📄 rfc1612.txt
字号:
Network Working Group R. AusteinRequest for Comments: 1612 Epilogue Technology CorporationCategory: Standards Track J. Saperia Digital Equipment Corporation May 1994 DNS Resolver MIB ExtensionsStatus of this Memo This document specifies an Internet standards track protocol for the Internet community, and requests discussion and suggestions for improvements. Please refer to the current edition of the "Internet Official Protocol Standards" (STD 1) for the standardization state and status of this protocol. Distribution of this memo is unlimited.Table of Contents 1. Introduction .............................................. 1 2. The SNMPv2 Network Management Framework ................... 2 2.1 Object Definitions ....................................... 2 3. Overview .................................................. 2 3.1 Resolvers ................................................ 3 3.2 Name Servers ............................................. 3 3.3 Selected Objects ......................................... 4 3.4 Textual Conventions ...................................... 4 4. Definitions ............................................... 5 5. Acknowledgements .......................................... 30 6. References ................................................ 30 7. Security Considerations ................................... 32 8. Authors' Addresses ........................................ 321. Introduction This memo defines a portion of the Management Information Base (MIB) for use with network management protocols in the Internet community. In particular, it describes a set of extensions which instrument DNS resolver functions. This memo was produced by the DNS working group. With the adoption of the Internet-standard Network Management Framework [4,5,6,7], and with a large number of vendor implementations of these standards in commercially available products, it became possible to provide a higher level of effective network management in TCP/IP-based internets than was previously available. With the growth in the use of these standards, it has become possible to consider the management of other elements of the infrastructure beyond the basic TCP/IP protocols. A key element ofAustein & Saperia [Page 1]RFC 1612 DNS Resolver MIB May 1994 the TCP/IP infrastructure is the DNS. Up to this point there has been no mechanism to integrate the management of the DNS with SNMP-based managers. This memo provides the mechanisms by which IP-based management stations can effectively manage DNS resolver software in an integrated fashion. We have defined DNS MIB objects to be used in conjunction with the Internet MIB to allow access to and control of DNS resolver software via SNMP by the Internet community.2. The SNMPv2 Network Management Framework The SNMPv2 Network Management Framework consists of four major components. They are: o RFC 1442 which defines the SMI, the mechanisms used for describing and naming objects for the purpose of management. o STD 17, RFC 1213 defines MIB-II, the core set of managed objects for the Internet suite of protocols. o RFC 1445 which defines the administrative and other architectural aspects of the framework. o RFC 1448 which defines the protocol used for network access to managed objects. The Framework permits new objects to be defined for the purpose of experimentation and evaluation.2.1. Object Definitions Managed objects are accessed via a virtual information store, termed the Management Information Base or MIB. Objects in the MIB are defined using the subset of Abstract Syntax Notation One (ASN.1) defined in the SMI. In particular, each object object type is named by an OBJECT IDENTIFIER, an administratively assigned name. The object type together with an object instance serves to uniquely identify a specific instantiation of the object. For human convenience, we often use a textual string, termed the descriptor, to refer to the object type.3. Overview In theory, the DNS world is pretty simple. There are two kinds of entities: resolvers and name servers. Resolvers ask questions. Name servers answer them. The real world, however, is not so simple.Austein & Saperia [Page 2]RFC 1612 DNS Resolver MIB May 1994 Implementors have made widely differing choices about how to divide DNS functions between resolvers and servers. They have also constructed various sorts of exotic hybrids. The most difficult task in defining this MIB was to accommodate this wide range of entities without having to come up with a separate MIB for each. We divided up the various DNS functions into two, non-overlapping classes, called "resolver functions" and "name server functions." A DNS entity that performs what we define as resolver functions contains a resolver, and therefore must implement the MIB groups required of all resolvers which are defined in this module. Some resolvers also implement "optional" functions such as a cache, in which case they must also implement the cache group contained in this MIB. A DNS entity which implements name server functions is considered to be a name server, and must implement the MIB groups required for name servers which are defined in a separate module. If the same piece of software performs both resolver and server functions, we imagine that it contains both a resolver and a server and would thus implement both the DNS Server and DNS Resolver MIBs.3.1. Resolvers In our model, a resolver is a program (or piece thereof) which obtains resource records from servers. Normally it does so at the behest of an application, but may also do so as part of its own operation. A resolver sends DNS protocol queries and receives DNS protocol replies. A resolver neither receives queries nor sends replies. A full service resolver is one that knows how to resolve queries: it obtains the needed resource records by contacting a server authoritative for the records desired. A stub resolver does not know how to resolve queries: it sends all queries to a local name server, setting the "recursion desired" flag to indicate that it hopes that the name server will be willing to resolve the query. A resolver may (optionally) have a cache for remembering previously acquired resource records. It may also have a negative cache for remembering names or data that have been determined not to exist.3.2. Name Servers A name server is a program (or piece thereof) that provides resource records to resolvers. All references in this document to "a name server" imply "the name server's role"; in some cases the name server's role and the resolver's role might be combined into a single program. A name server receives DNS protocol queries and sends DNS protocol replies. A name server neither sends queries nor receives replies. As a consequence, name servers do not have caches. Normally, a name server would expect to receive only those queries to which it could respond with authoritative information. However, if aAustein & Saperia [Page 3]RFC 1612 DNS Resolver MIB May 1994 name server receives a query that it cannot respond to with purely authoritative information, it may choose to try to obtain the necessary additional information from a resolver which may or may not be a separate process.3.3. Selected Objects Many of the objects included in this memo have been created from information contained in the DNS specifications [1,2], as amended and clarified by subsequent host requirements documents [3]. Other objects have been created based on experience with existing DNS management tools, expected operational needs, the statistics generated by existing DNS implementations, and the configuration files used by existing DNS implementations. These objects have been ordered into groups as follows: o Resolver Configuration Group o Resolver Counter Group o Resolver Lame Delegation Group o Resolver Cache Group o Resolver Negative Cache Group o Resolver Optional Counter Group This information has been converted into a standard form using the SNMPv2 SMI defined in [9]. For the most part, the descriptions are influenced by the DNS related RFCs noted above. For example, the descriptions for counters used for the various types of queries of DNS records are influenced by the definitions used for the various record types found in [2].3.4. Textual Conventions Several conceptual data types have been introduced as a textual conventions in the DNS Server MIB document and have been imported into this MIB module. These additions will facilitate the common understanding of information used by the DNS. No changes to the SMI or the SNMP are necessary to support these conventions. Readers familiar with MIBs designed to manage entities in the lower layers of the Internet protocol suite may be surprised at the number of non-enumerated integers used in this MIB to represent values such as DNS RR class and type numbers. The reason for this choice is simple: the DNS itself is designed as an extensible protocol,Austein & Saperia [Page 4]RFC 1612 DNS Resolver MIB May 1994 allowing new classes and types of resource records to be added to the protocol without recoding the core DNS software. Using non- enumerated integers to represent these data types in this MIB allows the MIB to accommodate these changes as well.4. Definitions DNS-RESOLVER-MIB DEFINITIONS ::= BEGIN IMPORTS MODULE-IDENTITY, OBJECT-TYPE, IpAddress, Counter32, Integer32 FROM SNMPv2-SMI TEXTUAL-CONVENTION, RowStatus, DisplayString FROM SNMPv2-TC MODULE-COMPLIANCE, OBJECT-GROUP FROM SNMPv2-CONF dns, DnsName, DnsNameAsIndex, DnsClass, DnsType, DnsQClass, DnsQType, DnsTime, DnsOpCode, DnsRespCode FROM DNS-SERVER-MIB; -- DNS Resolver MIB dnsResMIB MODULE-IDENTITY LAST-UPDATED "9401282250Z" ORGANIZATION "IETF DNS Working Group" CONTACT-INFO " Rob Austein Postal: Epilogue Technology Corporation 268 Main Street, Suite 283 North Reading, MA 10864 US Tel: +1 617 245 0804 Fax: +1 617 245 8122 E-Mail: sra@epilogue.com Jon Saperia Postal: Digital Equipment Corporation 110 Spit Brook Road ZKO1-3/H18 Nashua, NH 03062-2698 US Tel: +1 603 881 0480 Fax: +1 603 881 0120 E-mail: saperia@zko.dec.com" DESCRIPTION "The MIB module for entities implementing the client (resolver) side of the Domain Name System (DNS) protocol."Austein & Saperia [Page 5]RFC 1612 DNS Resolver MIB May 1994 ::= { dns 2 } dnsResMIBObjects OBJECT IDENTIFIER ::= { dnsResMIB 1 } -- (Old-style) groups in the DNS resolver MIB. dnsResConfig OBJECT IDENTIFIER ::= { dnsResMIBObjects 1 } dnsResCounter OBJECT IDENTIFIER ::= { dnsResMIBObjects 2 } dnsResLameDelegation OBJECT IDENTIFIER ::= { dnsResMIBObjects 3 } dnsResCache OBJECT IDENTIFIER ::= { dnsResMIBObjects 4 } dnsResNCache OBJECT IDENTIFIER ::= { dnsResMIBObjects 5 } dnsResOptCounter OBJECT IDENTIFIER ::= { dnsResMIBObjects 6 } -- Resolver Configuration Group dnsResConfigImplementIdent OBJECT-TYPE SYNTAX DisplayString MAX-ACCESS read-only STATUS current DESCRIPTION "The implementation identification string for the resolver software in use on the system, for example; `RES-2.1'" ::= { dnsResConfig 1 } dnsResConfigService OBJECT-TYPE SYNTAX INTEGER { recursiveOnly(1), iterativeOnly(2), recursiveAndIterative(3) } MAX-ACCESS read-only STATUS current DESCRIPTION "Kind of DNS resolution service provided: recursiveOnly(1) indicates a stub resolver. iterativeOnly(2) indicates a normal full service resolver. recursiveAndIterative(3) indicates a full-service resolver which performs a mix of recursive and iterative queries." ::= { dnsResConfig 2 } dnsResConfigMaxCnames OBJECT-TYPE SYNTAX INTEGER (0..2147483647) MAX-ACCESS read-writeAustein & Saperia [Page 6]RFC 1612 DNS Resolver MIB May 1994 STATUS current DESCRIPTION "Limit on how many CNAMEs the resolver should allow before deciding that there's a CNAME loop. Zero means that resolver has no explicit CNAME limit." REFERENCE "RFC-1035 section 7.1." ::= { dnsResConfig 3 } -- DNS Resolver Safety Belt Table dnsResConfigSbeltTable OBJECT-TYPE SYNTAX SEQUENCE OF DnsResConfigSbeltEntry MAX-ACCESS not-accessible STATUS current DESCRIPTION "Table of safety belt information used by the resolver when it hasn't got any better idea of where to send a query, such as when the resolver is booting or is a stub resolver." ::= { dnsResConfig 4 } dnsResConfigSbeltEntry OBJECT-TYPE SYNTAX DnsResConfigSbeltEntry MAX-ACCESS not-accessible STATUS current DESCRIPTION "An entry in the resolver's Sbelt table. Rows may be created or deleted at any time by the DNS resolver and by SNMP SET requests. Whether the values changed via SNMP are saved in stable storage across `reset' operations is implementation-specific." INDEX { dnsResConfigSbeltAddr, dnsResConfigSbeltSubTree, dnsResConfigSbeltClass } ::= { dnsResConfigSbeltTable 1 } DnsResConfigSbeltEntry ::= SEQUENCE { dnsResConfigSbeltAddr IpAddress, dnsResConfigSbeltName DnsName, dnsResConfigSbeltRecursion INTEGER, dnsResConfigSbeltPref INTEGER, dnsResConfigSbeltSubTreeAustein & Saperia [Page 7]RFC 1612 DNS Resolver MIB May 1994 DnsNameAsIndex, dnsResConfigSbeltClass DnsClass, dnsResConfigSbeltStatus RowStatus } dnsResConfigSbeltAddr OBJECT-TYPE SYNTAX IpAddress MAX-ACCESS not-accessible STATUS current DESCRIPTION "The IP address of the Sbelt name server identified by this row of the table." ::= { dnsResConfigSbeltEntry 1 } dnsResConfigSbeltName OBJECT-TYPE SYNTAX DnsName MAX-ACCESS read-create STATUS current DESCRIPTION "The DNS name of a Sbelt nameserver identified by this row of the table. A zero-length string indicates that the name is not known by the resolver." ::= { dnsResConfigSbeltEntry 2 } dnsResConfigSbeltRecursion OBJECT-TYPE SYNTAX INTEGER { iterative(1), recursive(2), recursiveAndIterative(3) } MAX-ACCESS read-create STATUS current DESCRIPTION "Kind of queries resolver will be sending to the name server identified in this row of the table: iterative(1) indicates that resolver will be directing iterative queries to this name server (RD bit turned off). recursive(2) indicates that resolver will be directing recursive queries to this name server (RD bit turned on). recursiveAndIterative(3) indicates that the resolver will be directing both recursive and iterative queries to the server identified in this row of the table." ::= { dnsResConfigSbeltEntry 3 }
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -