ALTO S. Kiesel Internet-Draft University of Stuttgart Intended status: Standards Track M. Stiemerling Expires: January 17, 2013 NEC Europe Ltd. N. Schwan M. Scharf Alcatel-Lucent Bell Labs H. Song Huawei July 16, 2012 ALTO Server Discovery draft-ietf-alto-server-discovery-04 Abstract The goal of Application-Layer Traffic Optimization (ALTO) is to provide guidance to applications, which have to select one or several hosts from a set of candidates that are able to provide a desired resource. Entities seeking guidance need to discover and possibly select an ALTO server to ask. This is called ALTO server discovery. This memo describes an ALTO server discovery mechanism based on several alternative mechanisms that are applicable when the resource consumer is co-located with the ALTO client. Kiesel, et al. Expires January 17, 2013 [Page 1] Internet-Draft ALTO Server Discovery July 2012 Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119]. Status of this Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on January 17, 2013. Copyright Notice Copyright (c) 2012 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Kiesel, et al. Expires January 17, 2013 [Page 2] Internet-Draft ALTO Server Discovery July 2012 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.1. Discovery Scenarios . . . . . . . . . . . . . . . . . . . 5 1.1.1. ALTO Server Discovery by Resource Consumers . . . . . 6 1.1.2. ALTO Server Discovery by a Third Party . . . . . . . . 7 1.2. Pre-Conditions . . . . . . . . . . . . . . . . . . . . . . 8 2. Protocol Overview . . . . . . . . . . . . . . . . . . . . . . 10 3. Retrieving the URI by U-NAPTR . . . . . . . . . . . . . . . . 12 3.1. Retrieving the Domain Name . . . . . . . . . . . . . . . . 12 3.1.1. Option 1: User input . . . . . . . . . . . . . . . . . 12 3.1.2. Option 2: DHCP . . . . . . . . . . . . . . . . . . . . 13 3.1.3. Option 3: PPP . . . . . . . . . . . . . . . . . . . . 13 3.2. U-NAPTR Resolution . . . . . . . . . . . . . . . . . . . . 14 4. Applicability . . . . . . . . . . . . . . . . . . . . . . . . 16 4.1. Applicability for Resource Consumer Server Discovery . . . 16 4.2. Applicability for Third Party Server Discovery . . . . . . 16 5. Deployment Considerations . . . . . . . . . . . . . . . . . . 18 5.1. DHCP option for DNS Suffix . . . . . . . . . . . . . . . . 18 5.2. PPP option for DNS Suffix . . . . . . . . . . . . . . . . 18 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 6.1. Registration of PPP IPCP Configuration Option Type . . . . 19 6.2. Registration of U-NAPTR application service tag . . . . . 19 7. Security Considerations . . . . . . . . . . . . . . . . . . . 20 7.1. General . . . . . . . . . . . . . . . . . . . . . . . . . 20 7.2. For U-NAPTR . . . . . . . . . . . . . . . . . . . . . . . 20 8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . 22 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 23 9.1. Normative References . . . . . . . . . . . . . . . . . . . 23 9.2. Informative References . . . . . . . . . . . . . . . . . . 23 Appendix A. Contributors List and Acknowledgments . . . . . . . . 25 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 26 Kiesel, et al. Expires January 17, 2013 [Page 3] Internet-Draft ALTO Server Discovery July 2012 1. Introduction The goal of Application-Layer Traffic Optimization (ALTO) is to provide guidance to applications, which have to select one or several hosts from a set of candidates, that are able to provide a desired resource [RFC5693]. The requirements for ALTO are itemized in [I-D.ietf-alto-reqs]. ALTO is realized by a client-server protocol. ALTO clients send queries to ALTO servers, in order to solicit guidance. Hence, ALTO clients need to know the contact information of ALTO servers, which can provide appropriate guidance for a given resource consumer. Typically the closer an ALTO server is to a resource consumer the more accurate guidance it can provide. Thus a design objective is to automatically discover an ALTO server topologically close to the network attachment point of the resource consumer, if available. The contact information of the ALTO server is retrieved by invoking the ALTO discovery procedure defined in this document. The ALTO protocol specification [I-D.ietf-alto-protocol] is based on HTTP. Therefore, it expects that the ALTO discovery procedure yields the HTTP(S) URI of the ALTO server's Information Resource Directory, which gives further information about the capabilities and services provided by that ALTO server. Further (DNS) lookups may be necessary in order to find out the ALTO server's IP address. The goal of this memo is to propose a discovery mechanism for ALTO client deployments in uncontrolled environments that is implementable and deployable at a fast pace, i.e., without creating other deployment dependencies for ALTO. We propose a schema which employs the U-NAPTR mechanism [RFC4848] to determine the URI of the ALTO server and where mutliple input methods to the U-NAPTR process can be used. U-NAPTR is used because the discovery mechanism must return an URI, and thus other discovery mechanisms are not applicable (e. g., DNS SRV records). There are various architectural options where to place the ALTO client and the ALTO server discovery procedure: o One option is that the ALTO client and the ALTO server discovery procedure are embedded directly in the resource consumer, i.e., the application protocol entity that will eventually initiate data transmission to/from the selected resource provider(s). In this case, the ALTO server discovery procedure might be able to interact with the user (i.e., prompt for a host name). Furthermore, it may use services such as DHCP, which are only available within the access network to which the resource consumer is connected. This option is described in this memo. Kiesel, et al. Expires January 17, 2013 [Page 4] Internet-Draft ALTO Server Discovery July 2012 o A similar architectural setup exists in controlled environments, e.g. a CDN that uses ALTO as information base for request redirection. However in such a controlled environment it is unlikely that a dynamic server discovery is necessary. Instead it is expected that the CDN administrator will manually configure the contact information of the ALTO server. Thus this draft is focused on allowing dynamic discovery in uncontrolled environments where provisioning contact information is impossible or undesireable. o Another option is to integrate the ALTO client and the ALTO server discovery procedure into a third party such as a resource directory ("peer-to-peer tracker"), which issues ALTO queries on behalf of various resource consumers. This third party may reside in a different part of the network (administrative domain) than the resource consumer. It may occur that said third party whishes to issue ALTO queries on behalf of a resouce consumer, but all it knows about the resource consumer is the source IP address of messages originating from it (i.e., the resource consumer's IP address or the "public" IP address of the outermost NAT in front of the resource consumer). A general server discovery procedure based solely on this IP address raises several issues, and are out of scope of this document. This option is further discussed in [I-D.kist-alto-3pdisc]. A more detailed discussion of various options where to place the funcional entities comprising the overall ALTO architecture can be found in [I-D.ietf-alto-deployments]. A general mechanism that redirects ALTO clients from one ALTO server to another, potentially closer, ALTO server raises several issues. First ALTO servers by definition provide Network Maps for the whole IP address space and thus can provide each client with a potentially useful answer. Second ALTO servers might be deployed independently by seperate administrative entities and are thus not necessarily aware of each other. The use of a redirection mechanism thus depends on the deployment scenario and is for this reason out of the scope of this document. Comments and discussions about this memo should be directed to the ALTO working group: alto@ietf.org. 1.1. Discovery Scenarios Figure 1 below shows an overview on the different entities of a generic ALTO framework. The ALTO Server discovery mechanism is used by the peer-to-peer (P2P) application in order retrieve the point of contact of the ALTO Service. Kiesel, et al. Expires January 17, 2013 [Page 5] Internet-Draft ALTO Server Discovery July 2012 +------+ +-----+ | Peers +-----+ +------+ +=====| |--+-oo | |......| |====+ oo+--*--+ o +-----+ +------+ | o * ooooooo Source of ALTO | o * Topological Service | o +--*--+ information +===o=| | Tracker o +-----+ /Super-peer ALTO Discovery o o Service o o +------+ o o | |oooooooooo +------+ Legend: === ALTO query protocol ooo ALTO service discovery protocol *** Application protocol (out of scope) ... Provisioning or initialization (out of scope) Figure 1: ALTO Discovery Overview Hereby the ALTO service discovery scenarios are classified into two types: one is the ALTO server discovery by the resource consumer, and the other is the ALTO server discovery by a third party, such as application trackers. Before the specification of the discovery mechanism the following section illustrates and discusses both scenarios. 1.1.1. ALTO Server Discovery by Resource Consumers The ALTO service discovery in some scenarios needs to be performed by the resource consumer itself. In particular in P2P applications without a tracker like DHTs and other conventional client/server applications. Also P2P application which are tracker based may embed the ALTO client into the resource consumer to allow peers a selection of peers after retrieving the peer list from the application tracker. The following figure illustrates this scenario, showing the relationship between the different entities as discussed before. Kiesel, et al. Expires January 17, 2013 [Page 6] Internet-Draft ALTO Server Discovery July 2012 +---------------+ | ALTO Server | +---------------+ ^ ^ +-----------+ | | | ALTO | | +---+---+ | Service | | |tracker| | Discovery | | +-------+ +---------+-+ | | o o +------------+--+ | o o |P2P Application|ooooo|oooooooooooo o | Client A | | o +---------------+ | o | o +--+-------------+ o | P2P Application|oooooo | Client B | +----------------+ Figure 2: Resource Consumer ALTO Server Discovery (Example) 1.1.2. ALTO Server Discovery by a Third Party Some P2P applications have trackers, and these applications might not need to have their clients looking for the ALTO server guidance. In these scenarios trackers query the ALTO servers for guidance themselves, and then return the final ranked result to the application clients. However, application clients are distributed among different network operators and autonomous systems. Trackers thus need to find different ALTO servers for the clients located in different operator networks or autonomous systems. In such scenarios the discovery is thus not performed by the resource consumer, but a third party entity on behalf of the resource consumer. This third party ALTO server discovery raises several issues. Typically only the IP address of the ALTO client is known that needs to suffice for finding the corresponding ALTO server. This requires detemining a FQDN as input for the U-NAPTR process. One option to do this is to use a reverse DNS lookup. However, reverse DNS has several limitations: First, there is no established unique way of maintaining the DNS tree, and there are different practices in different networks. Furthermore, it is possible that a lookup fails or that the returned value is not valid. For instance, it can point to a different domain. As a result, any potential use of reverse DNS lookup for service discovery must be able to deal with failures of lookup and react accordingly. Kiesel, et al. Expires January 17, 2013 [Page 7] Internet-Draft ALTO Server Discovery July 2012 Second, determining a domain name from IP addresses by tree climbing is problematic, in particular for IPv6. [RFC4472] discusses the issues for IPv6. Third, populating a DNS name space what looks like a reverse tree is a significant administrative DNS overhead. Finally it must be emphasized that any tree walking procedure raises several issues. There is no single best way tree, and heuristics are needed. Given these problems, this memo does not specify a reverve DNS mechanism to determine a FQDN. Thus, this draft is limited to scenarios where the discovery procedure is done by the resource consumer. In case a third party needs to know the contact information of the ALTO server it is RECOMMENDED that the resource consumer discovers the ALTO server and then sends the contact information directly to the third party. However, this requires a modification of the protocol used between the resource consumer and the third party, e.g., the tracker protocol in the peer-to-peer case. As a potential alternative, the client could provide a valid FQDN, so that the third party can use this as input for the U-NAPTR process, but this variant has no significant advantages. The options on how to transmit the contact information from the resource consumer to the third party are out of scope of this document. 1.2. Pre-Conditions In general, the ALTO server discovery SHOULD be based on the IP address that is used to communicate with other peers, i. e., it should return a server that can provide guidance for that address. In order to achieve this, the whole document assumes certain pre- conditions, in particular: o The ALTO server discovery procedure is executed on a per IP family base, i.e., seperate for IPv4 and IPv6. It is up to the ALTO client to decide which of the possible multiple results of different IP address families to use. The choice of whether to use IPv4 or IPv6 is out of scope of this document. o A change of the IP address at an interface invalidates the result of the ALTO server discovery procedure. For instance, if the IP address assigned to a mobile host changes due to host mobility, it is required to run the ALTO server discovery procedure for the new IP address without relying on earlier gained information. Kiesel, et al. Expires January 17, 2013 [Page 8] Internet-Draft ALTO Server Discovery July 2012 o The ALTO server discovery procedure is executed on a per IP address base. Multiple IP addresses per interface or multiple IP addresses assigned to different IP interfaces require to repeat the procedure for every IP address. It may be fine to group IP addresses according their domain suffixes and to perfom the procedure for such a group. However, this is out of scope of this document. o There are several challenges with DNS on hosts with multiple interfaces [RFC6418], which can affect the ALTO server discovery. If the DNS resolution is performed on the wrong interface, it can return an ALTO server that could provide sub-optimal or wrong guidance. Finding the best ALTO server for multi-interfaced hosts is outside the scope of this document. o The discovery procedure may need information about the public IP address and thus have to discover NATs. Details of NAT discovery are not discussed in this memo. o Similarly the discovery procedure may need information about the IP address of a proxy or agent if this address is used for communication. Examples include Virtual Private Network (VPN) or mobile IP triangular routing. Details on how to obtain the IP address that is used for communication in such setups are not discussed in this memo, since they can be specific to the network in use. Kiesel, et al. Expires January 17, 2013 [Page 9] Internet-Draft ALTO Server Discovery July 2012 2. Protocol Overview We define multiple alternatives to discover the IP address of the ALTO server, as there are a number of ways possible how such information can be provided to the ALTO client. The choice of method is up to the local network deployment. For instance, there can be deployments where the ALTO server in charge for ALTO client is provisioned by the network operator and communicated to the ALTO client's host via a DHCP option, while in other deployments no such means may exist. It should be noted that there is no silver bullet solution to the ALTO server discovery, as there too many deployment scenarios in the server discovery space. The following figure illustrates the different protocols that SHOULD be used to find the URI of a suitable ALTO server. Descending Order of Preference ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~> -------------- -------------- -------------- | User Input | | DHCP query | | PPP query | | in res.c. | | by res.c. | | by res.c. | -------------- -------------- -------------- | | | \ | / \--------+----------/ | V DNS suffix | V ------------------ | U-NAPTR lookup | ------------------ | V ALTO Server's Information Resource Directory URI Legend: res.c. : resource consumer Figure 3: Protocol Overview Figure 3 illustrates the U-NAPTR based resolution process to retrieve the ALTO Server URI As a precondition for resolution the U-NAPTR Kiesel, et al. Expires January 17, 2013 [Page 10] Internet-Draft ALTO Server Discovery July 2012 process needs the right DNS suffix as input. This domain name is typically the domain name of the client's IP address. In order to retrieve the DNS suffix we specify three options, which are listed in descending order of preference. A client SHOULD use the first method that determines a DNS suffix. It MAY try the other methods in case the U-NAPTR lookup failed. User input: A user MAY manually specify the DNS suffix on its own, either to access a 3rd party ALTO service provider or as it does know such information. This input MAY also origin from a web page where the user downloads the configuration, which is loaded as user input, or obtained by other discovery methods. DHCP: A network provider MAY provide the DNS suffix through a DHCP option. PPP: A network provider MAY provide the DNS suffix through a Point- to-Point (PPP) Internet Protocol Control Protocol (IPCP) extension. This discovery method SHOULD be repeated if the resource consumer moves, i. e., if its IP address changes. Given that DHCP and PPP are widely used for IP address configuration, this discovery method is applicable to many networks. The mechanism described in this document can also be applied to further networks if they provide an equivalent mechanism to retrieve the DNS suffix. Such straightforward extensions to other technologies are not specified in this memo. Instead of using the standard ALTO server discovery method, applications MAY also use own methods to discover an ALTO server. This variant is outside the scope of this document. Kiesel, et al. Expires January 17, 2013 [Page 11] Internet-Draft ALTO Server Discovery July 2012 3. Retrieving the URI by U-NAPTR This section specifies the U-NAPTR based resolution process. To start the U-NAPTR resolution process a domain name is required as input. Thus the section is devided into two parts: Section 3.2 describes the U-NAPTR resolution process itself. How the client identifies this DNS suffix of the access network where the resource consumer is registered in is described in Section 3.1. 3.1. Retrieving the Domain Name The U-NAPTR resolution process requires a domain name as input. The algorithm that SHOULD be applied to determine this domain name is described in this section. We specify three different options. In option 1 the user manually configures a specific ALTO service instance that he wants to use. Option 2 defines a DHCP and option 3 defines a PPP IPCP option to allow the network service provider a remote configuration of the client. In general, the ALTO server discovery SHOULD be based on the IP address that is used to communicate with other peers. The resource consumer may have private IP addresses and public IP addresses and depending on the deployment it might be necessary to determine for all IP addresses the ALTO server in charge of. To determine its public IP address the resource consumer may need to use STUN[RFC5389] or BEP24[bep24]. Determining the correct IP address out of multiple options strongly depends on the deployment scenario but is out of scope for this document, although we discuss it to some extend in Section 4. For the following examples we assume that the IP address of the resource consumer is a.b.c.d. 3.1.1. Option 1: User input A user may want to use a third party ALTO service instance. Therefore we allow the user to specify a DNS suffix on its own, for example in a config file option. The DNS suffix given by the user is combined with the IP address of the resource consumer to allow the third party ALTO service to direct the client to a suitable ALTO server based on the location of the client. A possible DNS suffix entered by the user may be: myaltoprovider.org In case not ALTO NAPTR records are found, we consider the discovery process based on user input as failed. A client MAY try one of the other options. Kiesel, et al. Expires January 17, 2013 [Page 12] Internet-Draft ALTO Server Discovery July 2012 3.1.2. Option 2: DHCP As a second option network operators MAY configure the domain name to be used for service discovery within an access network using DHCP. RFC 5986[RFC5986] defines DHCP IPv4 and IPv6 access network domain name options that identify a domain name that is suitable for service discovery within the access network. The ALTO server discovery procedure uses these DHCP options to retrieve the domain name as an input for the U-NAPTR resolution. One example could be: example.com 3.1.3. Option 3: PPP In some network deployments the PPP [RFC1661] IPCP [RFC1332] is used to do network configuration of residential user equipment, such as assigning an IP address or the name of a DNS server[RFC1877]. Thus as a third option a network operator MAY configure the domain name to be used for the server discovery using a PPP IPCP extension. The next section specifies the extension for configuration of the access network domain name, which can be used as input for the U-NAPTR process. One possible example yielded by this extension could be: example.com 3.1.3.1. Access Network Name Encoding This section describes the encoding of the domain name used in PPP IPCP extension Section 3.1.3.2. The domain name is encoded according to Section 3.1 of [RFC1035] whereby each label is represented as a one-octet length field followed by that number of octets. This is the same encoding as in DHCP according to RFC 5986[RFC5986]. Since every domain name ends with the null label of the root, a domain name is terminated by a length byte of zero. The high-order two bits of every length octet MUST be zero, and the remaining six bits of the length field limit the label to 63 octets or less. To simplify implementations, the total length of a domain name (i.e., label octets and label length octets) is restricted to 255 octets or less. For example, the domain "example.com." is encoded in 13 octets as: +---+---+---+---+---+---+---+---+---+---+---+---+---+ | 7 | e | x | a | m | p | l | e | 3 | c | o | m | 0 | +---+---+---+---+---+---+---+---+---+---+---+---+---+ Kiesel, et al. Expires January 17, 2013 [Page 13] Internet-Draft ALTO Server Discovery July 2012 3.1.3.2. Access Network Domain Name IPCP Configuration Option This Configuration Option defines a method for negotiating with the remote peer the name of the Access Network Domain Name to be used on the local end of the link. A summary of the Access Network Domain Name Configuration Option format is shown below. The fields are transmitted from left to right. Type Len Access Network Domain Name +-----+-----+-----+-----+-----+-----+-----+---- | TBD | n | s1 | s2 | s3 | s4 | s5 | ... +-----+-----+-----+-----+-----+-----+-----+---- The values s1, s2, s3, etc. represent the domain name labels in the domain name encoding. Note that the length field in the IPCP option represents the length of the entire domain name encoding, whereas the length fields in the domain name encoding (see Section 3.1.3.1) is the length of a single domain name label. Type: to be assigned by IANA Len: Length of the 'Access Network Domain Name' field in octets. Access Network Domain Name: The domain name of the Access Network for the client to use. 3.2. U-NAPTR Resolution The ALTO protocol specification [I-D.ietf-alto-protocol] expects that the ALTO discovery procedure yields the HTTP(S) URI of the ALTO server's Information Resource Directory, which gives further information about the capabilities and services provided by that ALTO server. The first step of the ALTO server discovery procedure (see Section 3.1) yielded an U-NAPTR/DDDS (URI-Enabled NAPTR/Dynamic Delegation Discovery Service) [RFC4848] application unique strings, in the form of a DNS name. An example is "example.com". In the second step, the ALTO Server discovery procedure needs to use the U-NAPTR [RFC4848] specification described below to obtain a URI (indicating host and protocol) for the ALTO server's Information Resource Directory. In this document, only the HTTP and HTTPS URL schemes are defined, as the ALTO protocol specification defines the access over both protocols, but no other [I-D.ietf-alto-protocol]. Note that the HTTP URL can be any valid HTTP(s) URL, including those Kiesel, et al. Expires January 17, 2013 [Page 14] Internet-Draft ALTO Server Discovery July 2012 containing path elements. The following two DNS entries show the U-NAPTR resolution for "example.com" to the HTTPS URL https://altoserver.example.com/secure/directory or the HTTP URL http://altoserver.example.com/directory, with the former being preferred. example.com. IN NAPTR 100 10 "u" "ALTO:https" "!.*!https://altoserver.example.com/secure/directory!" "" IN NAPTR 200 10 "u" "ALTO:http" "!.*!http://altoserver.example.com/directory!" "" There is a potential that retrieveing the domain name or the U-NAPTR lookup itself does not yield to a result, i.e. no ALTO NAPTR record is found. In this case the discovery procedure failed for this IP address. It is RECOMMENDED that clients give up the discovery process and wait a period of time before repeating the procedure. Clients MAY repeat the discovery procedure for a different IP address instantaneously. Kiesel, et al. Expires January 17, 2013 [Page 15] Internet-Draft ALTO Server Discovery July 2012 4. Applicability This section discusses the applicability of the proposed solution with respect to the resource consumer server discovery and the third party deployment scenarios. Each section discusses the proposed steps that are needed to determine the ALTO Server URI. 4.1. Applicability for Resource Consumer Server Discovery In this scenario the ALTO server discovery procedure is performed by the resource consumer, for example a peer in a P2P system. After the discovery the peer does the ALTO query on its own, or it might share the ALTO server contact information with a third party, for example a tracker, which then executes the ALTO query on behalf of the peer. To complete the ALTO server discovery process the resource consumer first SHOULD check whether the user has provider the domain name through manual configuration. If this is not the case the next step SHOULD be to check for the access network domain name DHCP option (Section 3.1.2). Finally the client SHOULD try to retrieve the domain name by the PPP option Section 3.1.3. A client can have several candidate IP addresses that it may use for the discovery process. For example if it is located behind a NAT, a private and a public IP address may be used for the discovery process. It depends on the deployment scenario which of the IP addresses is the correct one. Thus it is out-of-scope of this document to specify how exactly the client finds the right IP address. However in the following we list methods that may be used in order to determine these candidate IP addresses. Generally in P2P environments peers already have implemented mechanisms for NAT- traversal. This includes proprietary solutions to determine a peer's public IP address, for example by asking a neighbour peer about its record of the own IP address. Non-proprietary solutions that are favorable include the Session Traversal Utilities for NAT (STUN) [RFC5986] protocol to determine the public address. If the client is behind a residential gateway another option may be to use Universal Plug and Play (UPnP) [UPnP-IGD-WANIPConnection1] or the NAT Port Mapping Protocol (NAT-PMP) [I-D.cheshire-nat-pmp]. In case the ALTO discovery client has determined the domain name through one of the described options it proceedes with the U-NAPTR lookup as described in Section 3.2. 4.2. Applicability for Third Party Server Discovery In case of the third party server discovery deployment scenario the entity performing the ALTO server discovery process is different from Kiesel, et al. Expires January 17, 2013 [Page 16] Internet-Draft ALTO Server Discovery July 2012 the resource consumer. Typically the resource consumer is a peer whereas the ALTO client is a resource directory which seeks for ALTO guidance on behalf of the peer. Another use case for the third party discovery is an application that looks for ALTO guidance transparently for the resource consumer, for example a CDN. Here the ALTO server discovery process can also retrieve guidance through the PPP/DHCP options or manual user configuration, but only if the provided discovery information is forwarded by the resource consumer to the third party entity. In this case, additional mechanisms for the forwarding of this discovery information need to be specified. However these mechanisms are out of scope of this doument. In case the resource consumer needs guidance for a different IP address, for example one from a private network, we recommend that the resource consumer discovers the server itself and forwards the ALTO server contact information directly to the third party entity, which in turn can then do the third party ALTO query. Again, forwarding the contact information from the resource consumer to the third party entity is out of scope of this document. Kiesel, et al. Expires January 17, 2013 [Page 17] Internet-Draft ALTO Server Discovery July 2012 5. Deployment Considerations The mechanism specified in this document needs some configuration effort in order to work properly. 5.1. DHCP option for DNS Suffix Section 3.1.2 describes the usage of a DHCP option. It enables the network operator of the network, in which the ALTO client is located, to provide a DNS suffix. However, this assumes that this particular DHCP option is correctly passed from the DHCP server to the actual host with the ALTO client, and that the particular host understands this DHCP option. This memo assumes the client to be able to understand the proposed DHCP option, otherwise there is no further use of the DHCP option, but the client has to use the other proposed mechanisms. There are well-known issues with the handling of DHCP options in home gateways. One issue is that unkown DHCP options are not passed through some home gateways, effectively eliminating the DHCP option. Another well-known issues is the usage of home gateway specific DNS suffixes which "override" the DNS suffix provided by the network operator. For instance, a host behind a home gateway may receive a DNS suffix ".local" instead of "example.com". This suffix is not usuable for the server discovery procedure. 5.2. PPP option for DNS Suffix Section 3.1.3 describes the usage of a PPP option. It enables the network operator of the network, in which the ALTO client is located, to provide a DNS suffix. In residental networks, PPP is often terminated in the residential gateway. The ALTO client may run on hosts behind that gateway. As a result, the information may have to be passed to the client. The residential gateway could for instance use the DHCP option for that. Kiesel, et al. Expires January 17, 2013 [Page 18] Internet-Draft ALTO Server Discovery July 2012 6. IANA Considerations 6.1. Registration of PPP IPCP Configuration Option Type The IANA is requested to assign an Type code for the PPP IPCP Configuration Option Types for an Access Network Domain Name, as described in Section 3.1.3.2 of this document. [TO BE REMOVED: This registration should take place at the following location: http://www.iana.org/assignments/ppp-numbers] 6.2. Registration of U-NAPTR application service tag The IANA is requested to register the following U-NAPTR application service tag: Application Service Tag: ALTO Intended usage: Identifies a service that provides a Device with its location information. Defining Publication: The specification contained within this document. Contact information: The authors of this document Author/Change controller: The IESG Kiesel, et al. Expires January 17, 2013 [Page 19] Internet-Draft ALTO Server Discovery July 2012 7. Security Considerations 7.1. General There are two different failures for the ALTO server discovery, which can both be caused by malicious attacks or by configuration problems, e. g., in case of DNS configuration errors or multi-homed hosts. First, the discovery might not be able to discover an ALTO server, even if a suitable ALTO server exists. In that case, ALTO guidance will not be used. The resulting application performance and traffic distribution will correspond to a deployment scenario without ALTO guidance. But given that users cannot rely on the availability of an ALTO server, this results in no significant additional security risk. Second, the discovery procedure may discover a sub-optimal or wrong ALTO server. Such an ALTO server may either not be able to provide information for a given resource consumer (e. g., behind a NAT), thus rendering the ALTO service useless. Alternatively, it may provide sub-optimal or forged information. In the latter case, attackers could try to use ALTO to affect the traffic distribution or the performance of applications. Users may then observe performance problems, and network operators could detect traffic anormalities. A potential counter-measure is to disable the use of the ALTO service. Security issues of ALTO in general and potential solutions are also discussed in [I-D.ietf-alto-protocol]. 7.2. For U-NAPTR The address of an ALTO server is usually well-known within an access network; therefore, interception of messages does not introduce any specific concerns. The primary attack against the methods described in this document is one that would lead to impersonation of an ALTO server since a device does not necessarily have a prior relationship with an ALTO server. An attacker could attempt to compromise ALTO discovery at any of three stages: 1. providing a falsified domain name to be used as input to U-NAPTR 2. altering the DNS records used in U-NAPTR resolution 3. impersonation of the ALTO server This document focuses on the U-NAPTR resolution process and hence Kiesel, et al. Expires January 17, 2013 [Page 20] Internet-Draft ALTO Server Discovery July 2012 this section discusses the security considerations related to the DNS handling. The security aspects of obtaining the domain name that is used for input to the U-NAPTR process is described in respective documents, such as [RFC5986]. The domain name that is used to authenticated the ALTO server is the domain name in the URI that is the result of the U-NAPTR resolution. Therefore, if an attacker was able to modify or spoof any of the DNS records used in the DDDS resolution, this URI could be replaced by an invalid URI. The application of DNS security (DNSSEC) [RFC4033] provides a means to limit attacks that rely on modification of the DNS records used in U-NAPTR resolution. Security considerations specific to U-NAPTR are described in more detail in [RFC4848]. An "https:" URI is authenticated using the method described in Section 3.1 of [RFC2818]. The domain name used for this authentication is the domain name in the URI resulting from U-NAPTR resolution, not the input domain name as in [RFC3958]. Using the domain name in the URI is more compatible with existing HTTP client software, which authenticate servers based on the domain name in the URI. An ALTO server that is identified by an "http:" URI cannot be authenticated. If an "http:" URI is the product of the ALTO discovery, this leaves devices vulnerable to several attacks. Lower layer protections, such as layer 2 traffic separation might be used to provide some guarantees. Kiesel, et al. Expires January 17, 2013 [Page 21] Internet-Draft ALTO Server Discovery July 2012 8. Conclusion This document describes a general ALTO server discovery process and discusses how the process can be applied in different deployment scenarios. The discovery process uses U-NAPTR resolution based on input information obtained either manually, by DHCP, or by PPP. Kiesel, et al. Expires January 17, 2013 [Page 22] Internet-Draft ALTO Server Discovery July 2012 9. References 9.1. Normative References [RFC1035] Mockapetris, P., "Domain names - implementation and specification", STD 13, RFC 1035, November 1987. [RFC1332] McGregor, G., "The PPP Internet Protocol Control Protocol (IPCP)", RFC 1332, May 1992. [RFC1661] Simpson, W., "The Point-to-Point Protocol (PPP)", STD 51, RFC 1661, July 1994. [RFC1877] Cobb, S. and F. Baker, "PPP Internet Protocol Control Protocol Extensions for Name Server Addresses", RFC 1877, December 1995. [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818, May 2000. [RFC3958] Daigle, L. and A. Newton, "Domain-Based Application Service Location Using SRV RRs and the Dynamic Delegation Discovery Service (DDDS)", RFC 3958, January 2005. [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose, "DNS Security Introduction and Requirements", RFC 4033, March 2005. [RFC5389] Rosenberg, J., Mahy, R., Matthews, P., and D. Wing, "Session Traversal Utilities for NAT (STUN)", RFC 5389, October 2008. [RFC6418] Blanchet, M. and P. Seite, "Multiple Interfaces and Provisioning Domains Problem Statement", RFC 6418, November 2011. 9.2. Informative References [I-D.cheshire-nat-pmp] Cheshire, S., "NAT Port Mapping Protocol (NAT-PMP)", draft-cheshire-nat-pmp-03 (work in progress), April 2008. [I-D.ietf-alto-deployments] Stiemerling, M. and S. Kiesel, "ALTO Deployment Considerations", draft-ietf-alto-deployments-03 (work in progress), November 2011. Kiesel, et al. Expires January 17, 2013 [Page 23] Internet-Draft ALTO Server Discovery July 2012 [I-D.ietf-alto-protocol] Penno, R., Alimi, R., and Y. Yang, "ALTO Protocol", draft-ietf-alto-protocol-10 (work in progress), October 2011. [I-D.ietf-alto-reqs] Kiesel, S., Previdi, S., Stiemerling, M., Woundy, R., and Y. Yang, "Application-Layer Traffic Optimization (ALTO) Requirements", draft-ietf-alto-reqs-11 (work in progress), July 2011. [I-D.kist-alto-3pdisc] Kiesel, S. and M. Stiemerling, "3rd Party ALTO Server Discovery (3pdisc)", draft-kist-alto-3pdisc-00 (work in progress), July 2012. [RFC4472] Durand, A., Ihren, J., and P. Savola, "Operational Considerations and Issues with IPv6 DNS", RFC 4472, April 2006. [RFC4848] Daigle, L., "Domain-Based Application Service Location Using URIs and the Dynamic Delegation Discovery Service (DDDS)", RFC 4848, April 2007. [RFC5693] Seedorf, J. and E. Burger, "Application-Layer Traffic Optimization (ALTO) Problem Statement", RFC 5693, October 2009. [RFC5986] Thomson, M. and J. Winterbottom, "Discovering the Local Location Information Server (LIS)", RFC 5986, September 2010. [UPnP-IGD-WANIPConnection1] UPnP Forum, "Internet Gateway Device (IGD) Standardized Device Control Protocol V 1.0: WANIPConnection:1 Service Template Version 1.01 For UPnP Version 1.0", DCP 05-001, November 2001. [bep24] Harrison, D., "Tracker Returns External IP", BEP http://bittorrent.org/beps/bep_0024.html. Kiesel, et al. Expires January 17, 2013 [Page 24] Internet-Draft ALTO Server Discovery July 2012 Appendix A. Contributors List and Acknowledgments The initial version of this document was co-authored by Marco Tomsu . Hannes Tschofenig provided the initial input to the U-NAPTR solution part. Hannes and Martin Thomson provided excellent feedback and input to the server discovery. Olafur Gudmundsson provided an excellent DNS expert review on an earlier version of this document. The authors would also like to thank the following persons for their contribution to this document or its predecessors: Richard Alimi, David Bryan, Roni Even, Gustavo Garcia, Jay Gu, Xingfeng Jiang, Enrico Marocco, Victor Pascual, Y. Richard Yang, Yu-Shun Wang, Yunfei Zhang, Ning Zong. Nico Schwan is partially supported by the ENVISION project (http://www.envision-project.org), a research project supported by the European Commission under its 7th Framework Program (contract no. 248565). The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the ENVISION project or the European Commission. Michael Scharf is supported by the German-Lab project (http://www.german-lab.de) funded by the German Federal Ministry of Education and Research (BMBF). Martin Stiemerling is partially supported by the COAST project (COntent Aware Searching, retrieval and sTreaming, http://www.coast-fp7.eu), a research project supported by the European Commission under its 7th Framework Program (contract no. 248036). The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the COAST project or the European Commission. Kiesel, et al. Expires January 17, 2013 [Page 25] Internet-Draft ALTO Server Discovery July 2012 Authors' Addresses Sebastian Kiesel University of Stuttgart Computing Center Allmandring 30 Stuttgart 70550 Germany Email: ietf-alto@skiesel.de URI: http://www.rus.uni-stuttgart.de/nks/ Martin Stiemerling NEC Laboratories Europe Kurfuerstenanlage 36 Heidelberg 69115 Germany Phone: +49 6221 4342 113 Email: martin.stiemerling@neclab.eu URI: http://ietf.stiemerling.org Nico Schwan Alcatel-Lucent Bell Labs Lorenzstrasse 10 Stuttgart 70435 Germany Email: nico.schwan@alcatel-lucent.com URI: www.alcatel-lucent.com/bell-labs Michael Scharf Alcatel-Lucent Bell Labs Lorenzstrasse 10 Stuttgart 70435 Germany Email: michael.scharf@alcatel-lucent.com URI: www.alcatel-lucent.com/bell-labs Haibin Song Huawei Email: melodysong@huawei.com Kiesel, et al. Expires January 17, 2013 [Page 26]