Network Working Group Roland Hedberg Internet Draft Bruce Greenblatt Ryan Moats Expires in six months Mark Wahl A Tagged Index Object for use in the Common Indexing Protocol 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. Internet-Drafts may be updated, replaced, or made obsolete by other documents at any time. It is not appropriate to use Internet- Drafts as reference material or to cite them other than as a "working draft" or "work in progress". To learn the current status of any Internet-Draft, please check the 1id-abstracts.txt listing contained in the Internet-Drafts Shadow Direc- tories on ds.internic.net (US East Coast), nic.nordu.net (Europe), ftp.isi.edu (US West Coast), or munnari.oz.au (Pacific Rim). Distribution of this document is unlimited. Abstract This document defines a mechanism by which information servers can exchange indices of information from their databases by making use of the Common Indexing Protocol (CIP). This document defines the structure of the index information being exchanged, as well as a the appropriate meanings for the headers that are defined in the Common Indexing Proto- col. It is assumed that the structures defined here can be used by X.500 DSAs, LDAP servers, Whois++ servers, CCSO servers and many others. Hedberg, Greenblatt, Moats, Wahl [Page 1] Internet Draft January 1998 1. Introduction The Common Indexing Protocol (CIP) as defined in [1] proposes a mechanism for distributing searches across several instances of a single type of search engine with a view to creating a global directory. CIP provides a scalable, flexible scheme to tie individual databases into distributed data warehouses that can scale gracefully with the growth of the Internet. CIP provides a mechanism for meeting these goals that is independent of the access method that is used to access the actual data that underlies the indices. Separate from CIP is the definition of the Index Object that is used to contain the information that is exchanged among Index Servers. One such Index Object that has already been defined is the Centroid that is derived from the Whois++ protocol [2]. The Centroid does not meet all of the requirements for the exchange of index information amongst information servers. For example, it does not support the notion of incremental updates natively. For information servers that contain millions of records in their database, constant exchange of complete dredges of the database is bandwidth intensive. The Tagged Index Object is specifically designed to support the exchange of index update information. This design comes at the cost of an increase in the size of the index object being exchanged. The Centroid is also not tailored to always be able to give boolean answers to queries. In the Centroid Model, "an index server will take a query in standard Whois++ format, search its collections of centroids and other forward information, determine which servers hold records which may fill that query, and then notifies the user's client of the next servers to contact to submit the query." [2] Thus, the exchange of Centroids amongst index servers allows hints to be given as to which information server actually contains the information. The Tagged Index Object labels the various pieces of information with identifiers that tie the individual object attributes back to an object as a whole. This "tag- ging" of information allows an index server to be more capable of directing a specific query to the appropriate information server. Again, this feature is added to the Tagged Index Object at the expense of an increase in the size of the index object. 2. Background The Lightweight Directory Access Protocol (LDAP) is defined in [3], and it defines a mechanism for accessing a collection of information arranged hierarchically in such a manner as to provide a globally Hedberg, Greenblatt, Moats, Wahl [Page 2] Internet Draft January 1998 distributed database which is normally called the Directory Information Tree (DIT). Some distinguishing characteristics of LDAP servers are that it is normally the case that several servers cooperate to manage a common subtree of the DIT. LDAP servers are expected to respond to requests that pertain to portions of the DIT for which they have data, as well as for those portions for which they have no information in their database. For example, the LDAP server for a portion of the DIT in the United States (c=US) must be able to provide a response to a Search operation that pertains to a portion of the DIT in Sweden (c=se). Nor- mally, the response given will be a referral to another LDAP server that is expected to be more knowledgeable about the appropriate subtree. However, there is no mechanism that currently enables these LDAP servers to refer the LDAP client to the supposedly more knowledgeable server. Typically, an LDAP (v3) server is configured with the name of exactly one other LDAP server to which all LDAP clients are referred when their requests fall outside the subtree of the DIT for which that LDAP server has knowledge. This specification defines a mechanism whereby LDAP server can exchange index information that will allow referrals to point towards a clearly accurate destination. While the X.500 series of recommendations defines the Directory Information Shadowing Protocol (DISP) [4] which allows X.500 DSAs to exchange actual information in the DIT. Shadowing allows various infor- mation from various portions of the DIT to be replicated amongst partic- ipating DSAs. The design point of DISP is optimized at the exchange of entire portions of the DIT, whereas the design point of CIP and the Tagged Index Object is optimize at the exchange of structural index information about the DIT, and improving the performance of tree naviga- tion amongst various information servers. The Tagged Index Object is more appropriate for the exchange of index information than is DISP. DISP is more targeted at DIT distribution and fault tolerance. DISP is thus more appropriate for the exchange of the actual data in order to spread the load amongst several information servers. DISP is tailored specifically to X.500 (and other hierarchical directory systems), while the Tagged Index Object and CIP can be used in a wide variety of infor- mation server environments. While DISP allows an individual directory server to collect infor- mation about large parts of the DIT, it would require a huge database to collect all of the replicas for a meaningful portion of the DIT. Fur- thermore, as X.525 states: "Before shadowing can occur, an agreement, covering the conditions under which shadowing may occur is required. Although such agreements may be established in a variety of ways, such as policy statements covering all DSAs within a given DMD ...", where a DMD is a Directory Management Domain. This is due to the case that the actual data in the DIT is being exchanged amongst DSA rather than only Hedberg, Greenblatt, Moats, Wahl [Page 3] Internet Draft January 1998 the information required to maintain an Index. In many environments such an agreement is not appropriate, and in order to collect informa- tion for a meaningful portion of the DIT, a large number of agreements may need to be arranged. 3. Object What is desired is to have an information server (or network of information servers) that can quickly respond to real world requests, like: - What is Tim Howes' email address? This is much harder than, What is Tim Howes at Netscape's email address. - What is the X.509 certificate for Fred Smith at compuserve.com? One certainly doesn't want to search CompuServe's entire directory tree to find out this one piece of information. I also don't want to have to shadow the entire CompuServe directory subtree onto my server. If this request is being made because Fred is trying to log into my server, I'd certainly want to be able to respond to the BIND in real time. - Who are all of the people at Novell that have a title of program- mer? All of these requests can reasonably be translated into LDAP or Whois++, and other directory access protocol queries. They can also be serviced in a straightforward manner by the users home information server if it has the appropriate reference information into the database that contains the source data. In this situation, the first server would be able to "chain" the request on behalf of the user. Alterna- tively, a precise referral could be returned. If the home information server wants to service (i.e chain) the request based on the index information that it has on hand, this servicing could be done by any number of means: - issuing LDAP operations to the remote directory server - issuing DSP operations to the remote directory server - issuing DAP operations to the remote directory server Hedberg, Greenblatt, Moats, Wahl [Page 4] Internet Draft January 1998 - issuing Whois++ operations to the remote Whois++ server - ... 4. The Tagged Index Object This section defines a Tagged Index Object that can be exchanged by Information Servers using CIP. While in many cases it is acceptable for Information Servers to make use of the Centroid construct (as defined in [2]) to exchange index information, the goals in defining a new con- struct are multi-pronged: - When the Information Server receives a search request that warrants that a referral be returned, allow the server to return a referral that will point client to a server that is most likely able to answer the request correctly. False positive referrals (the search turns up hits in the index object that generate referrals to servers that don't hold the desired information) can be reduced, depending on the choice of attribute tokenization types that are used. - When the Information Server receives a search request that is not operating against local data, allow the Information Server itself to "chain" the request to the appropriate remote Information Server. Note that LDAP itself does not define how Chaining works, but X.500 does. This seems very similar to the first "prong". - Finally, when a collection of Information Servers are operating against a large distributed directory, allow them to distribute index information amongst themselves (ala CIP) so that as their own searches can be carried out with some degree of efficiency. 4.1. The Agreement Before a Tagged Index Object can be exchanged, the organization which administers the object supplier and the organization which admin- isters the object consumer must reach an agreement on how the servers will communicate. This agreement contains the following: - "version":The version of the agreement and the index type. This specification describes the index type "x-tagged-index-1" - "dsi": An OID which uniquely identifies the subtree and scope. This field is not explicitly necessary, as it may not provide information beyond that which is contained in the "base-uri" below. Hedberg, Greenblatt, Moats, Wahl [Page 5] Internet Draft January 1998 - "base-uri": One or more URI's which will form the base of any referrals created based upon the index object that is governed by this agreement. For example, in the LDAP URL format [8] the base- uri would specify (among other items): the LDAP host, the base object to which this index object refers (e.g. c=SE), and the scope of the index object (e.g. single container). - "supplier": The hostname and listening port number of the supplier server, as well as any alternative servers holding that same naming contexts, in case the supplier is unavailable. - "consumeraddr": This is a URI of the "mailto:" form, with the RFC 822 email address of the consumer server. Subsequent versions of this draft allow other forms of URI, so that the consumer may retrieve the update via the WWW, FTP or CIP - "updateinterval": The maximum duration in seconds between occu- rances of the supplier server generating an update. If the con- sumer server has not received an update from the supplier server after waiting this long since the previous update, it is likely that the index information is now out of date. A typical value for a server with frequent updates would be 604800 seconds, or every week. Servers whose DITs are only modified annually could have a much longer update interval. - "securityoption": Whether and how the supplier server should sign and encrypt the update before sending it to the consumer server. Options for this version of the specification are: "none" - the update is sent in plaintext "PGP/MIME": the update is digitally signed and encrypted using PGP [9] "S/MIME": the update is digitally signed and encrypted using S/MIME [10] "SSLv3": the update is digitally signed and encrypted using an SSLv3 connection [11] "Fortezza": the update is digitally signed and encrypted using Fortezza [5] It is recommended that the "PGP/MIME" option be used when exchang- ing sensitive information across public networks, and both the supplier and consumer have PGP keys. The "Fortezza" option is intended for use in environments where security protocols are based on Fortezza-compatible devices. The "S/MIME" option can be used with both the supplier and Hedberg, Greenblatt, Moats, Wahl [Page 6] Internet Draft January 1998 consumer have RSA keys and can make use of the PKCS protocols defined in the S/MIME specification. The "SSLv3" option can be used when both the supplier and consumer have access to SSL services, have server certifi- cates, and can mutually authenticate each other. Should these be IANA registered things??? - Security Credentials: The long-term cryptographic credentials used for key exchange and authentication of the consumer and supplier servers, if a security option was selected. For "PGP/MIME", this will be the trusted public keys of both servers. For "Fortezza", this will be the certificate paths of both servers to a common point of trust. For "S/MIME" and "SSLv3" these will be the certifi- cates of the supplier and consumer. Note that if the index server maintains the information that would appear in the agreement in a directory according to the definitions in [7], then no real formal agreement between the two parties needs to be put in place, and the information that is required for communication between the two index servers is derived automatically from the direc- tory. 4.2. Content Type The update consists of a MIME object of type application/cip-index- object. The parameters are: "type": this has value "application/index.obj.tagged". "dsi": the DSI (if any) from the agreement. "base-uri". A set of URIs, separated by spaces. In each URI, the hostname/portno must be distinct, and based on the "supplier" part of the agreement. The payload is mostly textual data but may include bytes with the high bit set. The originating information server should set the con- tent-transfer-encoding as appropriate for the information included in the payload. This object may be encapsulated in a wrapper content (such as mul- tipart/signed) or be encrypted as part of the security procedures. The resulting content can the distributed, for example via electronic mail. For example, From: supplier@sup.com Date: Thu, 16 Jan 1997 13:50:37 -0500 Message-Id: <199701161850.NAA29295@sup.com>; To: consumer@consumer.com <<-- from consumer server address Hedberg, Greenblatt, Moats, Wahl [Page 7] Internet Draft January 1998 Reply-to: supplier-admin@sup.com MIME-Version: 1.0 Content-Type: application/index.obj.tagged; dsi=1.3.6.1.4.1.1466.85.85.1.2.3.4.5.6.7.8.9.10.11.12.13.14.15.16; base-uri="ldap://sup.com/dc=sup,dc=com ldap://alt.com/dc=sup,dc=com" The payload is series of CRLF-terminated lines. The payload only includes characters from a subset of the printable US-ASCII subset of UTF-8. Attribute values that occur outside of this subset are encoded as defined below. As more experience is gained with index objects and UTF-8 data, a future version of this specification may allow for the native transfer of UTF-8 data with requiring this special encoding. No other character sets are permitted by this version of the specification. Some supplier servers may only be able to generate the printable US- ASCII subset, but all consumer servers must be able to handle the full range of Unicode characters when decoding the attribute values (in the "attr-value" field in the BNF below). 4.3. Tagged Index BNF The Tagged Index object has the following grammar, expressed in modified BNF format: index-object = 0*(io-part SEP) io-part io-part = header SEP schema-spec SEP index-info header = version-spec SEP update-type SEP this-update SEP last-update SEP context-size version-spec = "version:" *SPACE "x-tagged-index-1" update-type = "updatetype:" *SPACE ( "total" | "incremental") this-update = "thisupdate:" *SPACE TIMESTAMP last-update = [ "lastupdate:" *SPACE TIMESTAMP ] context-size = [ "contextsize:" *SPACE 1*DIGIT ] schema-spec = "BEGIN IO-Schema" SEP 1*(schema-line SEP) "END IO-Schema" schema-line = attribute-name ":" token-type token-type = "FULL" | "TOKEN" | "RFC822" | "UUCP" | "DNS" index-info = full-index | incremental-index full-index = "BEGIN Index-Info" SEP 1*(index-block SEP) "END Index-Info" incremental-index = 1*(add-block | delete-block | update-block) add-block = "BEGIN Add Block" SEP 1*(index-block SEP) "END Add Block" delete-block = "BEGIN Delete Block" SEP 1*(index-block SEP) "END Delete Block" update-block = "BEGIN Update Block" SEP 1*(index-block SEP) Hedberg, Greenblatt, Moats, Wahl [Page 8] Internet Draft January 1998 "END Update Block" index-block = first-line 0*(SEP cont-line) first-line = attr-name ":" *SPACE taglist "/" attr-value cont-line = "-" taglist "/" attr-value taglist = tag 0*("," tag) tag = 1*DIGIT ["-" 1*DIGIT] attr-value = 0*(UTF8) attr-name = 1*(NAMECHAR) UTF8 = ASCII | "%" HEX HEX TIMESTAMP = 1*DIGIT ASCII = DIGIT | UPPER | LOWER | OTHER NAMECHAR = DIGIT | UPPER | LOWER | "-" | ";" | "." SPACE = ; SEP = (CR LF) | LF CR = ; LF = ; HEX = "a" | "b" | "c" | "d" | "e" | "f" | DIGIT DIGIT = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" UPPER = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" | "J" | "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" | "S" | "T" | "U" | "V" | "W" | "X" | "Y" | "Z" LOWER = "a" | "b" | "c" | "d" | "e" | "f" | "g" | "h" | "i" | "j" | "k" | "l" | "m" | "n" | "o" | "p" | "q" | "r" | "s" | "t" | "u" | "v" | "w" | "x" | "y" | "z" OTHER = "(" | ")" | "+" | "," | "-" | "." | "/" | ":" | "=" | "?" | "@" | ";" | "$" | "_" | "!" | "~" | "*" | "'" | "\" | """ | "#" | "&" | "<" | ">" | "[" | "]" | "^" | "`" | "{" | "|" | "}" Characters that are allowed to appear unescaped in attr-values are the printable subset of (low) ASCII minus the "%" characters, i.e. hex 21 through hex 7e inclusive with the exception of hex 25 (which is the "%" character). Any other UTF-8 encoding of a character that appears in an attr-value must be excaped by using the "%" character and two hex digits that encode the character. For example, The UCS-2 sequence "A." (0041, 2262, 0391, 002E) may be encoded in UTF-8 as follows: 41 E2 89 A2 CE 91 2E If this character sequence appears in an attribute that is in a Tagged Index Object attr-value, then it is encoded as: 41 25 65 32 25 38 39 25 61 32 25 63 65 25 39 31 2E When viewed as an character string the encoding appears as: "A%e2%89%a2%ce%91." Hedberg, Greenblatt, Moats, Wahl [Page 9] Internet Draft January 1998 The set of characters allowed to appear in the attr-name field is limited to the set of characters used in LDAP and WHOIS++ attribute names. For other services that have attribute name character sets that are larger than these, it is suggested that those services create a pro- file that maps the names onto object identifiers, and the sequence of digits and periods is used by those services in creating the attr-name fields for their Tagged Index Objects. Note that the attribute value may only be empty in the case of an incremental update that contains a "Update Block" in which the index object indicates that certain attributes of objects are being removed. This specification only supports the replacement of entire attributes, so that in the case of a multi-valued attribute, all of the values must be specified in the Replace Block, not just the newly added values. The intention of the Tagged Index Object is to supply a snapshot of the cur- rent index of the directory. 4.3.1. Header Descriptions The header section consists of one or more "header lines". The following header lines are defined: "version": This line must always be present, and have the value "x- tagged-index-1" for this version of the specification. "updatetype": This line must always be present. It takes as the value either "total" or "incremental". The first update sent by a supplier server to a consumer server for a DSI must be a "total" update (why?). "thisupdate": This line must always be present. The value is the number of seconds from 00:00:00 UTC January 1, 1970 at which the supplier constructed this update. "lastupdate": This line must be present if the "updatetype" list has the value "incremental". The value is the number of seconds from 00:00:00 UTC January 1, 1970 at which the supplier constructed the previous update sent to the consumer. This field allows the consumer to determine if a previous update was missed. "contextsize": This line may be present at the supplier's option. The value is a number, which is the approximate total number of entries in the subtree. This information is provided for statisti- cal purposes only. Hedberg, Greenblatt, Moats, Wahl [Page 10] Internet Draft January 1998 4.3.2. Tokenization Types The Tagged Index Object inherits the "TOKEN" scheme for tokeniza- tion as specified in [2]. In addition, there are several other tok- enization schemes defined for the Tagged Index Object. The following table presents these schemes and what character(s) are used to delimit tokens. Token Type Tokenization Characters FULL none TOKEN white space, "@" RFC822 white space, ".", "@" UUCP white space, "!" DNS any character note a number, letter, or "-" 4.3.3. Tag Conventions In the tag list, multiple consecutive tags may be shortened by using "#-#". For example, the list "3,4,5,6,7,8,9,10" may be shortened to "3-10". Tags are to be applied to the data on a per entry level. Thus, if two index lines in the same index object contain the same tag, then it is always the case that those two lines refer back to the same "record" in the directory. In LDAP terminology, the two lines would refer back to the same directory object. Additionally if two index lines in the same index object contain different tags, then it is always the case that those two lines refer back to different records in the directory. The tags in the index object are meaningful only in the context of that transmission. The tag applied to the same underlying record in two separate transmissions of a full-update may be different. Thus, receiv- ing index servers should make no assumptions about the values of the tags across index object boundaries. If the recieving index server is implemented in such a way that it maintains a structure similar to the one that exists in the tagged index object with numbered tags attached to various records, then these "internal" tags are distinct from the tags that appear in the index object as created by the transmitting index server. 4.4. Incremental Indexing The tagged index object format supports the ability of information servers to distribute only delta index data, rather than distributing total index information each. This scenario, known as incremental indexing supports three basic types of operations: add, delete and Hedberg, Greenblatt, Moats, Wahl [Page 11] Internet Draft January 1998 replace. If th incremental updatetype is specified in the tagged index object, then the index object contains a snapshot of only the changes that have been made since the index object specified in the lastupdate header was distributed. If the receiving index server did not receive that index object, it should request a total index object. If the CIP protocol supports it, the index server may request the specific index object that it missed. If the tagged index object contains an Add Block, then the lines in the Add Block refer to new records that were added to the information base of the transmitting index server. It can be guaranteed that those records did not exist in any previously received tagged index object, and the receiving index server can insert this index information in the index that it already maintains for the transmitting index server. If the receiving index server is maintaining internal tags, then a new internal tag should be created for each tag in the Add Block. If the tagged index object contains a Delete Block, then the Delete Block contains lines each of which refers to the "key" field (in the attr-name area of the index line) from a record in the information server that has been deleted since the last update (specified in the lastupdate header field). This key field is assumed to be the unique identifier on the transmitting information server for the record that has been deleted. In the case of LDAP servers, this field would have an attr-name of "dn". Other forms of information servers would use the appropriate unique identifier. Thus, the unique identifier must have previously been sent by the transmitting index server. If the receiving index server has never received information for the record refered to by a line in the Delete Block, then it should be ignored, with the proviso that the receiving index server has more than likely "lost" some infor- mation previously distributed by the transmitting index server. If the receiving index server is maintaining internal tags, then after process- ing the Delete Block, the internal tag numbers may be reordered so as to not have "holes" in the sequence. If the tagged index object contains an Update Block, then the lines in the Update Block refer to records that were changed in the informa- tion base of the transmitting index server. As was mentioned in clause 4.3, if any portion of an attribute in the information server has been changed, then the entire attribute must be specified, and all index information from all values of a multi-valued attribute must be speci- fied. If the attribute was removed from the record in the information server, the attribute value specified in the attr-value field should be empty. Attributes which have not been changed in the record are not specified. The Update Block also supports the idea of indexing new attributes which were not previously included in the tagged index object. For example, if the transmitting index server began including index information on postal addresses, then it could include an Update Hedberg, Greenblatt, Moats, Wahl [Page 12] Internet Draft January 1998 Block in the index object that included all of the index information on postal addresses for all records in its information base, and indicate that nothing else has changed. If the receiving index server is main- taining internal tags, then after processing the Update Block, the internal tag numbers should remain the same. 5. Example As an example, the following LDIF [6] entries and the resulting Tagged Index Object are presented. dn: cn=Barbara Jensen, ou=Product Development, o=Ace Industry, c=US objectclass: top objectclass: person objectclass: organizationalPerson cn: Barbara Jensen cn: Barbara J Jensen cn: Babs Jensen sn: Jensen uid: bjensen telephonenumber: +1 408 555 1212 description: A big sailing fan. dn: cn=Bjorn Jensen, ou=Accounting, o=Ace Industry, c=US objectclass: top objectclass: person objectclass: organizationalPerson cn: Bjorn Jensen sn: Jensen telephonenumber: +1 408 555 1212 dn: cn=Gern Jensen, ou=Product Testing, o=Ace Industry, c=US objectclass: top objectclass: person objectclass: organizationalPerson cn: Gern Jensen cn: Gern O Jensen sn: Jensen uid: gernj telephonenumber: +1 408 555 1212 dn: cn=Horatio Jensen, ou=Product Testing, o=Ace Industry, c=US objectclass: top objectclass: person objectclass: organizationalPerson cn: Horatio Jensen cn: Horatio N Jensen sn: Jensen uid: hjensen telephonenumber: +1 408 555 1212 Hedberg, Greenblatt, Moats, Wahl [Page 13] Internet Draft January 1998 The Tagged Index Object for this example would be: version: x-tagged-index-1 updatetype: total thisupdate: 855938804 BEGIN IO-Schema dn: FULL ou: TOKEN o: TOKEN c: TOKEN objectclass: FULL cn: TOKEN sn: FULL uid: FULL title: TOKEN END IO-Schema BEGIN Index-Info dn: 1/cn=Barbara Jensen,ou=Product Development,o=Ace Industry,c=US -2/cn=Bjorn Jensen,ou=Accounting,o=Ace Industry,c=US -3/cn=Gern Jensen,ou=Product Testing,o=Ace Industry,c=US -4/cn=Horatio Jensen,ou=Product Testing,o=Ace Industry,c=US ou: 1,3-4/Product -1/Development -2/Accounting -3-4/Testing o: */Ace -*/Industry c: */US objectclass: */top -*/person -*/organizationalPerson cn: 1/Barbara -1/J -1/Babs -*/Jensen -2/Bjorn -3/Gern -3/O -4/Horatio -4/N sn: */Jensen uid: 1/bjensen -3/gernj -4/hjensen title: 1/product 1/manager 1/rod 1/and Hedberg, Greenblatt, Moats, Wahl [Page 14] Internet Draft January 1998 1/reel 1/division END Index-Info As an example of the Incremental Index Object, consider an update that occurs when Barbara Jensen's entry above changes to: dn: cn=Barbara Jensen-Smith, ou=Product Development, o=Ace Industry, c=US objectclass: top objectclass: person objectclass: organizationalPerson cn: Barbara Jensen-Smith cn: Barbara J Jensen-Smith cn: Babs Jensen-Smith sn: Jensen-Smith uid: bjensen telephonenumber: +1 408 555 1212 description: A big sailing fan. The Tagged Index Object for this example would be: version: x-tagged-index-1 updatetype: incremental lastupdate: 855940000 thisupdate: 855938804 BEGIN IO-schema dn: FULL cn: TOKEN sn: FULL END IO-Schema BEGIN Delete Block dn: 1/cn=Barbara Jensen,ou=Product Development,o=Ace Industry,c=US cn: 1/Jensen sn: 1/Jensen END Delete Block BEGIN Add Block dn: 1/cn=Barbara Jensen-Smith,ou=Product Development,o=Ace Industry,c=US cn: 1/Jensen-Smith sn: 1/Jensen-Smith END Add Block In this next example, consider an LDIF file containing a series of change records and comments. # Add a new entry dn: cn=Fiona Jensen, ou=Marketing, o=Ace Industry, c=US changetype: add objectclass: top Hedberg, Greenblatt, Moats, Wahl [Page 15] Internet Draft January 1998 objectclass: person objectclass: organizationalPerson cn: Fiona Jensen sn: Jensen uid: fiona telephonenumber: +1 408 555 1212 jpegphoto:< /usr/local/directory/photos/fiona.jpg # Delete an existing entry dn: cn=Robert Jensen, ou=Marketing, o=Ace Industry, c=US changetype: delete # Modify an entry's relative distinguished name dn: cn=Paul Jensen, ou=Product Development, o=Ace Industry, c=US changetype: modrdn newrdn: cn=Paula Jensen deleteoldrdn: 1 # Rename and entry and move all of its children to a new location in # the directory tree (only implemented by LDAPv3 servers). dn: ou=PD Accountants, ou=Product Development, o=Ace Industry, c=US changetype: modrdn newrdn: ou=Product Development Accountants deleteoldrdn: 0 newsuperior: ou=Accounting, o=Ace Industry, c=US # Modify an entry: add an additional value to the postaladdress attribute, # completely delete the description attribute, replace the telephonenumber # attribute with two values, and delete a specific value from the # facsimiletelephonenumber attribute dn: cn=Paula Jensen, ou=Product Development, o=Ace Industry, c=US changetype: modify add: postaladdress postaladdress: 123 Anystreet $ Sunnyvale, CA $ 94086 - delete: description - replace: telephonenumber telephonenumber: +1 408 555 1234 telephonenumber: +1 408 555 5678 - delete: facsimiletelephonenumber facsimiletelephonenumber: +1 408 555 9876 - The Tagged Index Object for this example would be: version: x-tagged-index-1 updatetype: incremental thisupdate: 855938804 lastupdate: 855912345 BEGIN IO-Schema Hedberg, Greenblatt, Moats, Wahl [Page 16] Internet Draft January 1998 dn: FULL ou: TOKEN o: TOKEN c: TOKEN objectclass: FULL cn: TOKEN sn: FULL uid: FULL title: TOKEN END IO-Schema BEGIN Add Block objectclass: top objectclass: person objectclass: organizationalPerson c: 1/us o: 1/Ace o: 1/Industry ou: 1/Marketing cn: 1/Fiona cn: 1/Jensen sn: 1/Jensen uid: 1/Fiona END Add Block BEGIN Delete Block dn: 1/cn=Robert Jensen, ou=Marketing, o=Ace Industry, c=us END Delete Block BEGIN Update Block dn: 1/ou=PD Accountants, ou=Product Development, o=Ace Industry, c=US -2/cn=Paula Jensen, ou=Product Development, o=Ace Industry, c=US rdn: 1/Product Development Accountants description: 2/ telephonenumber: 2/+1 408 555 5678 facsimilenumber: 2/ postaladdress: 2/123 -2/AnyStreet -2/Sunnyvale -2/CA -2/94086 END Update Block END Index-Info 6. Aggregation Hedberg, Greenblatt, Moats, Wahl [Page 17] Internet Draft January 1998 6.1. Aggregation of Tagged Index Objects Aggregation of two tagged index objects is done by merging the two lists of values and rewriting each tag list. The tag list rewriting process is done so that the resulting index object appears as if it came from a single source. Tags from one of the two tagged index objects are "mapped" to the number space above that used by the other tagged index object. An index server that aggregates tagged index objects for export MUST ensure that the export URL (i.e. the base-uri of the CIP object) for the aggregate index object will route all queries that have "hits" on the index object to that server (otherwise, query routing will not succeed). 7. Recommendations TBD 8. Security Considerations This specification provides a protocol for transfering information between two servers. The actual information transfered may be protected by laws in many countries, so care must be taken in the methods used to tokenize the data in order to ensure that protected data may not be reconstructed in full by the receiving server. This protocol does not have any inherent protection against spoofing or eavesdropping. How- ever, since this protocol is transported in MIME messages (as are all CIP index objects), it inherits all of the security capabilities and liabilities of other MIME messages. Specifically, those wanting to pre- vent eavesdropping or spoofing may use some of the various techniques for signing and encrypting MIME messages. Information Server administrators must decide what portions of their databases are appropriate for inclusion in the Tagged Index Object. For distribution of information outside of the enterprise, information server developers are encouraged to allow for facilities that hide the organizational structure when generating the Tagged Index Object from the underlying information database. In order to allow for the secure transmission of Tagged Index Objects across the Internet, Index Servers should make use of SSL to carry out the connection. In order to strongly verify the identity of the peer index server on the other side of the connection, SSL version 3 certificate exchange should be implemented, and the identity in the peer's certificate verify with the Public Key Infrastructure. If electronic mail is used to exchange the Tagged Index Objects, then a secure messaging facility, such as PGP/MIME or S/MIME should be used to sign or encrypt (or both) the Hedberg, Greenblatt, Moats, Wahl [Page 18] Internet Draft January 1998 information. 9. References [1] J. Allen, M. Mealling, "The Architecture of the Common Indexing Protocol (CIP)," Internet Draft (work in progress) June 1997. [2] C. Weider, J. Fullton, S. Spero, "Architecture of the Whois++ Index Service. RFC 1913, February 1996. [3] M. Wahl, T. Howes, S. Kille, "Lightweight Directory Access Protocol (v3)," Internet Draft (work in progress), June 1997. [4] ITU, "X.525 Information Technology - Open Systems Interconnection - The Directory: Replication", November 1993. [5] "FORTEZZA Application Implementors Guide for the FORTEZZA Crypto Card (Production Version)", Document #PD4002102-1.01, SPYRUS, 1995. [6] The LDAP Data Interchange Format (LDIF). Internet Draft (work in progress), 25 November 1996. [7] R. Hedberg, "LDAPv2 client Vs the Index Mesh". Internet Draft (work in progress), November 1997. [8] T. Howes, M. Smith, "The LDAP URL Format". Internet Draft (work in progress), June 1997. [9] M. Elkins, "MIME Security with Pretty Good Privacy (PGP)", RFC2015, October 1996. [10] Blake Ramsdell, "S/MIME Version 3 Message Specification", Internet Draft, (work in progress), May 1997. [11] C. Allen, T. Dierks, "The TLS Protocol Version 1.0", Internet Draft, (work in progress), November 1997. 10. Author's Addresses Hedberg, Greenblatt, Moats, Wahl [Page 19] Internet Draft January 1998 Roland Hedberg Umdac Umea University 901 87 Umea Sweden Email: Roland.Hedberg@umdac.umu.se Bruce Greenblatt RSA Data Security 100 Marine Parkway Suite 500 Redwood City, CA 94065 USA Email: bgreenblatt@rsa.com Phone: +1-650-595-8782 Ryan Moats AT&T 15621 Drexel Circle Omaha, NE 68135-2358 USA EMail: jayhawk@ds.internic.net Phone: +1 402 894-9456 Mark Wahl Critical Angle, Inc. 4815 W Braker Lane #502-385 Austin, TX 78759 Email: M.Wahl@critical-angle.com Hedberg, Greenblatt, Moats, Wahl [Page 20] Internet Draft January 1998 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3. Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4. The Tagged Index Object . . . . . . . . . . . . . . . . . . . . . 5 4.1. The Agreement . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.2. Content Type . . . . . . . . . . . . . . . . . . . . . . . . . 7 4.3 Tagged Index BNF . . . . . . . . . . . . . . . . . . . . . . . . 8 4.3.1. Header Descriptions . . . . . . . . . . . . . . . . . . . . . 10 4.3.2. Tokenization types . . . . . . . . . . . . . . . . . . . . . 11 4.3.3. Tag Conventions . . . . . . . . . . . . . . . . . . . . . . . 11 4.4. Incremental Indexing . . . . . . . . . . . . . . . . . . . . . 11 5. Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 6. Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 6.1 Aggregation of Tagged Index Objects . . . . . . . . . . . . . . 18 7. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . 18 8. Security Considerations . . . . . . . . . . . . . . . . . . . . . 18 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 10. Author's Addresses . . . . . . . . . . . . . . . . . . . . . . . 19 Hedberg, Greenblatt, Moats, Wahl [Page 21]