” of $50.00 that would be charged each year beginning year 3.

Sprint intends to continue to offer its Customers the option of having Sprint register their Domain names with the InterNIC. However, the Customer will be responsible for ensuring that any fees associated with registration are paid directly to the InterNIC within 30 days of Domain name activation. The InterNIC Registrar will batch process and electronically send invoices each night. These invoices will include domain names registered the preceding day. Should the Customer fail to pay the registration fee, the Registrar will delete the domain name from the database on the day after the payment due date.

For existing Domain names, the InterNIC Registrar will provide 60 days advance notice by Email, to the domain name billing contact or the administrative contact for renewal of the Maintenance Fee on the anniversary date. The Customer will be responsible for payment of this fee by 12:00 PM(Eastern Time) on the anniversary of the activation date or the last workday preceding the anniversary date if the anniversary date falls on a weekend or holiday. Should the Customer fail to pay the registration fee, the Registrar will delete the domain name from the database on the day after the payment due date.

24

Domain Names that are deleted from the database will be placed in a 60 day “hold”. After the 60 day period the Domain name will become available for reuse by the InterNIC.

Should users require any additional information on the billing and payment of these fees please see the Appendix - InterNIC Bulletin, Fee for Registration of Domain Names. Additionally, questions regarding the policy may be sent directly to [email protected].

Customers who want Sprint to register their domain names are required to provide their top three selections for each domain name requested at the time of service ordering. Should the first selection already be in use, Sprint attempts to register the second selection and so on. Please be prepared to supply customers' current root domain names and requests for additional domain names during the SprintLink order process.

3.3Autonomous System (AS) NumberAn AS number is only necessary when the customer wishes to "peer" with SprintLink. For example, customers configured with an exterior routing protocol (e.g., BGP-4), in contrast to static routing, are considered "peers." Customers who require their own AS number should contact the InterNIC directly. Templates are available from ftp://rs.internic.net in the templates directory.

3.4Network Address Change Request (NACR)Because of the decommissioning of the NSFnet, NACR is no longer required for SprintLink routing.

3.5Domain Name Service (DNS)SprintLink customers may use SprintLink for primary DNS, secondary DNS, or both. Customers who do not need either must provide information about their domain name service to SprintLink Implementations to complete their Internet registrations.

3.5.1Primary DNS

Primary DNS is offered to non-ISP customers according to the guidelines below:

∑ Primary DNS is available for top-level or subdomains.∑ Primary DNS is available for all hosts on one Class C network address

or up to 64 hosts across multiple, Sprint allocated Class C network addresses.

25

∑ Customers must run a local SMTP gateway (sendmail server).∑ New SprintLink customers with an existing connection to another

service provider must inform the InterNIC or have their current service provider inform the InterNIC of any change to their existing Primary DNS. This includes the addition of the new SprintLink IP address. The InterNIC will only accept changes to the current Primary DNS from either the customer or the customer's current provider.

3.5.2Secondary DNS

Most customers obtain Secondary DNS from Sprint even if they provide their own Primary DNS. This service allows them to automatically have disparate name servers without any additional cost. Sprint will provide Secondary DNS to all customers upon request via the name servers specified in the table below.

SprintLink's Secondary DNS servers

Host name IP Address

ns1.sprintlink.net 204.117.214.10



3.6IN-ADDR.ARPAAddress-to-name mapping is often used as an authorization check and to produce output (in log files, for example) that's easier to read. The in-addr.arpa names space is used for this IP address-to-domain name mapping. SprintLink Implementation personnel coordinate in-addr.arpa registration with the customer during the implementation process.

3.7Network News FeedSprintLink can provide a full or partial news feed to all SprintLink customers. SprintLink will not provide Network News to customers of customers.

Customers must send an e-mail to [email protected] after their account team has submitted a valid SprintLink order for news. Sending the e-mail will trigger an auto-response that includes instructions on how to subscribe to either a full or partial news feed. Customers desiring a Clarinet news feed may integrate this feed with a standard news feed after ordering the Clarinet news

26

service directly from Clarinet.

27

4Change Management

Change management defines the procedures involved in modifying or changing a customer's SprintLink configuration. Included among these changes are:

∑ Requesting additional IP address space∑ Requesting additional root domains∑ Submitting routing changes∑ Submitting and requesting changes to DNS∑ Requesting or modifying a Network News feed.

4.1IP Address RequestsSprintLink assigns customers additional Class C IP addresses from its CIDR address block as needed. Generally, Sprint requires that 80% of the requested address space be utilized immediately. This is to encourage the efficient use of the IP address space. Before additional IP addresses can be requested, 90% of the current address space must be exhausted. To request additional IP addresses, send an e-mail message to [email protected].

Sprint initially forwards assignment information to InterNIC so it is not necessary for customers to also send assignment information. However, if ISPs reassign those IP addresses to their customers, the ISP is responsible for reassignment as noted below.

Note:Subsequent allocations of IP block addresses will be based on need; this need will be demonstrated based on the number of reassignment actions that have been transmitted to the InterNIC Registry. Reassignment information is to be forwarded to the InterNIC within 7 days of the assignment so that the WHOIS may be maintained efficiently. Transmission of reassignment information is also necessary for the following reasons:

a) To ensure that a provider has exhausted, or is about to exhaust, its current IP Block allocation such that an additional allocation is justified.

b) To allow operational people to see which organization is using the network and who to contact in the event of operational problems, security issues, etc.

c) To assist in IP allocation studies.

28

Reassignment actions can be submitted by utilizing the database exchange format defined by the Shared WHOIS Project (SWIP) project. Information regarding SWIP may be obtained via anonymous FTP from rs.internic.net (198.41.0.5). The files may be found in the ftp/pub/swip directory.

All ISPs should SWIP the reassignment information. ISPs should get a maintainer ID necessary to SWIP the reassignment information. This can be done by sending an e-mail request to [email protected].

ISPs are required to assign address space based on utilization efficiency. To this end, ISPs should have documented justification available for each assignment. SprintLink may, at any time, ask to see this justification. Future allocations could be impacted if justification is not available.

4.2Routing ChangesTo request routing changes, send an e-mail message to [email protected].

4.3DNS Changes

4.3.1Primary DNS

All DNS changes should be sent via e-mail to [email protected]. The change interval is three business days.

4.3.2Secondary DNS

Any changes in the status of the primary DNS provider must immediately be sent via e-mail to [email protected]. Normal response time is within three business days.

4.4Network News FeedChanges to a customer’s network news feed can be submitted directly by the customer via an e-mail message to [email protected]. The change interval is three business days.

29

5Trouble Management

If customers experience a service problem, they should report it to Sprint’s Service Management Center (SMC) as soon as possible. Customer support representatives isolate problems and provide quick, efficient resolutions.

5.1Trouble Reporting ProcedureTo report service issues, 24 hours a day, 365 days a year, customers should follow these steps:

1. Contact the SMC at 800-877-5045 or 404-859-7600. Follow the recorded instructions to reach the customer support team.

2. Be prepared to provide the following information to the support team representative when reporting a trouble.

∑ Company name∑ Customer contact name and callback number∑ Contact hours∑ 449 network address∑ IP address∑ Private line number∑ Domain name∑ Problem description∑ Additional information, as requested∑

1. When customers call, their SMC representative will assign a trouble report number. Use of this number will be helpful when communicating with the service management representative. Service management representatives keep customers appraised of the status of their problem and notify them when it is resolved. The SMC’s charter is to facilitate trouble ticket resolution and be the customer’s advocate.

5.2Escalation GuidelinesThe following escalation procedure and timetables are followed when managing SprintLink trouble reports.

30

Critical Condition:

Ticket DurationEscalationLevel Notification Contact

Initial outage 1 The SMC escalates immediately to fix agency lead

1 hour 2 The SMC escalates to fix agency supervisor

2 hours 3 The SMC escalates to fix agency manager

3 hours 4 The SMC escalates to fix agency director

4 hours 5 The SMC escalates to fix agency vice president

Out of Service:


1 hour 1 The SMC escalates to fix agency lead

2 hours 2 The SMC escalates to fix agency supervisor




31

Service Impairment:


2 hours 1 The SMC escalates to fix agency lead





Minimal Impact or RFO:


4 hours 1 The SMC escalates to fix agency lead





Non-Service Affecting Conditions: Escalations as necessary every 24 hours.

5.3Status GuidelinesThe following status guidelines are followed when managing SprintLink trouble reports.

32

Ticket Status Guidelines

Ticket Severity Status Guideline

Critical condition Initial status within 30 minutes of ticket entry, then every hour thereafter

Service impairment Initial status within one hour of ticket entry, then every hour thereafter as necessary

Minimal impact or RFO

Status is provided every two hours as necessary

Non-service affecting conditions

Minimum of once a day or as necessary. The status of these tickets can vary greatly depending on the customers needs

33

6Reseller Policy

6.1Internet Service Providers (ISPs)An ISP or reseller is defined as a SprintLink customer who in turn sells Internet access to businesses and/or individuals by assigning IP addresses. This includes SprintLink customers offering dedicated Internet connections or SLIP/PPP dial-up connections to others, as these require assigning an IP network address, subnet address or host address to the end user. This definition is consistent with the CIX’s definition of ISPs or resellers.

Sprint expects all ISPs to be Internet-knowledgeable and capable of independently configuring TCP/IP communications, applications and routing software in their local network environments.

All standard SprintLink access speeds are available to ISPs. Pricing is based on published SprintLink list rates.

6.2Customer Premise Equipment (CPE)The equipment required for a SprintLink connection consists of an IP router and a CSU/DSU. This equipment resides on the ISP's premise and is commonly referred to as CPE. ISP CPE must comply with these requirements:

∑ The ISP is required to provide its own Cisco router for the connection to SprintLink. The Cisco router must run Cisco 10.2 or greater and have at least 4 Mb of flash memory for static routing and 8 Mb of flash memory for multi-homed customers running BGP-4.

∑ The ISP is fully responsible for installation, maintenance (hardware and software), initial configuration and on-going management of the CPE router.

∑ The ISP is fully responsible for the configuration of its local network environment and applications. This includes installation and configuration of Internet servers for mail, news, and the DNS.

∑ A Sprint-provided or Sprint-certified CSU/DSU is required for the connection to SprintLink. CSU/DSUs not provided by Sprint must be maintained and monitored by the ISP. Standard CPE pricing applies to all Sprint-provided CSU/DSUs.

∑ The service demarcation point is the Sprint-provided CSU/DSU, LEC provided smartjack, or central office. Sprint will monitor the network connection 24 hours a day, seven days a week and troubleshoot as

34

appropriate. However, problems that are isolated to the CPE router or the ISP's local network environment and applications are the full responsibility of the ISP.

6.3Routing ConfigurationThere are two general routing configurations for ISPs customers: static routing and dynamic routing.

∑ ISPs who have a single network connection to the Internet are considered single-homed and are configured with static routes.

∑ Multi-homed ISPs, those ISPs with two or more diverse connections to SprintLink or another service provider, should use an exterior routing protocol (preferably BGP-4) to announce their routes to SprintLink and other service providers. This requires that the ISP have an Internet AS number. In addition, the ISP is fully responsible for configuring exterior routing protocols on the CPE router for the SprintLink connection. The ISP is also responsible for configuring routing within its local network environment.

6.4Internet RegistrationsThere are several registrations required for an Internet connection. These include an Internet network number, an Internet domain name, and possibly an AS number (e.g., required for multi-homed environments only). Sprint will perform these Internet registrations on behalf of the ISP:

∑ Internet network number (i.e. IP addresses) registration: SprintLink will assign ISPs Class C IP addresses from its CIDR address block as needed. Generally, Sprint requires that 80% of the requested address space is utilized immediately. This is to encourage the efficient use of the IP address space.

∑ Domain name registration: It is recommended that ISPs provide their top three selections for each domain name requested at the time of service ordering. Should the first selection already be in use, Sprint attempts to register the second selection and so on. Please be prepared to supply the ISPs' current root domain names and requests for additional domain names during the SprintLink order process. ISPs are fully responsible for the registration of domain names for their customers. (Sprint will not register domain names for a customer of an ISP).

∑ AS number registration: The InterNIC Templates appendix has an autonomous system number registration template that lays out the process for obtaining an Internet AS number. An AS number is only necessary when the ISP wishes to "peer" with SprintLink. ISPs who require their own AS number should contact the InterNIC directly.

35

Additional templates are available from ftp://rs.internic.net in the templates directory.

6.5Domain Name Service (DNS)DNS is a distributed hierarchical system for resolving host names (e.g. mnemonic names) into IP addresses (e.g. numeric addresses). Each enterprise that registers an Internet domain name must provide multiple name servers for the domain: a primary server and a secondary server for backup purposes. Ideally, the primary and secondary name servers should run on independent and redundant systems so that availability of name service for the domain isn't affected by a single point of failure. Specific requirements regarding DNS for SprintLink ISPs are:

∑ The ISP is fully responsible for providing primary DNS for its own domain and for its customers' domains (if required). Sprint will not provide primary DNS for ISPs or their customers.

∑ Sprint will provide secondary DNS for the ISP's domain only. There is no additional charge for this service.

∑ The ISP is fully responsible for providing secondary DNS for its own customers' domains. Sprint will not provide secondary DNS for a ISP's customers.

Questions regarding DNS may be submitted via e-mail to [email protected].

6.6Trouble ManagementOnce the SprintLink Service has been implemented, the ISP is expected to follow standard SprintLink trouble management procedures. Sprint's SMC is the ISP's single point of contact for resolving troubles with SprintLink once initial service implementation has been completed. The SMC is available 24 hours per day, seven days a week. Specific requirements regarding SMC support for SprintLink resellers are:

∑ The ISP is fully responsible for resolving troubles within their local network and application environment. The ISP is also fully responsible for providing trouble resolution services to its own customers.

∑ The Sprint SMC will only accept calls from the ISP. The SMC has no responsibility for the ISP's customers and will not accept calls from the ISP's customers.

6.7Commercial Internet Exchange (CIX)The CIX is an association of commercial Internet ISPs formed in 1991 to offer Internet routing services for purely commercial traffic. Currently, the CIX is

36

comprised of over 55 Internet ISPs around the world, including regional, national and international network providers. Sprint has been a member of the CIX since its inception in 1991.

The largest CIX members (e.g., national backbone providers such as SprintLink) interconnect their networks in Santa Clara, California via a router-based system called the CIX exchange point or CIX gateway. Small or local ISPs (e.g., resellers) almost always obtain their connection to the CIX through other member networks. (Note that the term "CIX" -- pronounced "kicks" -- is used to refer to the physical exchange point as well as the trade association.)

Below are some current guidelines regarding CIX membership for reseller of SprintLink Service:

∑ On July 13, 1994, the CIX board voted to block traffic from ISPs (i.e., resellers) who are not CIX members. The consequence is that SprintLink ISPs' traffic may be blocked by filters established at the CIX gateway, unless such ISPs join the CIX. The annual fee for CIX membership is approximately $7,500 per year.

∑ Note that SprintLink does not (and will not) filter traffic from non-CIX member ISPs. Sprint was the only CIX board member to vote against the filtering proposal. Regardless of its voting position, Sprint recommends that ISPs consider joining the CIX.

∑ It should also be noted that the routing services at the CIX router are becoming less important over time, while its trade association characteristics are becoming increasingly valuable. NSPs, such as SprintLink, connect at several other higher-bandwidth interconnection points. SprintLink will deliver its customers traffic to any appropriate interconnection points to which SprintLink connects. Sprint does not recommend that a customer or ISP rely solely on CIX router connectivity for connection with the rest of the Internet.

∑ For ISPs who choose not to join the CIX, Sprint is not responsible for any restrictions placed on the ISP's traffic by the other CIX member networks.

For details regarding CIX membership requirements, procedures and annual fees, please contact the CIX directly using one of the following mechanisms:

∑ E-mail (auto-responder): [email protected]. Sending an e-mail message to this address results in an auto-response, which includes a brief description of the CIX and a CIX membership agreement. Since responses are automated, this address should not be used for general questions and answers about the CIX.

∑ Regular e-mail address: [email protected]. This mailbox is monitored by people who send customized responses. General questions and answers about the CIX should be sent here.

∑ Help-desk phone number: (703) 824-9249. The CIX help desk is staffed 8:30 a.m. to 4:30 p.m. EST Monday through Friday.

37

∑ Help-desk fax number: (703) 824-1611.

38

7Appendix - Sprint Templates

7.1IP Address Template (for ISPs)This template must be completed by Internet Service Providers (ISP) to request additional IP addresses. To obtain a block of Internet numbers, please submit the template via electronic mail to [email protected]. In the subject of the message enter, "IP REQUEST".

Following the template is a detail description of each group of fields. If you have any additional questions please send e-mail to [email protected].

39

ISP IP Address Template

1a. Name of Organization........................................1b. Postal address of Organization.........................1c. Complete Domain Name....................................1d. Type of Routing with Sprint.............................. (static or dynamic)

Technical POC Information2a. NIC handle (if known)........................................2b. Name (Last, First)..............................................2c. Title.....................................................................2d. Phone Number...................................................2f. E-Mailbox............................................................

Reassignment Information3a. Current Block(s).................................................3b. Assigned.............................................................3c. Reserved.............................................................3d. Available.............................................................3e. Last Block Allocated (Date)...............................

4. Block requested..................................................

Network Information5a. Two month projection for dial up customers.............................................................5b. Two month projection for leased line customers......................................................5c. Additional information.......................................

Name Server (for inverse mapping) Information6a. Prime Server Netaddress..................................6b. Prime Server Hostname....................................6c. Second Server Netaddress................................6d Second Server Hostname..................................

40

Instructions For Requesting An ISP IP Block

Section 1Those customers who have a single network connection to the Internet are considered single-homed and are configured with static routes. Multi-homed customers (those with two or more diverse connections to SprintLink or another ISP and running an exterior routing protocol like BGP-4) are configured with dynamic routing.

Section 3 - Reassignment Information:Examples of appropriate responses include:

current block: 204.123.0 - 204.123.31, 204.123.212 - 204.123.220assigned: 30reserved: 5available: 5 (network numbers available for customer use)

Section 4 - Size of IP Block requested:Please keep in mind that allocations are made based on a slow-start model. The initial allocations will be relatively small. Subsequent allocated blocks may be increased based on utilization verification supplied to the SprintLink.

Section 5 - JustificationPlease list your 2 month projections for dial up customers and leased line customers. Please include any additional information you feel is pertinent to help detail your requirement.

Section 6 - Name Server for inverse mappingThis information is needed for in-addr.apra registration.

Note:Subsequent allocations of IP block addresses will be based on need; this need will be demonstrated based on the number of reassignment actions that have been transmitted to the InterNIC Registry. Reassignment information is to be forwarded to the InterNIC within 7 days of the assignment so that the WHOIS may be maintained efficiently. Transmission of reassignment information is also necessary for the following reasons:

a) To ensure that a provider has exhausted, or is about to exhaust its current IP Block allocation such that an additional allocation is justified.

b) To allow operational people to see which organization is using the network and who to contact in the event of operational/security problems, etc.

c) To assist in IP allocation studies.

This can be done by: Shared WHOIS Project (SWIP)

41

Reassignment actions can be submitted by utilizing the database exchange format defined by the SWIP project. Information regarding SWIP may be obtained via anonymous FTP from rs.internic.net (198.41.0.5). The files may be found under the ftp/pub/swip directory.

[ URL = ftp://rs.internic.net/ftp/pub/swip ]

ALL ISPs, should SWIP the reassignment information. ISPs should get a maintainer ID necessary to SWIP the reassignment information. This can be done sending your request to [email protected].

ISPs are required to assign address space based on utilization efficiency. To this end, ISPs should have documented justification available for each assignment. SprintLink may at any time ask to see this justification, if not available, this could impact future allocations.

Recommended Reading (available via anonymous FTP from ds.internic.net (198.49.45.10)

Gerich, E. Guidelines for Management of IP Address Space, Ann Arbor, MI: Merit Network, Inc.; May 1993; RFC 1466. 10 p. < URL: ftp://ds.internic.net/rfc/rfc1466.txt>

Rekhter, Y., Moskowitz. B., Karrenberg, D., de Groot, G. Address Allocation for Private Internets, IBM Corp., Chrysler Corp., RIPE NCC; March 1994; RFC 1597. 8 p. < URL: ftp://ds.internic.net/rfc/rfc1597.txt>

Braden, R.T.; Postel, J.B. Requirements for Internet Gateways. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1987 June; RFC 1009. 55 p. < URL: ftp://ds.internic.net/rfc/rfc1009.txt>

Internet Engineering Task Force, Braden, R.T. Requirements for Internet Hosts -- Communication Layers. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; October 1989; RFC 1122. 116 p. < URL: ftp://ds.internic.net/rfc/rfc1122.txt>

Internet Engineering Task Force, Braden, R.T. Requirements for Internet Hosts -- Application and Support. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; October 1989; RFC 1123. 98 p. < URL: ftp://ds.internic.net/rfc/rfc1123.txt>

Internet Activities Board. Internet Official Protocol Standards. 1994 March; RFC 1600. 34p. (DS.INTERNIC.NET POLICY RFC1600.TXT). [Note: the current version is always available as "STD 1".] < URL: ftp://ds.internic.net/rfc/rfc1600.txt>

Postel, J.B. Internet Standard Subnetting Procedure. Stanford, CA: Stanford University; 1995 March; Obsoletes RFC1720 RFC 1780 < URL: ftp://ds.internic.net/rfc/rfc1780.txt>

42

Postel, J.B. Internet Control Message Protocol. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1981 September; RFC 792. 21 p. < URL: ftp://ds.internic.net/rfc/rfc792.txt>

Postel, J.B. Transmission Control Protocol. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1981 September; RFC 793. 85 p. < URL: ftp://ds.internic.net/rfc/rfc793.txt>

Postel, J.B. User Datagram Protocol. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1980 August 28; RFC 768. 3 p. < URL: ftp://ds.internic.net/rfc/rfc768.txt>

Postel, J.B. Internet Protocol. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1981 September; RFC 791. 45 p. < URL: ftp://ds.internic.net/rfc/rfc1791.txt>

Reynolds, J.K.; Postel, J.B. Assigned Numbers. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1994 October; RFC 1700. < URL: ftp://ds.internic.net/rfc/rfc1700.txt>

43

7.2IP Address Template (for End-Users)This template must be completed by SprintLink customers (non-ISPs) to request additional IP addresses. To obtain a Internet numbers, please submit the template via electronic mail to [email protected]. In the subject field of the message enter, "IP REQUEST".

Following the template is a detail description of each group of fields. If you have any additional questions please send e-mail to [email protected].

44

End-User IP Address Template1a. Name of Organization..............................1b. Postal address of Organization.................1c. Complete Domain Name...........................

Technical POC Information2a. NIC handle (if known)..............................2b. Name (Last, First)....................................2c. Title..........................................................2d. Phone Number.........................................2f. E-Mailbox..................................................

Reassignment Information3a. Current Block(s).......................................3b. Last Block Allocated (Date)......................

4. Block requested........................................

Network InformationPlease provide counts of the number of end system IP addresses you will need to be accessible to the Internet (be sure to count all devices that will need unique IP addresses which will require full connectivity to the global Internet such as PCs, terminal servers, smart hubs, servers, etc.).

5a. Current number of end systems...............5b. Additional information.............................

Name Server for inverse mapping6a. Prime Server Netaddress.........................6b. Prime Server Hostname...........................6c. Second Server Netaddress.......................6d Second Server Hostname.........................

45

BGP Guidelines For Multi-Homed Customers

1. Are you multi-homed? The definition of multi-homed is: connected to SprintLink by more than one link or connected to SprintLink and at least one other Network Service Provider (NSP). Note: all links must be dedicated

2. What are your "other" connections (to other NSPs?)

3. Were any of your connections acquired with the intention of doing some type of load balancing? If so please explain.

4. What type(s) of router will interface with SprintLink and what is the hardware/software configuration?

5. What type of routing policy do you intend to implement? This should be a description of how packets go to/from different places in the Internet from/to their networks. What is the plan to achieve this (full routing/partial routing, default networks, routing with other provider(s), etc.). This should also include a diagram of the customer's network and itsconnections to other providers.

Based on the described policy, engineering needs to "design" a routing plan before implementation can begin. Depending on the policy, coordination with the customer's "other" provider and possibly changes will be required.

6. Do you have a registered Autonomous System Number (ASN) or numbersIf so what are they? If not it is your obligation to acquire them fromthe appropriate registering source. (Internic, Ripe, APnic)

What Sprint is looking for:

Customers must supply Sprint with copies of their border router configurations for analysis. Sprint will be looking for:

1. There should be NO unfiltered redistribution from interior routing into BGP. Explicit distribute-list or network statements should be used to prevent injections of bogus routes into global tables.

2. There should be NO redistribution from BGP into interior routing protocols, as it corrupts as-path information.

3. There should be as-path filters preventing leakage of routing information from customer's other service provider to us and vice versa. Those filters should be inclusive, rather than exclusive (i.e. they should list customer as-es instead of excluding other provider's as-es).

4. Blocks of specific routes should be aggregated as much as possible.

46

5. Networks listed in configuration should be sane (i.e. no networks assigned to other customers, subnets should never be announced outside, etc.).

6. There should be no obvious security problems with the configuration (i.e. no cleartext passwords, or unlimited r/w SNMP access).

7.3IN-ADDR.ARPA TemplateThis template must be completed as part of the implementation process and when requesting submitting change requests. Additional templates are available anonymously at ftp://ftp.sprintlink.net/templates. Initial in-addr.arpa registrations are finalized during the implementation process. SprintLink Implementations will complete this template, based on information gathered directly from the customer, for initial SprintLink installations. Customers submitting change requests should e-mail completed registration templates to [email protected]. If e-mail is not available please fax your request to 703-689-5957:

DNS AdministratorSprintFax: 703-689-5697Attn: in-addr.arpa Request

The Internet uses a special domain to support gateway location and Internet address to host mapping. The intent of this domain is to provide a guaranteed method to perform host address to host name mapping and to facilitate queries to locate all gateways on a particular network in the Internet.

The following information is needed for delegation of registered networks in domains for inclusion in the in-addr.arpa zone files:

∑ In-addr.arpa domain∑ Network name∑ Hostnames of the two hosts on networks that will be acting as IN-ADDR

servers

In-ADDR domains are represented using the network number in reverse. For example:

IN-ADDr domain

Network Domain Network name IN-ADDR Servers(Hostname)(Net address)(CPUType/OpSYS)

41.192.IN-ADDR.ARPA NET-TEST-ONE BAR.F00.EDU123.45.67.89

47

VAX-II/VMSONE.ABC.COM98.76.54.32SUN/UNIX

For example:

Network 123.45.67.0’s IN-ADDR domain is represented as 67.45.123.in-addr.arpa.

Note: Unless specified, new hosts registered as IN-ADDR servers will be registered in the root servers only and will not appear in the HOSTS.TXT file.

Please have the network coordinator complete and return the following information for those networks needing IN-ADDR registration:

IN-ADDR DomainNetwork NameIN-ADDR Servers

7.4IN-ADDR.ARPA Express [email protected] receives many requests regarding updates to network DNS. While the InAddr concept requires the netnumbers to process in reverse, our databases can also accept netnumbers in forward format, for example:

137.39.0.0 <not 39.137.In-ADDR.ARPA>

The best format for the quickest processing is as follows:

Company Name.........................InAddr ExpressTechnical POC .........................Joe Smith

Single NetNumber....................223.4.5.38 orBeginning of blocknumber........223.4.5.0Ending of blockrecord...............223.4.8.0

Primary Hostname....................ns.inaddrexp.comPrimary Address........................223.4.5.100

Secondary Hostname................dns.example.netSecondary Address....................123.45.67.8

48

8Appendix - InterNIC Templates

The following templates can be used to receive services directly from the InterNIC.

8.1Domain Name RegistrationThis template is taken from:

ftp://www.internic.net/templates/domain-template.txt. [9/95]

It is included here for the convenience of those customers who, for various reasons, must register their own domain name.

---------------------------------------------- CUT HERE -------------------------------------------------

InterNIC Domain Name Template

************ Please DO NOT REMOVE Version Number ************

Domain Version Number: 2.0

**************** Please see attached detailed instructions *****************

****** Only for registrations under ROOT, COM, ORG, NET, EDU, GOV *****

0. (N)ew (M)odify (D)elete....:

1. Purpose/Description........:

2. Complete Domain Name.......:

Organization Using Domain Name

3a. Organization Name..........:3b. Street Address.............:3c. City.......................:3d. State......................:3e. Postal Code................:3f. Country....................:

49

Administrative Contact

4a. NIC Handle (if known)......:4b. Name (Last, First).........:4c. Organization Name..........:4d. Street Address.............:4e. City.......................:4f. State......................:4g. Postal Code................:4h. Country....................:4i. Phone Number...............:4j. E-Mailbox..................:

Technical Contact


Billing Contact


Primary Name Server

7a. Primary Server Hostname....:7b. Primary Server Netaddress..:

Secondary Name Server(s)

8a. Secondary Server Hostname..:8b. Secondary Server Netaddress:

Invoice Delivery

50

9. (E)mail (P)ostal...........:

A domain name registration fee of $100.00 US is applicable. This charge will cover the $50.00 maintenance fee for two (2) years. After the two year period, an invoice will be sent on an annual basis.

The party requesting registration of this name certifies that, to her/his knowledge, the use of this name does not violate trademark or other statutes.

Registering a domain name does not confer any legal rights to that name andany disputes between parties over the rights to use a particular name are to be settled between the contending parties using normal legal methods (see RFC 1591).

By applying for the domain name and through the use or continued use of the domain name, the applicant agrees to be bound by the terms of NSI's then current domain name policy (the 'Policy Statement') which is available at ftp://rs.internic.net/policy/internic/internic-domain-1.txt.

(If this application is made through an agent, such as an Internet Service Provider, that agent accepts the responsibility to notify the applicant of the conditions on the registration of the domain name and to provide the applicant a copy of the current version of the Policy Statement, if so requested by the applicant.) The applicant acknowledges and agrees that NSI may change the terms and conditions of the Policy Statement from time to time as provided in the Policy Statement.

The applicant agrees that if the use of the domain name is challenged by any third party, or if any dispute arises under this Registration Agreement, as amended, the applicant will abide by the procedures specified in the Policy Statement.

This Registration Agreement shall be governed in all respects by and construed in accordance with the laws of the United States of America and of the State of California, without respect to its conflict of law rules.

This Registration Agreement is the complete and exclusive agreement of the applicant and NSI ("parties") regarding domain names. It supersedes, and its terms govern, all prior proposals, agreements, or other communications between the parties. This Registration Agreement may only be amended as provided in the Policy Statement.

----------------------------------------------- CUT HERE ----------------------------------------------

General Instructions

51

Use the form above for registering new domain names, for making changes to existing domain name records and for removing a domain name from the InterNIC database and root servers. The form, and only the form, should be sent via e-mail to [email protected]. Please do not send hardcopy registrations to the InterNIC. Your provider will be able to send e-mail applications if you are not connected.

In the Subject of the message, use the words, "NEW DOMAIN", "MODIFY DOMAIN", or "REMOVE DOMAIN" as appropriate, followed by the name of the domain to assist in sorting and locating incoming registration requests.

In response to the submission of a form, you should receive an auto-reply with a tracking number. Use the number in the Subject of any messages you send regarding that registration action. When the registration is completed you will receive a notification via e-mail.

If you regularly submit domain applications, you will note three changes in the registration form. Section 4 has been removed. We assume that the domain will essentially be operational when the template is submitted and have found the "operational date" to be of little use. Removal of Section 4 moves the Administrative and Technical Contact sections to Section 4 and 5 respectively. A Billing Contact section was added as Section 6, so that invoices and other billing information can be sent to the Billing Contact. Finally, additional fields have been added to the parts of the form that deal with addresses. Street Address, City, State, Postal Code and Country MUST be placed in separate fields.

Please do not modify the form nor remove the version number. The computer program that scans and parses the form is looking for section numbers, followed by a period, followed by a colon. Information following the colon is compared with and inserted into the database as appropriate. Please send only one form per message.

When completing the form, make use of "whois" at rs.internic.net to check to see if the domain name, organization name, people, and name servers have been registered. Use the information in the database where appropriate. This will minimize the number of registrations that are returned for incomplete or inaccurate information.

The instructions for completing each field are in the following three sections - one each for NEW, MODIFY and DELETE. The interpretation of missing or blank fields can vary between the New and Modify registration actions. Please read the instructions carefully and make sure the form is properly completed to accomplish the action you desire.

The information on billing of domains and invoices is available at

ftp://rs.internic.net/billing/README

52

8.1.1Registering A New Domain Name

Section 0 - Registration Action Type

Following the colon, place the character "N" or the word "NEW" to indicate a NEW domain registration. Changing a name from DOMAINAME1 to DOMAINAME2 is treated as a New registration. The transfer of a name from one organization to another is also treated as a New registration. If the intent is to effect a transfer from one organization to another, clearly indicate this in Section 1 and include a statement from the current holder of the name that it is being transferred.

Section 1 - Purpose of Registration

Briefly describe the organization and/or the purpose for which this domain name is being registered. The description should support the choice of top-level domain in Section 2. If the domain name is for an organization that already has a domain name registered, describe the purpose of this domain and why the additional name is needed. Indicate why existing names cannot be used or why the proposed second-level name cannot be used as a third-level name under a domain name that is already registered.

Section 2 - Complete Domain Name

Top-level country domains may be registered by inserting the two-letter country code in this section. See RFC1591 for the duties and responsibilities of top-level domain administrators.

For second-level domain names under COM, ORG, NET, EDU, GOV insert the two-part name of the domain you wish to register, for example, ABC.COM. The total length of the two-part name may be up to 24 characters. The only characters allowed in a domain name are letters, digits and the dash (-). A domain name cannot begin or end with a dash (see RFC952). Consult RFC1591 to determine the most appropriate top-level domain to join. Briefly:

∑ COM is for commercial, for-profit organizations∑ ORG is for miscellaneous usually non-profit organizations∑ NET is for network infrastructure machines and organizations∑ EDU is for 4-year, degree granting institutions∑ GOV is for United States federal government agencies

US state and local government agencies, schools, libraries, museums, and individuals should register under the US domain. See RFC1480 for a complete description of the US domain and registration procedures.

GOV registrations are limited to top-level US Federal Government agencies (see RFC1816).

53

Section 3 - Organization using the Domain Name

The domain name is considered to be registered to an organization, even if the "organization" is an individual. It is important in this section to list the name and address of the end-user organization, not the provider organization.

If the organization has the same name as one that is already registered, explain this in Section 1 above.

Item 3b may be copied as many times as necessary to reflect different lines of the Street Address. If item 3c, 3d or 3e is not applicable for your country, leave that item blank.

Section 4 - Administrative Contact

The administrative contact is the person who can speak on behalf of the organization listed in Section 3. This person should be able to answer non-technical questions about the organization's plans for the name, and procedures for establishing sub-domains, and should be able to represent the organization regarding use of the name. See RFC1032 for more detail on administrative contacts.

Each person in the InterNIC database is assigned a "handle" - a unique tag consisting of the person's initials and a serial number. This tag is used on records in the database to indicate a point of contact for a domain name, network, name server or other entity. Each person should have only one handle.

If the person's handle is known, insert just the handle in item 4a and leave the rest of Section 4 blank.

If a person's handle is unknown or the person has never been registered, leave item 4a blank. The registration software will check for an existing user record. If a matching user record is found, the record will be updated with any new information contained in the template.

All contacts MUST include information in items 4i and 4j. Templates that do not have this information will be returned.

Section 5 - Technical Contact

The technical contact is the person who tends to the technical aspects of maintaining the domain's name server, resolver software, and database files. This person keeps the name server running, and interacts with technical people in other domains to solve problems that affect the domain. The Internet Service Provider often performs this role.

54

The procedures for completing Section 5 are the same as those for Section 4. If Section 5 is left blank, the information from Section 4 will be assumed and vice-versa.

Section 6 - Billing Contact

The billing contact will be sent invoices for new domain registrations and re-registrations.

The procedures for completing Section 6 are the same as those for Section 4. If Section 6 is left blank, the information from Section 4 will be assumed.

Section 7 - Primary Name Server

Domains MUST provide at least two independent servers for translating names to addresses for hosts in the domain. The servers should be in physically separate locations and on different networks if possible. The servers should be active and responsive to DNS queries BEFORE this application is submitted. Incomplete information in Sections 7 and 8 or inactive servers will result in delay of the registration.

The registration software makes a cross check between the host name given and the IP addresses given to see if there are matches with either in the database. If a match with an IP number in the database is found, the name in the database will be assumed. Neither the name nor number of a registered name server will be changed as a result of a new domain registration. A Modify registration request MUST be sent to change either of these values.

Please provide the fully-qualified name of the machine that is to be the name server; for example: "machine.domainame.com" not just "machine"

Section 8 - Secondary Name Server(s)

The procedures for completing Section 8 are the same as for Section 7. At least one secondary name server is required. If several secondary servers are desired, copy Section 8 as many times as needed. Please do not renumber or change the copied section.

Section 9 - Invoice Delivery

If you wish to receive your invoice electronically, place the character "E" or the word "EMAIL" in item 9. If you wish to receive your invoice by postal mail, place the character "P" or the word "POSTAL" in item 9.

8.1.2Modifying A Domain Name Record

Changing an existing record is done by replacement. That is, the contents of various fields in the database are replaced with new information from the form.

55

If the modification involves first registering a person or name server that is not in the database, the instructions for completing Sections 4-8 "REGISTERING A NEW DOMAIN NAME" apply. Use "whois" if you are unsure about the current information for a domain, name server, or individual.

Changes will be made if it appears that the modification request has come from a "reasonable" source. This source may be from a listed contact for the domain, others in the same organization, the current provider, or a new provider that is about to provide support for the domain.

Notification of the change and the approximate time the change will take effect will be sent to:

∑ the requester∑ if contacts are changing, both old and new contacts ∑ if name servers are changing, the technical contacts for the domains in

which the old and new primary name servers reside

This dissemination of change information is to ensure that all parties involved are aware and concur with the change.

Transfer of a name from one organization to another is considered a new registration. See, "REGISTERING A NEW DOMAIN NAME."


Following the colon, place the character "M" or the word "MODIFY" to indicate a modification to an existing domain registration. Transfer of a name from one organization to another is considered a new registration. See, "REGISTERING A NEW DOMAIN NAME."


Briefly describe the purpose of the modification. If the intent is to change either the name or the IP address of name servers, make this very clear, otherwise changes to name servers will not be made.

If you wish to change the name of the domain itself, file a separate NEW registration and a DELETE registration when you are ready to have the old name removed. Make it clear in this section that you are changing from one domain name to another and estimate how long you will need the old domain name.


Insert the two-part name of the domain name you wish to modify, for example, ABC.COM. This section MUST be completed, even if you are making minor changes, like the phone number of a point of contact, for example.

56

Section 3 - Organization Using the Domain Name

The domain name is considered to be registered to an organization, even if the "organization" is an individual. Therefore, a change in the organization name should be explained in Section 1. Transfer of a domain name from one organization to another is considered a new domain registration. See, "REGISTERING A NEW DOMAIN NAME."

Section 4 - Administrative Contact

If a handle is given in item 4a, the database will be updated with any information on the form in items 4b through 4j. If any item is missing or blank, the corresponding information in the database will not be altered.

If no handle is given, the database will be searched to see if the person is already registered. If so, the person's record will be updated with any new information on the form. If not registered, the person will be entered into the database and assigned a handle. The handle will be attached to the domain name record, thus replacing the old contact with the new one. Both old and new contacts will be notified of the change.

Section 5 - Technical Contact

The procedures for modifying information in Section 5 are the same as those for Section 4.

Section 6 - Billing Contact

The procedures for modifying information in Section 6 are the same as those for Section 4.

Section 7 - Primary Name Server

If any changes in name servers are being made, provide a complete list of name servers (both primary and secondary). The list in the application will replace the list in the database.

If the intention is to change the name or IP address of a name server, clearly state that in Section 1 above. The default is to NOT change either of these values unless so instructed in Section 1 and the requester is authorized to do so.

Section 8 - Secondary Name Server(s)

If any changes in name servers are being made, provide a complete list of name servers (both primary and secondary). The list in the application will replace the list in the database.

Section 9 - Invoice Delivery

57

Leave this section blank.

8.1.3Deleting A Domain Name Record

A request to remove a registered domain name from the database may come from the administrative or technical contact for the domain or from a provider who is no longer supporting the name. A third party who believes that a domain is inactive and wants to become the holder of the name should NOT submit a Delete form. Instead, a New form should be submitted with a description in Section 1 of attempts to contact the listed domain name holder.

When the name is removed, it will no longer be visible via whois, wais, gopher, or Web browsers that query the InterNIC database. It will also be removed from the root name servers on the subsequent update.

Notification of the deletion and the approximate time it will take effect will be sent to:

∑ the requester∑ the contacts currently listed ∑ the technical contact for the domain in which the primary name server

resides

When a name is transferred from one organization to another, the InterNIC will first delete the existing record and then process a new registration.

If an invoice is not paid within the specified time period, the domain name will be automatically deleted.

If the domain name has been placed on hold for more than 90 days because of lack of name service, the domain name may be deleted.


Following the colon, place the character "D" or the word "DELETE" to indicate a deletion of an existing domain registration.


Briefly state the reason for removing the name.

If additional records, such as the point(s) of contact or name servers should also be deleted, state that in this section. These will not be automatically removed. The InterNIC will check further to see if they are used on other domain, network, or autonomous system number records before removing them.


58

Insert the two-part name of the domain name you wish to have deleted.

Section 3 - Organization Using the Domain Name

This section MUST be filled completely.

Section 4 - Section 9

These sections should be left blank.

RECOMMENDED READING

Albitz, P., C. Liu, "DNS and Bind" Help for UNIX System Administrators, O'Reilly and Associates, Inc., October 1992.

Postel, J., "Domain Name System Structure and Delegation". Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1994 March; RFC 1591. 7 p. <URL: ftp://rs.internic.net/policy/rfc1591.txt>

Cooper, Postel, "The US Domain". Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1992 December; RFC 1480. 31 p. <URL: ftp://rs.internic.net/policy/rfc1480.txt> Stahl, M.K. Domain Administrators Guide. Menlo Park, CA: SRI International, DDN Network Information Center; 1987 November; RFC 1032. 14 p. <URL: ftp://rs.internic.net/policy/rfc1032.txt>

Lottor, M. Domain Administrators Operations Guide. Menlo Park, CA: SRI International, DDN Network Information Center; 1987 November; RFC 1033. 22 p. <URL: ftp://rs.internic.net/policy/rfc1033.txt>

Mockapetris, P. Domain Names - Concepts and Facilities. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1987 November; RFC 1034. 55 p. <URL: ftp://rs.internic.net/policy/rfc1034.txt>

Mockapetris, P. Domain names - Implementation and Specification. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1987 November; RFC 1035. 55 p. <URL: ftp://rs.internic.net/policy/rfc1035.txt>

Mockapetris, P. DNS Encoding of Network Names and Other Types. Marina del Rey, CA: University of Southern California, Information Sciences Inst.; 1989 April; RFC 1101. 14 p. <URL: ftp://rs.internic.net/policy/rfc1101.txt>

59

8.2AS Number RegistrationRegistering for an AS number implies an intent to implement one or more gateways and use them to connect networks in the Internet. Customers should provide Sprint with details about their plans and information about the administrative authority they have for participating in the Internet. It is strongly advised that customers follow the development of inter-autonomous systems protocols in the IAB task forces.

Send the completed application via e-mail to [email protected]. If e-mail is not available, please fax to:

Network SolutionsFAX: 703-742-4811

Questions may be sent to the registration services hostmaster by e-mail at the above address, or by phone at (800) 444-4345 or (703) 472-4777.

To obtain an AS number the following information must be provided:

Note: The key people must have electronic mailboxes and NIC “handles,” which are unique NIC database identifiers. Customers with access to WHOIS, can check to see if they are registered and, if so, make sure the information is current. Include only the Handle and any changes(if any) that need to be made. Customers who do not have access to WHOIS or are not registered in the NIC WHOIS database should provide all the information requested. A NIC Handle will be assigned.

1. The name, title, mailing address, phone number and organization of the administrative head of the organization. This is the contact point for administrative and policy questions about the autonomous system. In the case of a research organization this should be the principal investigator. The on-line mailbox and handle (if any) of this person should also be included.

Example:Administrator Organization: Network SolutionsName: Last name, first nameTitle: Principal investigatorMail address: Network Solutions, Inc.

505 Huntmar Park DriveHerndon, Virginia 22090

Phone number: (703) 742-4777Net mailbox: [email protected] handle BB34

2. The name, title, mailing address, phone number and organization of the technical contact. The on-line mailbox and NIC handle (if any) of the technical contact should also be included. This is the contact point for problems with the

60

autonomous system and for updating information about the autonomous system. Also, the technical contact may be held responsible for the behavior to gateways in this autonomous system.

Example:Technical contact organization: Network Solutions, Inc.Name: Last name, first nameTitle: Computer ScientistMail address Network Solutions

505 Huntmar Park DriveHerndon, Virginia 22070

Phone number: (703) 742-4777Net mailbox [email protected] handle cc56

3. The name of the autonomous system (up to 12 characters). This is the name that will be used in tables and lists associating autonomous systems and numbers.

Examples: ALPHABETA

4. A description of the gateway that implements the inter-autonomous system protocol for interaction with other autonomous systems. Currently the exterior gate protocol (EGP) is being used for this purpose (RFC 904). This gateway should comply with RFC 1009, Requirements for Internet Gateways.

Example: This is a new gateway developed by the XYZ Group at the ABC Corporation.

Example: This is an instance of the well known IJK company gateway, but set up as a distinct autonomous system.

5. A description of the gateway hardware, including CPU and interfaces.

Example: DEC PDP-11/23, ARPANET Interface by ACC, Ethernet Interfaces by 3 COM.

6. A description of the gateway software, including operating system and languages.

Example: The gateway is implemented using language “C” and runs Berkeley 4.2 UNIX.

7. The deployment schedule for the autonomous system.

a. Initiallyb. Within one yearc. Two years andd. Five years

61

Examples:a. Initially = 1b. One year = 3c. Two years = 5d. Five years = 10

8. What networks will be interconnected by these gateways? What are the internet addresses of each gateway?

Example: One gateway, interconnecting BNET and LEWIS-PRNET2, with address 192.8.15.3 on BNET and 128.108.12.10 on LEWIS-PRNET2.

For further information contact:

InterNIC Registration Services13221 Woodland Park RoadHerndon, Virginia 22071

Electronic mail: [email protected]: 1-800-444-4345

Recommended Reading

Mills, D.L. Exterior Gateway Protocol Formal Specification. 1984. April; RFC 904. 30P (RS.INERNIC.NETPOLICY RFC904.TXT)

Braden, R.T.:Postel, J.B. Requirements for Internet Gateways 1987 June; RFC 1009. 55 p (RS.InterNIC.NETPOLICY RFC1009.TXT).

62

9Appendix - InterNIC Bulletin: Fee for Registration of Domain Names

The following is a reprint of the InterNIC Fee for Registration of Domain Names that can be found on the InterNIC worldwide web page at http://rs.internic.net/announcements/fee-policy.html.

------------------------

Fee for Registration of Domain Names

1. Introduction

1.1 Summary

Since March 1, 1993, the National Science Foundation has funded theadministration of the "COM", "ORG", "NET", "EDU", and "GOV" and root domainsthrough a Cooperative Agreement with Network Solutions, the InterNIC Registrar.Beginning not later than September 13, 1995, the Registrar will require directpayment from domain name applicants and holders for registration andmaintenance of the domain names at the second level of the five listed top-leveldomains. The funds received from those fees will replace the funding provided bythe National Science Foundation, and will provide "program income" which willoffset costs related to the intellectual infrastructure of the Internet.

1.2 Background

Originally, the Internet began as a research experiment and network known as theARPANET, which supported the exchange of files and data among governmentcontractors and researchers. As the TCP/IP suite was developed in the early 1980s,the Domain Name System (DNS) emerged as the replacement system to the originalARPANET hosts.txt mechanism. The high-level structure of names used by DNSeventually evolved into five world-wide generic domains ("COM", "ORG", "NET","EDU", and "INT"), two U.S. only generic domains ("MIL" and "GOV"), and countrycode domains (e.g., "US" for the United States, "AU" for Australia, etc.)

The exponential growth of the Internet, due mostly to the connecting of commercialorganizations to the Internet over the past couple years, has had a directlyproportional affect on the registration activity of the Registrar. The increasedactivity, with the corresponding growth of operating costs, have resulted in fundingrequirements exceeding the National Science Foundation's budget. In addition, it isappropriate that Internet users, instead of the U.S. Federal Government, pay the

63

costs of domain name registration services. Accordingly, the Registrar will begincharging a fee for the registration and maintenance of domain names in the "COM","ORG", "NET", "EDU", and "GOV" domains.

1.3 Guidelines, Procedures, and Policies

The guidelines, procedures, and policies that are currently operative with regard tothe Domain Name System will not change (RFCs 1031 through 1035, RFC 1480, andRFC 1591).

2. Definitions

"Applicant" The party or organization, such as a company or service provider, thatis applying for a new domain name.

"Domain" Any root-level domain (i.e. . within the "COM", "ORG", "NET", "EDU",and "GOV" root domains. (Note: This does not include domains "COM", "ORG","NET", "EDU", and "GOV" which exist under a country domain. For example,"COM.AU" is the commercial subdomain in Australia)

"Contact" The person responsible for a particular domain or aspect of a particulardomain. There are four types of contacts: administrative, technical, zone, and billingcontacts. These contacts are listed in the InterNIC's WHOIS database.Communication with contacts is via the email address that is listed in this database.

"Domain Name" The operational name used by TCP/IP applications that identifiesan organization connected to the Internet.

"DNS" The Domain Name System handles mapping from hostnames to Internetaddresses. Configuration information from the InterNIC Registration Servicesdatabase is released into DNS three times a week (Monday, Wednesday, and Fridayat 5 p.m. EST)

"WHOIS" A database utility that allows queries of domain records which includecompany, contact, and operational information from the InterNIC RegistrationServices database.

"New Domain Name" Names that have not yet been entered into the InterNICRegistration Services database, or names that have expired and have been removedfrom the InterNIC database and made available for reissue to an applicant at a latertime.

"Existing Domain Name" Names that are registered within the InterNICRegistration Services database and accessible through WHOIS.

"Internet Service Provider" Commercial companies that provide connectivitybetween the domain name holder and the Internet, and that assist with domainname registrations.

3. Charges

64

The Registrar will apply two types of charges with respect to domain names. Thefirst is a "Registration Fee" (or initial fee) for new domain names; the other is a"Maintenance Fee" (or recurring fee) for existing domain names that are alreadyregistered. These initial and recurring fees for each domain name cover the costs foran unlimited number of update requests, including changes in domain name, foreach domain name. The Registrar requires that all applications continue to be sentelectronically, and that the payments are made by any of the methods outlinedbelow. Payment must be in U.S. dollars. (Please note that "IN-ADDR.ARPA"domains are not subject to Registration Fees.) The Registrar will announce anyfuture price changes, which are subject to National Science Foundation approval, bynotice on "[email protected]."

3.1 New Domain Names

3.1.1 Registration Fee

The Registration Fee for a new domain name is $100.00. New domain names arevalid for two years from the date that the Registrar activates the domain name. TheRegistrar will remove domain names from the database upon the request of thedomain name holder; however, the Registration Fee is non-refundable. DomainNames deleted from the database will be available for reuse as described in Section3.3.

3.1.2 Payment of Fee

The Registrar will activate domain names upon request, on a first-come, first-servebasis. Payment of the Registration Fee is due by 12:00 PM (Eastern Time) on the 30thday after the activation date, or on the last work day preceding the due date if the30th day falls on a weekend or holiday. The Registrar will delete the domain namefrom the database on the day after the due date if payment is not received. DomainNames deleted from the database will be available for reuse as described in Section3.3.

3.2 Existing Domains

3.2.1 Maintenance Fee

For all registered domain names, there will be an Maintenance Fee of $50.00 per yearper domain name, due upon the anniversary date of the domain name activation.This annual fee will keep the domain name valid for one year. Payment must bemade in advance on an annual basis. The payment is non-refundable

3.2.2 Payment of Fee

Payment of the Maintenance Fee is due by 12:00 PM (Eastern Time) on theanniversary of the activation date, or the last workday preceding the anniversarydate if the anniversary date falls on a weekend or holiday. The Registrar will delete

65

the domain name from the database on the day after the due date if payment is notreceived. Domain Names deleted from the database will be available for reuse asdescribed in Section 3.3.

3.2.3 Notification of Maintenance Fee Due

The Registrar will provide 60 days advance notice, by email, to the domain namebilling contact (or administrative contact, in the absence of a separately identifiedbilling contact), with a copy to the administrative, technical, and zone contacts, thatthe Maintenance Fee is due on the anniversary of the activation date. In addition,email notification will also be sent to the address that is advertised via DNS withinthe Start of Authority (SOA) record.

The Registrar will send reminder notices 30 days and 15 days prior to the activationanniversary date, unless it receives payment. If the Registrar does not receivepayment by the due date, it will remove the name from the DNS and notify thecontacts that such action is being taken. Domain Names deleted from the databasewill be available for reuse as described in Section 3.3.

The Registrar will not attempt to notify the contacts by any means other than email.It is the responsibility of the administrative, billing, technical, and zone contacts tokeep their records up to date in the InterNIC Registration Services' database.

3.3 Domain Name Deletions

Domain names deleted from the database will become available for reuse after a 60day "hold" period. The purpose of the 60 day period is to minimize the probabilitythe next holder of that domain name will receive messages intended for theprevious holder.

3.4 Special Initial Grace Period

To allow domain name applicants and holders time to prepare for the new billingprocedures, the Registrar will grant an initial grace period of 90 days on all domainname registration transactions. All new applications during the three calendarmonths following the implementation of fees will have 90 days to pay for theirdomain names. All domain names with annual maintenance fees due during thethree calendar months following implementation of fees will have 90 days to pay fortheir domain names. All new domain name applications received before theimplementation date will be exempt from the initial domain name Registration Fee;however, they will be charged the Maintenance Fee on each activation anniversarydate.

4. Invoicing and Payment

The Registrar requires that all applications be submitted electronically to"[email protected]", and that the payment be made by any of the methodsdescribed in this section. All fees are to be paid in U.S. currency. Facsimile numbersand U.S. postal addresses are:

66

FAX: (703) 742-4811 (InterNIC Registration Services)

U. S. Mail InterNIC Registration Services P.O. Box 1656 Herndon, VA 22070 USA

4.1 Special Payments

4.1.1 "EDU" and "GOV" Domains

The National Science Foundation will pay the fees associated with domain nameregistrations and maintenance for the academic institutions registered in the "EDU"domain. The National Science Foundation will pay the fees, on an interim basis,associated with domain name registrations and maintenance for the U. S. FederalGovernment agencies registered in the "GOV" domain..

4.1.2 Internet Service Providers

As an alternative to domain name by domain name billing, the Registrar will allowInternet Service Providers to establish and maintain special accounts against whicheach registration and maintenance action will be debited. The Registrar will requireeach Internet Service Provider using this optional payment method to maintain apositive balance in their account. The Registrar will not process domain nameapplications and renewals if there is an insufficient balance in the respectiveaccount.

4.2 Invoices

Except as set forth in Section 4.1, the Registrar will batch process and electronicallysend invoices each night. The invoices will include domain names registered thepreceding day.

4.3. Checks and Money Orders

All checks and money orders should be made out to "InterNIC RegistrationServices." For single domain registrations, the applicant must list the domain nameon the check. For multiple domain registrations, except in the special cases describedin Section 4.1 above, the applicant must list on the payment template the domainnames for which payment is being made.

Insufficient payments or failure to indicate the domain name(s) associated with thepayment will result in a returned check. Returned checks will not change the duedate for payment. Failure to resubmit proper payment by the due date will result inthe loss of the domain name.

4.4 Credit Cards

Except in the special cases described in Section 4.1 above, organizations may pay theRegistration and Maintenance Fees by American Express, MasterCard, or VISA. TheCard Number, Expiration Date, and Name, as it appears on the card, must be

67

supplied with each charge on the hardcopy InterNIC Registration Services paymentform which will be attached to the electronic domain name applicationacknowledgment. The payment form is also available from the InterNICRegistration Services Help Desk, or electronically at"ftp://rs.internic.net/templates/pmt-template.txt". For security reasons, InterNICwill not accept credit card information submitted by email. For single domainregistrations, the applicant must list the domain name on the form. For multipledomain registrations, except in the special cases described in Section 4.1 above, theapplicant must indicate on the payment template the domain names for whichpayment is being made.

The Registrar can not process payments unless the domain names are identified,and will not process payments which are insufficient to pay for the cost of domainname registration or renewal. The Registrar will return improperly submittedpayments to the domain name holder. Payments returned because of impropersubmission will not change the due date for payment. Failure to resubmit properpayment by the due date will result in the loss of the domain name. Credit cardpayments for domain name applications and renewals may be faxed to the numberlisted above.

4.5 Submission by U.S. Postal Mail

Checks, money orders, and credit card information may be sent to InterNICRegistration Services by U.S. Postal Mail to the address listed in Section 4 above.

4.6 Future Methods of Payment The Registrar is considering other paymentmethods, such as submission by electronic means. Of particular concern is thesecurity associated with electronic traffic over the Internet. The Registrar plans tophase electronic payment implementation after security analyses are complete.

4.7 Form of Payment Not Accepted

The Registrar will not accept currency, coin, purchase orders, or stamps as payment.The Registrar will not accept credit card information over the telephone or sent byemail.

4.8 Taxes

Because the InterNIC is selling a service from the Commonwealth of Virginia,Registration Fees and Maintenance Fees are exempt from sales tax.

5. Domain Name Registration Funds

All funds collected as the result of charging a fee for domain name registration are"program income." The Registrar will use 70 percent of the income to offset the costsof providing InterNIC Registration Services.

The Registrar will dedicate 30 percent of the funds to offset the costs which areassociated with the intellectual infrastructure which underlies the operation of theInternet. Areas of the infrastructure to be supported will be discussed in the

68

InterNIC Registrar's Annual Report and Project Plan to the National ScienceFoundation, and will be subject to oversight by an advisory panel. The panel willconsist of representatives from the Internet community. A public accounting will bemade of the expenditures toward the Internet infrastructure.

6. Suggestions and Complaints

Billing questions should be send to "[email protected]." Comments should beaddressed to [email protected]

69

10Appendix - Glossary

Access ListA list kept by routers to control access to or from the router for a number of services (for example, to restrict packets with a certain IP address from leaving a particular interface on the router).

ARPANETAdvanced Research Projects Agency Network. A landmark packet-switching network developed in the early 1970s by BBN and funded by ARPA (and later DARPA). The ARPANET evolved into the Internet, and the term ARPANET was officially retired in 1990. See also Internet.

ASautonomous system. A collection of networks under a common administration sharing a common routing strategy. An AS must be given a unique 16-bit number that is assigned by the InterNIC.

ATMAsynchronous Transfer Mode. The CCITT (now ITU-T) standard for cell relay wherein information for multiple service types (such as voice, video, or data) is conveyed in small, fixed-size cells.

AUPAcceptable Use Policy. Formally required by the National Science Foundation as a commitment to not pass “unacceptable” traffic across the NSFnet.

BGP-4Border Gateway Protocol version 4. An interdomain routing protocol that is a potential replacement for EGP. BGP is defined by RFC 1105.

BSGBusiness Services Group. A division of Sprint Long Distance.

CIDRClassless interdomain routing. Technique supported by BGP-4 and based on route aggregation. Allows routers to group routes together in order to cut down on the quantity of routing information carried by core routers. With CIDR, several IP networks are combined, from the point of view of networks outside the group, into a single, larger entity.

CIXCommercial Internet Exchange. A network exchange point for

70

interconnecting commercial traffic

CLINA GSD billing platform used to bill SprintLink services ordered before March 27, 1995.

CLLICommon Language Location Indicator

CPECustomer Premise Equipment

CSCSprint’s Customer Service Center.

CSUchannel service unit. A digital interface device that connects end-user equipment to the local digital telephone loop.

DAFDigital Access Facility.

DNSDomain Name System. A system used in the Internet for translating names of network nodes into addresses.

DSUData service unit. A device used in digital transmission for connecting DTE, such as a router, to DCE or to a service.

ESEnterprise System.

FDDIFiber Distributed Data Interface. An ANSI-defined standard specifying a 100-Mbps token-passing network using fiber-optic cable, with transmission distances of up to 2 km. Uses a dual-ring architecture to provide redundancy.

FIXFederal Internet Exchange. A network exchange point for interconnecting government traffic.

Frame RelayA protocol used across the interface between user devices (for example, hosts and routers) and network equipment (for example, switching nodes). Frame Relay is more efficient than X.25, the protocol for which it is generally considered a replacement.

FTP

71

File Transfer Protocol. An IP application protocol for transferring files between network nodes.

GSDGovernment Services Division. A division of Sprint Long Distance.

IABInternet Activities Board. A group of internetwork researchers who meet regularly to discuss issues pertinent to the Internet. This board sets much of the policy for the Internet through decisions and assignment of task forces to various issues.

ICMPInternet Control Message Protocol. A network layer Internet protocol that provides message packets to report errors and other information relevant to IP packet processing. Documented in RFC 792.

InterNICInternet Network Information Center.

InternetTerm used to refer to the world’s largest internetwork, connecting thousands of networks worldwide and having a “culture” that focuses on simplicity, research, and standardization based on real life use. Much of today’s leading-edge network technology came from the Internet community. The Internet evolved from ARPANET. Sometimes called the DARPA Internet. Not to be confused with the general term internet. See also ARPANET.

internetShort for internetwork. Not to be confused with the Internet. See internetwork.

Internet addressA 32-bit address assigned to host using TCP/IP. The address is written as 4 octets separated with periods (dotted decimal format) that are made up of a network section, an optional subnet section, and a host section. Also call an IP Address.

InternetworkA collection of networks interconnected by routers that functions (generally) as a single network. Sometimes called the internet, which is not to be confused with the Internet.

IPInternet Protocol. A network layer protocol that contains addressing information and some control information that allows packets to be routed. Documented in RFC 791.

IP multicast

72

Routing technique that allows IP traffic to be propagated from one source to a number of destinations or from many sources to many destinations. Rather than sending one packet to each destination, one packet is sent to a multicast group identified by a single IP destination group address.

ISPInternet Service Provider. A reseller of Internet access services

LANlocal-area network. A network covering a relatively small geographic area (usually not larger than a floor or small building). Compared to WANs, LANs are usually characterized by relatively high data rates and relatively low error rates. See also WAN.

MAE-EAST and MAE-EAST+

MBONEmulticast backbone. The multicast backbone of the Internet.

Multi-homed host.Host attached to multiple physical network segments in an OSI CLNS network.

MOMA BSG order tracking system

NACRNetwork Announcement Change Request.

NAPNetwork Access Point.

NSFNational Science Foundation. A body that funds scientific research in the United States.

NSFnetNational Science Foundation Network. Decommissioned in early 1995, the NSFnet was a large network controlled by the NSF that provided networking services in support of education and research in the United States.

NSPNational Service Provider

NTPNetwork Time Protocol. A TCP/IP protocol that allows time synchronization between a network user’s clock and the network clock.

OE

73

Order Entry.

OFDCOrder for Data Communications.

PBXPrivate Branch Exchange. A device used to manage an entities’ telephone system

PLSCSprint’s Private Line Service Center.

PPPPoint-to-Point Protocol. A successor to SLIP, this protocol provides router-to-router and host-to-network connections over synchronous and asynchronous circuits. See also SLIP.

RFCRequest For Comments. Document series used as the primary means for communicating information about the Internet. Some RFCs are designated by the IAB as “Internet standards.” Most RFCs document protocol specifications such as Telnet and FTP, but some are humorous or historical. RFCs are available on-line from the NIC.

RouterAn OSI network layer (Layer 3) device that can decide which of several paths network traffic will follow based on some optimality metric. Routers forward packets from one network to another, based on network layer information. Occasionally called a gateway (although this definition of gateway is becoming increasingly outdated).

SCASpecial Customer Arrangement. A standard Sprint process for requesting non-standard product offerings for a single customer.

SLIPSerial Line Internet Protocol. A standard for point-to-point serial connections using TCP/IP. Predecessor of PPP. See also PPP.

SMCSprint’s Service Management Center.

SMTPSimple Mail Transfer Protocol. An Internet Protocol providing electronic mail services.

SOL

Sales Office Log.

SSR

74

Sales System Representative.

Static routeA route that is manually entered into the routing table. Static routings take precedence over routes chosen by dynamic routing protocols.

T1Bell system terminology referring to a digital carrier facility used for transmission of data through the telephone hierarchy. The rate of transmission is 1.544 Mbps.

TCP/IPTransmission Control Protocol / Internet Protocol. The common name for the suite of protocols developed by the U.S. DOD in the 1970s to support the construction of worldwide internetworks. TCP and IP are the two best-known protocols in the suite. TCP corresponds to the transport layer (Layer 4) of the OSI reference model and provides reliable transmission of data. IP corresponds to the network layer (Layer 3) of the OSI reference model and provides connectionless datagram service.

UDPUser Datagram Protocol. A connectionless transport layer protocol belonging to the Internet protocol family. UDP adds reliability and multiplexing to IP datagrams.

VBSVirginia Billing System.

WANWide-area network. A network spanning a wide geographic area. See also LAN.

splk product guide - rxn communicationsuffda/archive/internet/sprtlink.doc · web viewtitle: splk...

Documents