INTERNET-DRAFT November 16, 1998 Expires in six month K. Tsuchiya, Hitachi H. Higuchi, Hitachi Y. Atarashi, Hitachi Dual Stack Hosts using the "Bump-in-the-Stack" Technique Status of this Memo This document is an Internet-Draft. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other docu- ments at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as ``work in pro- gress.'' To learn the current status of any Internet-Draft, please check the ``1id-abstracts.txt'' listing contained in the Internet-Drafts Sha- dow Directories on ftp.ietf.org (US East Coast), nic.nordu.net (Europe), ftp.isi.edu (US West Coast), or munnari.oz.au (Pacific Rim). Abstract Especially in the early stage of the migration from IPv4 to IPv6, it is hard to prepare IPv6 applications completely. This memo pro- poses a mechanism of dual stack hosts using the technique called "Bump-in-the-Stack" in the IP security area. The mechanism enables the hosts to communicate with other IPv6 hosts using IPv4 legacy applications. Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 1] INTERNET-DRAFT November 1998 1. Introduction RFC1933 [TRANS-MECH] proposed mechanisms to migrate from IPv4 [IPv4] to IPv6 [IPv6], including dual stack and tunneling, for the early stage. Accordingly, hosts and routers are developed for the IPv6 migration. But there are few applications for IPv6 compared to IPv4, where a huge number of applications are available. In order to advance the migration to IPv6 smoothly, it is highly desirable to increase the availability of IPv6 applications to the same level as IPv4. But unfortunately this is expected to take a very long time. This memo proposes a mechanism of dual stack hosts using the tech- nique called "Bump-in-the-Stack" in the IP security area. The tech- nique inserts modules into the hosts which snoop data that flows between a TCP/IPv4 module and network card driver modules, and translate IPv4 into IPv6 and vice versa. It enables the hosts to communicate with other IPv6 hosts using IPv4 legacy applications; thus making it seem as if the hosts have applications for both IPv4 and IPv6. This document uses the words defined in [IPV4], [IPV6], and [TRANS-MECH]. 2. Components Dual stack hosts defined in RFC1933 [TRANS-MECH] need applications, TCP/IP modules and addresses for both IPv4 and IPv6. The proposed hosts in this memo have 3 modules instead of IPv6 applications, and communicate with other IPv6 hosts using IPv4 applications. The 3 modules are a translator, an extension name resolver and an address mapper. Figure 1 illustrates a host which has the 3 modules described above installed. Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 2] INTERNET-DRAFT November 1998 The proposed dual stack host +----------------------------------------------------------+ | +----------------------------------------------------+ | | | IPv4 applications | | | +----------------------------------------------------+ | | +----------------------------------------------------+ | | | TCP/IPv4 | | | | +-------------------------------------------+ | | | | +-----------+ +---------+ +------------+ | | | | | extension | | address | | translator | | | | | | name | | mapper | +------------+ | | | | | resolver | | | +------------+ | | | | | | | | | IPv6 | | | +--------+ +-----------+ +---------+ +------------+ | | +----------------------------------------------------+ | | | Network card drivers | | | +----------------------------------------------------+ | +----------------------------------------------------------+ +----------------------------------------------------------+ | Network cards | +----------------------------------------------------------+ Figure 1 2.1 Translator The translator translates IPv4 into IPv6 and vice versa using the IP conversion mechanism defined in [SIIT]. When receiving IPv4 packets from IPv4 applications, the translator converts IPv4 packet headers into IPv6 packet headers, then frag- ments the IPv6 packets (because header length of IPv6 is typically 20 bytes larger than that of IPv4), and sends them to IPv6 net- works. When receiving IPv6 packets from the IPv6 networks, it works symmetrically to the previous case, except that there is no need to fragment the packets. 2.2 Extension Name Resolver The extension name resolver returns a "proper" answer in response to the IPv4 application's request. The application typically sends a query to its name server to resolve 'A' records for the target host name. The translator snoops the query, then creates another query to resolve both 'A' and Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 3] INTERNET-DRAFT November 1998 'AAAA' records for the host name, and sends the query to the server. If the 'A' record is resolved, it returns the 'A' record to the application. In this case, there is no need for translation by the translator above. If only the 'AAAA' record is available, it requests the mapper to assign an IPv4 address corresponding to the IPv6 address. Then it creates the 'A' record for the assigned IPv4 address and returns the 'A' record to the application. 2.3 Address mapper The address mapper maintains an IPv4 address spool. The spool, for example, consists of private addresses [PRIVATE]. Also, it main- tains pairs consisting of an IPv4 address and an IPv6 address in a table. When the resolver or the translator requests the mapper to assign an IPv4 address for an IPv6 address, it selects and returns an IPv4 address out of the spool, and then registers a new entry into the table dynamically. The registration occurs in the following 2 cases: (1) When the resolver gets only an 'AAAA' record for the target host name and there is not a mapping entry for the IPv6 address. (2) When the translator receives an IPv6 packet and there is not a mapping entry for the IPv6 source address. NOTE: There is one exception to above. When initializing the table, it registers a pair of its own IPv4 address and IPv6 address into the table statically. 3. Action Examples This section describes action of the proposed dual stack host called "dual stack," which communicates with an IPv6 host called "host6" using an IPv4 application. 3.1 Originator behavior This subsection describes the originator behavior of "dual stack." The communication is triggered by "dual stack." Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 4] INTERNET-DRAFT November 1998 The application sends a query to its name server to resolve 'A' records for "host6." The resolver snoops the query, and then creates another query to resolve both 'A' and 'AAAA' records for the host name and sends it to the server. In the case, only the 'AAAA' record is resolved, so the resolver requests the mapper to assign an IPv4 address corresponding to the IPv6 address. NOTE: In the case of communication with an IPv4 host, the 'A' record is resolved. The resolver then returns it to the applica- tion, and there is no need for translation as follows. The mapper selects an IPv4 address out of the spool and returns it to the resolver. The resolver creates the 'A' record for the assigned IPv4 address and returns it to the application. NOTE: See subsection 4.3 about influence on other hosts caused by the assigned IPv4 address. The application sends an IPv4 packet to "host6." The IPv4 packet reaches the translator. The translator tries translating the IPv4 packet into an IPv6 packet but does not know how to translate the IPv4 destination address and the IPv4 source address. So the translator requests the mapper to provide mapping entries for them. The mapper checks its mapping table and finds entries for each of them, and then returns the IPv6 destination address and the IPv6 source address to the translator. NOTE: The mapper will register its own IPv4 address and IPv6 address into the table beforehand. See subsection 2.3. The translator translates the IPv4 packet into an IPv6 packet then fragments the IPv6 packet if necessary and sends it to an IPv6 net- work. The IPv6 packet reaches "host6." Then "host6" sends a new IPv6 packet to "dual stack." Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 5] INTERNET-DRAFT November 1998 The IPv6 packet reaches the translator of "dual stack." The trans- lator gets mapping entries for the IPv6 destination address and the IPv6 source address from the mapper in the same way as before. Then the translator translates the IPv6 packet into an IPv4 packet and tosses it up to the application. The following diagram illustrates the action described above: "dual stack" "host6" IPv4 TCP/ extension address translator IPv6 appli- IPv4 name mapper cation resolver | | | | | | | <> | | | | | | | | | |------|------>| Query of 'A' records for "host6." | Name | | | | | | | Server | | |---------|-------|-----------|------|------>| | | | Query of 'A' and 'AAAA' records for "host6." | | | | | | | | | | |<--------|-------|-----------|------|-------| | | | Reply only with the 'AAAA' record.| | | | | | | | | | |<> | | | | | | | | | |-------->| Request an IPv4 address | | | | | corresponding to the IPv6 address. | | | | | | | | | | |<> | | | | | | | | | |<--------| Reply with the IPv4 address. | | | | | | | | | |<> | | | | | | | |<-----|-------| Reply with the 'A' record. | | | | | | | | | Figure 2(1/2) Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 6] INTERNET-DRAFT November 1998 "dual stack" "host6" IPv4 TCP/ extension address translator IPv6 appli- IPv4 name mapper cation resolver | | | | | | | <>| | | | | | | | | | |======|=======|=========|======>| An IPv4 packet. | | | | | | | | | | | |<------| Request IPv6 addresses | | | | | corresponding to the IPv4 | | | | | addresses. | | | | | | | | | | | |------>| Reply with the IPv6 | | | | | addresses. | | | | | | | | | | | |<> | | | | | | | | | |An IPv6 packet. |===========|=====>| | | | | | | | | | | | <> | | | | | | | | | | |An IPv6 packet. |<==========|======| | | | | | | | | | | | |<> | | | | | | | |<=====|=======|=========|=======| An IPv4 packet. | | | | | | | | Figure 2(2/2) 3.2 Recipient behavior This subsection describes the recipient behavior of "dual stack." The communication is triggered by "host6." "host6" resolves the 'AAAA' record for "dual stack" through its name server, and it then sends an IPv6 packet to the resolved IPv6 address. The IPv6 packet reaches the translator of "dual stack." The trans- lator tries translating the IPv6 packet into an IPv4 packet but does not know how to translate the IPv6 destination address and the IPv6 source address. So the translator requests the mapper to Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 7] INTERNET-DRAFT November 1998 provide mapping entries for them. The mapper checks its mapping table with each of them and finds a mapping entry for the IPv6 destination address. NOTE: The mapper will register its own IPv4 address and IPv6 address into the table beforehand. See subsection 2.3. But there is not a mapping entry for the IPv6 source address, so the mapper selects an IPv4 address out of the spool for it, and then returns the IPv4 destination address and the IPv4 source address to the translator. NOTE: See subsection 4.3 about influence on other hosts caused by the assigned IPv4 address. The translator translates the IPv6 packet into an IPv4 packet and tosses it up to the application. The application sends a new IPv4 packet to "host6." The following behavior is the same as that described in subsection 3.1. The following diagram illustrates the action described above: Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 8] INTERNET-DRAFT November 1998 "dual stack" "host6" IPv4 TCP/ extension address translator IPv6 appli- IPv4 name mapper cation resolver | | | | | | | <> | | | | | | | | | | | |An IPv6 packet. |<==========|======| | | | | | | | | | | |<------| Request IPv4 addresses | | | | | corresponding to the IPv6 | | | | | addresses. | | | | | | | | | | |------>| Reply with the IPv4 | | | | | addresses. | | | | | | | | | | | | |<> | | | | | | | |<=====|=======|=========|=======| An IPv4 packet. | | | | | | | | <> | | | | | | | | | |======|=======|=========|======>| An IPv4 packet. | | | | | | | | | | | | |<> | | | | | | | | | |An IPv6 packet. |===========|=====>| | | | | | | | Figure 3 4. Considerations This section considers some issues with the proposed dual stack hosts. 4.1 IP conversion In common with NAT [NAT], IP conversion needs to translate IP addresses embedded in application layer protocols, which are typi- cally found in FTP [FTP]. So it is hard to translate all such applications completely. Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 9] INTERNET-DRAFT November 1998 4.2 IPv4 address spool and mapping table The spool, for example, consists of private addresses [PRIVATE]. So a large address space can be used for the spool. Nonetheless, IPv4 addresses in the spool may be exhausted and cannot be assigned to IPv6 target hosts if the host communicates with great many other IPv6 hosts and the mapper never frees entries registered into the mapping table once. To solve the problem, for example, it is desir- able for the mapper to free the oldest entry in the mapping table and re-use the IPv4 address for creating a new entry. 4.3 Internally assigned IPv4 addresses IPv4 addresses, which are internally assigned to IPv6 target hosts out of the spool, never flow out from the host, and so do not nega- tively affect other hosts. 5. References [SIIT] Erik Nordmark, "Stateless IP/ICMP Translator (SIIT)", Inernet-Draft, Work in Progress, November 1998. [IPV4] J. Postel, "Internet Protocol", RFC 791, September 1981. [FTP] J. Postel, J.K. Reynolds, "File Transfer Protocol", RFC959, October 1985. [NAT] Kjeld Borch Egevang and Paul Francis, "The IP Network Address Translator (NAT)", RFC1631, May 1994. [IPV6] S. Deering and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 1883, January 1996. [PRIVATE] Y. Rekhter, B. Moskowitz, D. Karrenberg, G. J. de Groot and E. Lear, "Address Allocation for Private Internets", RFC1918, February 1996. [TRANS-MECH] R. Gilligan and E. Nordmark, "Transition Mechanisms for IPv6 Hosts and Routers", RFC 1933, April 1996. [BUMP] D.A. Wagner and S.M. Bellovin, "A Bump in the Stack Encryptor for MS-DOS Systems", The 1996 Symposium on Network and Distributed Systems Security (SNDSS'96) Proceedings. Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 10] INTERNET-DRAFT November 1998 6. Acknowledgments The authors gratefully acknowledge the many helpful suggestions of the members of the WIDE Project, Kazuhiko YAMAMOTO, Jun MURAI, Munechika SUMIKAWA, Ken WATANABE, and Takahisa MIYAMOTO. 7. Author's Addresses Kazuaki TSUCHIYA Server & Network Development Division, Hitachi, Ltd. 810 Shimoimaizumi, Ebina-shi, Kanagawa-ken, 243-0435 JAPAN Phone: +81-462-32-2111 Fax: +81-462-35-8325 Email: tsuchi@ebina.hitachi.co.jp Hidemitsu HIGUCHI Server & Network Development Division, Hitachi, Ltd. 810 Shimoimaizumi, Ebina-shi, Kanagawa-ken, 243-0435 JAPAN Phone: +81-462-32-2111 Fax: +81-462-35-8325 Email: h-higuti@ebina.hitachi.co.jp Yoshifumi ATARASHI Server & Network Development Division, Hitachi, Ltd. 810 Shimoimaizumi, Ebina-shi, Kanagawa-ken, 243-0435 JAPAN Phone: +81-462-32-2111 Fax: +81-462-35-8325 Email: atarashi@ebina.hitachi.co.jp Tsuchiya draft-ietf-ngtrans-dual-stack-hosts-00.txt [Page 11]