idnits 2.17.1 draft-ietf-mif-current-practices-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (April 24, 2011) is 4745 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-15) exists of draft-ietf-mif-problem-statement-13 -- Obsolete informational reference (is this intentional?): RFC 3484 (Obsoleted by RFC 6724) -- Duplicate reference: RFC3484, mentioned in 'RFC3484', was also mentioned in 'ANDROID-RFC3484'. -- Obsolete informational reference (is this intentional?): RFC 3484 (Obsoleted by RFC 6724) -- Duplicate reference: RFC3484, mentioned in 'WNDS-RFC3484', was also mentioned in 'RFC3484'. -- Obsolete informational reference (is this intentional?): RFC 3484 (Obsoleted by RFC 6724) Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Engineering Task Force M. Wasserman 3 Internet-Draft Painless Security, LLC 4 Intended status: Informational P. Seite 5 Expires: October 26, 2011 France Telecom - Orange 6 April 24, 2011 8 Current Practices for Multiple Interface Hosts 9 draft-ietf-mif-current-practices-10 11 Abstract 13 An increasing number of hosts are operating in multiple-interface 14 environments. This document summarizes current practices in this 15 area, and describes in detail how some common operating systems cope 16 with challenges ensue from this context. 18 Status of this Memo 20 This Internet-Draft is submitted in full conformance with the 21 provisions of BCP 78 and BCP 79. 23 Internet-Drafts are working documents of the Internet Engineering 24 Task Force (IETF). Note that other groups may also distribute 25 working documents as Internet-Drafts. The list of current Internet- 26 Drafts is at http://datatracker.ietf.org/drafts/current/. 28 Internet-Drafts are draft documents valid for a maximum of six months 29 and may be updated, replaced, or obsoleted by other documents at any 30 time. It is inappropriate to use Internet-Drafts as reference 31 material or to cite them other than as "work in progress." 33 This Internet-Draft will expire on October 26, 2011. 35 Copyright Notice 37 Copyright (c) 2011 IETF Trust and the persons identified as the 38 document authors. All rights reserved. 40 This document is subject to BCP 78 and the IETF Trust's Legal 41 Provisions Relating to IETF Documents 42 (http://trustee.ietf.org/license-info) in effect on the date of 43 publication of this document. Please review these documents 44 carefully, as they describe your rights and restrictions with respect 45 to this document. Code Components extracted from this document must 46 include Simplified BSD License text as described in Section 4.e of 47 the Trust Legal Provisions and are provided without warranty as 48 described in the Simplified BSD License. 50 Table of Contents 52 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 53 2. Summary of Current Approaches . . . . . . . . . . . . . . . . 3 54 2.1. Centralized Connection Management . . . . . . . . . . . . 3 55 2.2. Per Application Connection Settings . . . . . . . . . . . 4 56 2.3. Stack-Level Solutions to Specific Problems . . . . . . . . 4 57 2.3.1. DNS Resolution Issues . . . . . . . . . . . . . . . . 5 58 2.3.2. First hop selection . . . . . . . . . . . . . . . . . 5 59 2.3.3. Address Selection Policy . . . . . . . . . . . . . . . 5 60 3. Current Practices in Some Operating Systems . . . . . . . . . 6 61 3.1. Mobile Handset Operating Systems . . . . . . . . . . . . . 6 62 3.1.1. Nokia S60 3rd Edition, Feature Pack 2 . . . . . . . . 7 63 3.1.2. Microsoft Windows Mobile and Windows Phone 7 . . . . . 9 64 3.1.3. RIM BlackBerry . . . . . . . . . . . . . . . . . . . . 10 65 3.1.4. Google Android . . . . . . . . . . . . . . . . . . . . 11 66 3.1.5. Qualcomm Brew . . . . . . . . . . . . . . . . . . . . 12 67 3.1.6. Leadcore Tech. Arena . . . . . . . . . . . . . . . . . 13 68 3.2. Desktop Operating Systems . . . . . . . . . . . . . . . . 13 69 3.2.1. Microsoft Windows . . . . . . . . . . . . . . . . . . 14 70 3.2.1.1. First hop selection . . . . . . . . . . . . . . . 14 71 3.2.1.2. Outbound and Inbound Addresses . . . . . . . . . . 14 72 3.2.1.3. DNS Configuration . . . . . . . . . . . . . . . . 14 73 3.2.2. Linux and BSD-based Operating Systems . . . . . . . . 15 74 3.2.2.1. First hop selection . . . . . . . . . . . . . . . 16 75 3.2.2.2. Outbound and Inbound Addresses . . . . . . . . . . 16 76 3.2.2.3. DNS Configuration . . . . . . . . . . . . . . . . 17 77 3.3. Focus on access network selection . . . . . . . . . . . . 18 78 3.3.1. Android/HTC magic . . . . . . . . . . . . . . . . . . 18 79 3.3.2. RIM BlackBerry . . . . . . . . . . . . . . . . . . . . 19 80 4. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 20 81 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 20 82 6. Security Considerations . . . . . . . . . . . . . . . . . . . 20 83 7. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 21 84 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 21 85 8.1. Normative References . . . . . . . . . . . . . . . . . . . 21 86 8.2. Informative References . . . . . . . . . . . . . . . . . . 22 87 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 23 89 1. Introduction 91 Multiple-interface hosts face several challenges not faced by single- 92 interface hosts, some of which are described in the MIF problem 93 statement, [I-D.ietf-mif-problem-statement]. This document 94 summarizes how current implementations deal with the problems 95 identified in the MIF problem statement. 97 Publicly-available information about the multiple-interface solutions 98 implemented in some widely used operating systems, including both 99 mobile handset and desktop operating systems, is collected in this 100 document, including: Nokia S60 [S60], Microsoft Windows Mobile 101 [WINDOWSMOBILE], Blackberry [BLACKBERRY], Google Android [ANDROID], 102 Microsoft Windows, Apple Mac OS X, Linux and BSD-based operating 103 systems. 105 2. Summary of Current Approaches 107 This section summarizes current approaches that are used to resolve 108 the multi-interface issues described in the Multiple Interface 109 Problem Statement [I-D.ietf-mif-problem-statement]. These approaches 110 can be broken down into three major categories: 112 o Centralized connection management 114 o Per-application connection settings 116 o Stack-level solutions to specific problems 118 2.1. Centralized Connection Management 120 It is a common practice for mobile handset operating systems to use a 121 centralized connection manager that performs network interface 122 selection based on application or user input. The information used 123 by the connection manager may be programmed into an application or 124 provisioned on a handset-wide basis. When information is not 125 available to make an interface selection, the connection manager will 126 query the user to choose between available choices. 128 Routing tables are not typically used for network interface selection 129 when a connection manager is in use, as the criteria for network 130 selection is not strictly IP-based but is also dependent on other 131 properties of the interface (cost, type, etc.). Furthermore, 132 multiple overlapping private IPv4 address spaces are often exposed to 133 a multiple-interface host, making it difficult to make interface 134 selection decisions based on prefix matching. 136 2.2. Per Application Connection Settings 138 In mobile handsets, applications are often involved in choosing what 139 interface and related configuration information should be used. In 140 some cases, the application selects the interface directly, and in 141 other cases the application provides more abstract information to a 142 connection manager that makes the final interface choice. 144 2.3. Stack-Level Solutions to Specific Problems 146 In most desktop operating systems, multiple interface problems are 147 dealt with in the stack and related components, based on system- 148 level configuration information, without the benefit of input from 149 applications or users. These solutions tend to map well to the 150 problems listed in the problem statement: 152 o DNS resolution issues 154 o Routing 156 o Address selection policy 158 The configuration information for desktop systems comes from one of 159 three sources: DHCP, proprietary configuration systems or manual 160 configuration. While these systems universally accept IP address 161 assignment on a per-interface basis, they differ in what set of 162 information can be assigned on a per-interface basis and what can be 163 configured only on a per-system basis. 165 When choosing between multiple sets of information provided, these 166 systems will typically give preference to information received on the 167 "primary" interface. The mechanism for designating the "primary" 168 interface differs by system. 170 There is very little commonality in how desktop operating systems 171 handle multiple sets of configuration information, with notable 172 variations between different versions of the same operating system 173 and/or within different software packages built for the same 174 operating system. Although these systems differ widely, it is not 175 clear that any of them provide a completely satisfactory user 176 experience in multiple-interface environments. 178 The following sections discuss some of the solutions used in each of 179 the areas raised in the MIF problem statement. 181 2.3.1. DNS Resolution Issues 183 There is very little commonality in how desktop operating systems 184 handle the DNS server list. Some systems support per-interface DNS 185 server lists, while others only support a single system-wide list. 187 On hosts with per-interface DNS server lists, different mechanisms 188 are used to determine which DNS server is contacted for a given 189 query. In most cases, the first DNS server listed on the "primary" 190 interface is queried first, with back off to other servers if an 191 answer is not received. 193 Systems that support a single system-wide list differ in how they 194 select which DNS server to use in cases where they receive more than 195 one DNS server list to configure (e.g. from DHCP on multiple 196 interfaces). Some accept the information received on the "primary" 197 interface, while others use either the first or last set DNS server 198 list configured. 200 2.3.2. First hop selection 202 Routing information is also handled differently on different desktop 203 operating systems. While all systems maintain some sort of routing 204 cache, to handle redirects and/or statically configured routes, most 205 packets are routed based on configured default gateway information. 207 Some systems do allow the configuration of different default router 208 lists for different interfaces. These systems will always choose the 209 default gateway on the interface with the lowest routing metric, with 210 different behavior when two or more interfaces have the same routing 211 metric. 213 Most systems do not allow the configuration of more than one default 214 router list, choosing instead to use the first or last default router 215 list configured and/or the router list configured on the "primary" 216 interface. 218 2.3.3. Address Selection Policy 220 There is somewhat more commonality in how desktop hosts handle 221 address selection. Applications typically provide the destination 222 address for an outgoing packet, and the IP stack is responsible for 223 picking the source address. 225 IPv6 specifies a specific source address selection mechanism in 226 [RFC3484], and several systems implement this mechanism with similar 227 support for IPv4. However, many systems do not provide any mechanism 228 to update this default policy, and there is no standard way to do so. 230 In some cases, the routing decision (including which interface to 231 use) is made before source address selection is performed, and a 232 source address is chosen from the outbound interface. In other 233 cases, source address selection is performed before, or independently 234 from outbound interface selection. 236 3. Current Practices in Some Operating Systems 238 The following sections briefly describe the current multiple- 239 interface host implementations on some widely-used operating systems. 240 Please refer to the References section for pointers to original 241 documentation on most of these systems, including further details. 243 3.1. Mobile Handset Operating Systems 245 Cellular devices typically run a variety of applications in parallel, 246 each with different requirements for IP connectivity. A typical 247 scenario is shown in figure 1, where a cellular device is utilizing 248 WLAN access for web browsing and GPRS access for transferring 249 multimedia messages (MMS). Another typical scenario would be a real- 250 time VoIP session over one network interface in parallel with best 251 effort web browsing on another network interface. Yet another 252 typical scenario would be global Internet access through one network 253 interface and local (e.g. corporate VPN) network access through 254 another. 256 Web server MMS Gateway 257 | | 258 -+--Internet---- ----Operator network--+- 259 | | 260 +-------+ +-------+ 261 |WLAN AP| | GGSN | 262 +-------+ +-------+ 263 | +--------+ | 264 +--------|Cellular|--------+ 265 |device | 266 +--------+ 268 A cellular device with two network interfaces 270 Figure 1 272 Different network access technologies require different settings. 273 For example, WLAN requires Service Set Identifier (SSID) and the GPRS 274 network requires the Access Point Name (APN) of the Gateway GPRS 275 Support Node (GGSN), among other parameters. It is common that 276 different accesses lead to different destination networks (e.g. to 277 "Internet", "intranet", cellular network services, etc.). 279 3.1.1. Nokia S60 3rd Edition, Feature Pack 2 281 S60 is a software platform for mobile devices running on the Symbian 282 OS. S60 uses the concept of an Internet Access Point (IAP) [S60] 283 that contains all information required for opening a network 284 connection using a specific access technology. A device may have 285 several IAPs configured for different network technologies and 286 settings (multiple WLAN SSIDs, GPRS APNs, dial-up numbers, and so 287 forth). There may also be 'virtual' IAPs that define parameters 288 needed for tunnel establishment (e.g. for VPN). 290 For each application, a correct IAP needs to be selected at the point 291 when the application requires network connectivity. This is 292 essential, as the wrong IAP may not be able to support the 293 application or reach the desired destination. For example, MMS 294 application must use the correct IAP in order to reach the MMS 295 Gateway, which typically is not accessible from the public Internet. 296 As another example, an application might need to use the IAP 297 associated with its corporate VPN in order to reach internal 298 corporate servers. Binding applications to IAPs avoids several 299 problems, such as choosing the correct DNS server in the presence of 300 split DNS (as an application will use the DNS server list from its 301 bound IAP), and overlapping private IPv4 address spaces used for 302 different interfaces (as each application will use the default routes 303 from its bound IAP). 305 If multiple applications utilize the same IAP, the underlying network 306 connection can typically be shared. This is often the case when 307 multiple Internet-using applications are running in parallel. 309 The IAP for an application can be selected in multiple ways: 311 o Statically: e.g. from a configuration interface, via client 312 provisioning/device management system, or at build-time. 314 o Manually by the user: e.g. each time an application starts the 315 user may be asked to select the IAP to use. This may be needed, 316 for example, if a user sometimes wishes to access his corporate 317 intranet and other times would prefer to access the Internet 318 directly. 320 o Automatically by the system: after the destination network has 321 been selected statically or dynamically. 323 The static approach is fine for certain applications, like MMS, for 324 which configuration can be provisioned by the network operator and 325 does not change often. Manual selection works, but may be seen as 326 troublesome by the user. An automatic selection mechanism needs to 327 have some way of knowing which destination network the user, or an 328 application, is trying access. 330 S60 3rd Edition, Feature Pack 2, introduces a concept of Service 331 Network Access Points (SNAPs) that group together IAPs that lead to 332 the same destination. This enables static or manual selection of the 333 destination network for an application and leaves the problem of 334 selecting the best of the available IAPs within a SNAP to the 335 operating system. 337 When SNAPs are used, it is possibly for the operating system to 338 notify applications when a preferred IAP, leading to the same 339 destination, becomes available (for example, when a user comes within 340 range of his home WLAN access point), or when the currently used IAP 341 is no longer available and applications have to reconnect via another 342 IAP (for example, when a user goes out of range of his home WLAN and 343 must move to the cellular network). 345 In S60 3.2 does not support RFC 3484 for source address selection 346 mechanisms. Applications are tightly bound the network interface 347 selected for them or by them. E.g. an application may be connected 348 to IPv6 3G connection, IPv4 3G connection, WLAN connection, or VPN 349 connection. The application can change between the connections, but 350 uses only one at a time. If the interface happens to be dual-stack, 351 then IPv4 is preferred over IPv6. 353 DNS configuration is per-interface; an application bound to an 354 interface will always use the DNS settings for that interface. Hence 355 the device itself remembers these pieces of information for each 356 interface separately. 358 The S60 3.2 manages with totally overlapping addresses spaces. Each 359 interface can even have same IPv4 address configured on it without 360 issues. This is so because interfaces are kept totally separate from 361 each other. This also implies that the interface selection has to be 362 done at application layer, as from network layer point of view device 363 is not multihomed in the IP-sense. 365 Please see the source documentation for more details and screenshots: 366 [S60]. 368 3.1.2. Microsoft Windows Mobile and Windows Phone 7 370 Microsoft Windows Mobile leverages on a Connection Manager 371 [WINDOWSMOBILE] to handle multiple network connections. This 372 architecture centralizes and automates network connection 373 establishment and management, and makes it possible to automatically 374 select a connection, to dial-in automatically or by user initiation, 375 and to optimize connection and shared resource usage. Connection 376 Manager periodically re-evaluates the validity of the connection 377 selection. The Connection Manager uses various attributes such as 378 cost, security, bandwidth, error rate, and latency in its decision 379 making. 381 The Connection Manager selects the best possible connection for the 382 application based on the destination network the application wishes 383 to reach. The selection is made between available physical and 384 virtual connections (e.g. VPN, GPRS, WLAN, and wired Ethernet) that 385 are known to provide connectivity to the destination network, and the 386 selection is based on the costs associated with each connection. 387 Different applications are bundled to use the same network connection 388 when possible, but in conflict situations when a connection cannot be 389 shared, higher priority applications take precedence, and the lower 390 priority applications lose connectivity until the conflict situation 391 clears. 393 During operation, Connection Manager opens new connections as needed, 394 and also disconnects unused or idle connections. 396 To optimize resource use, such as battery power and bandwidth, 397 Connection Manager enables applications to synchronize network 398 connection usage by allowing applications to register their 399 requirements for periodic connectivity. An application is notified 400 when a suitable connection becomes available for its use. 402 In comparison to Windows Mobile connection management, Windows phone 403 7 updates the routing functionality in the case where the terminal 404 can be attached simultaneously to several interfaces. Windows Phone 405 7 selects the first hop corresponding to the interface which has a 406 lower metric. When there are multiple interfaces, the applications 407 system will, by default, choose from an ordered list of available 408 interfaces. The default connection policy will prefer wired over 409 wireless and WLAN over cellular. Hence, if an application wants to 410 use cellular 3G as the active interface when WLAN is available, the 411 application needs to override the default connection mapping policy. 412 An application specific mapping policy can be set via a microsoft API 413 or provisioned by the Mobile Operator. The application, in 414 compliance with the security model, can request connection type by 415 interface (WLAN, cellular), by minimum interface speed (x kbps, y 416 mbps), or by name (Access Point Name). 418 In dual-stack systems, Windows mobile and Windows phone 7 implement 419 adress selection rules as per [WNDS-RFC3484]. An administrator can 420 configure a policy table that can override the default behavior of 421 the selection algorithms. It is reminded that the policy table 422 specifies precedence values and preferred source prefixes for 423 destination prefixes (see [RFC3484], section 2.1, for details). If 424 the system has not been configured, then the default policy table 425 specified in [RFC3484] is used. 427 3.1.3. RIM BlackBerry 429 Depending on the network configuration, applications in reasearch In 430 Motion (RIM) BlackBerry devices [BLACKBERRY] can use can use direct 431 TCP/IP connectivity or different application proxys to establish 432 connections over the wireless network. For instance, some wireless 433 service providers provide an Internet gateway to offer direct TCP/IP 434 connectivity to the Internet while some others can provide a WAP 435 gateway that allows HTTP connections to occur over the WAP (Wireless 436 Application Protocol) protocol. It is also possible to use the 437 BlackBerry Enterprise Server [BLACKBERRY] as a network gateway, The 438 BlackBerry Enterprise Server provides an HTTP and TCP/IP proxy 439 service to allow the application to use it as a secure gateway for 440 managing HTTP and TCP/IP connections to the intranet or the Internet. 441 An application connecting to the Internet, can use either the 442 BlackBerry Internet Service or the Internet gateway of the wireless 443 server provider or direct Internet connectivity over WLAN to manage 444 connections. The problem of gateway selection is supposed to be 445 managed independently by each application. For instance, an 446 application can be designed to always use the default Internet 447 gateway, while another application can be designed to use a preferred 448 proxy when available. 450 A BlackBerry device [BLACKBERRY] can be attached to multiple networks 451 simultaneously (wireless/wired). In this case, Multiple network 452 interfaces can be associated to a single IP stack or multiple IP 453 stacks. The device, or the application, can select the network 454 interface to be used in various ways. For instance, the device can 455 always map the applications to the default network interface (or the 456 default access network). When muliple IP stacks are associated to 457 multiple interfaces, the application can select the source address 458 correponding to the preferred network interface. Per-interface IP 459 stacks also allow to manage overlapping addresses spaces. When 460 multiple network interfaces are aggregated into a single IP stack, 461 the device associates each application to the more appropriate 462 network interface. The selection can be based on cost, type-of- 463 service and/or user preference. 465 The BlackBerry uses per-interface DNS configuration; applications 466 bound to a specific interface will use the DNS settings for that 467 interface. 469 3.1.4. Google Android 471 Android is based on a Linux kernel and, in many situations, behaves 472 like a Linux device as described in Section 3.2.2. As per Linux, 473 Android can manage multiple routing tables and rely on policy based 474 routing associated with packet filtering capabilities (see 475 Section 3.2.2.1 for details). Such a framework can be used to solve 476 complex routing issue brought by multiple interfaces terminals, e.g. 477 address space overlapping. 479 For incoming packets, Android implements the weak host model 480 [RFC1122] on both IPv4 and IPv6. However, Android can also be 481 configured to support the strong host model. 483 Regarding DNS configuration, Android does not list the DNS servers in 484 the file /etc/resolv.conf, used by Linux. However, as per Linux, DNS 485 configuration is node-scoped, even if DNS configuration can rely on 486 the DHCP client. For instance, the udhcp client [UDHCP], which is 487 also available for Linux, can be used on Android. Each time new 488 configuration data is received by the host from a DHCP server, 489 regardless of which interface it is received on, the DHCP client 490 rewrites the global configuration data with the most recent 491 information received. 493 Actually, the main difference between Linux and Android is on the 494 address selection mechanism. Android version prior to 2.2 simply 495 prefers IPv6 connectivity over IPv4. However, it should be noted 496 that, at the time of writing, IPv6 is available only on WiFi and 497 virtual interfaces, but not on the cellular interface (without IPv6 498 in IPv4 encapsulation). Android 2.2 has been updated with 499 [ANDROID-RFC3484], which implements some of the address selection 500 rules defined in [RFC3484]. All RFC3484 rules are supported, except 501 rule 3 (avoid deprecated addresses), 4 (prefer home addresses) and 7 502 (prefer native transport). Also, rule 9 (use longest matching 503 prefix) has been modified so it does not sort IPv4 addresses. 505 The Android reference documentation describes the android.net package 506 [ANDROID] and the ConnectivityManager class that applications can use 507 to request the first hop to a specified destination address via a 508 specified network interface (3GPP or WLAN). Applications also ask 509 Connection Manager for permission to start using a network feature. 510 The Connectivity Manager monitors changes in network connectivity and 511 attempts to failover to another network if connectivity to an active 512 network is lost. When there are changes in network connectivity, 513 applications are notified. Applications are also able to ask for 514 information about all network interfaces, including their 515 availability, type and other information. 517 3.1.5. Qualcomm Brew 519 This section describes how multi-interface support is handled by 520 Advanced Mobile Station Software (AMSS) that comes with Brew OS for 521 all Qualcomm chipsets (e.g., MSM, Snapdragon etc). AMSS is a low 522 level connectivity platform, on top of which manufacturers can build 523 to provide the necessary connectivity to applications. The 524 interaction model between AMSS, the Operating System, and the 525 applications is not unique and depend on the design chosen by the 526 manufacturer. The Mobile OS can let an application invoke the AMSS 527 directly (via API), or provide its own connection manager that will 528 request connectivity to the AMSS based on applications needs. The 529 interaction between the OS connection manager and the applications is 530 OS dependent. 532 AMSS supports a concept of netpolicy which allows each application to 533 specify the type of network connectivity desired. The netpolicy 534 contains parameters such as access technology, IP version type and 535 network profile. Access technology could be a specific technology 536 type such as CDMA or WLAN or could be a group of technologies, such 537 as ANY_Cellular or ANY_Wireless. IP version could be one of IPv4, 538 IPv6 or Default. The network profile identifies a type of network 539 domain or service within a certain network technology, such as 3GPP 540 APN or Mobile IP Home Agent. It also specifies all the mandatory 541 parameters required to connect to the domain such authentication 542 credentials and other optional parameters such as QoS attributes. 543 Network Profile is technology specific and set of parameters 544 contained in the profile could vary for different technologies. 546 Two models of network usage are supported: 548 o Applications requiring network connectivity specify an appropriate 549 netpolicy in order to select the desired network. The netpolicy 550 may match one or more network interfaces. AMSS system selection 551 module selects the best interface out of the ones that match the 552 netpolicy based on various criteria such as cost, speed or other 553 provisioned rules. Application explicitly starts the selected 554 network interface and, as a result, the application also gets 555 bound to the corresponding network interface. All outbound 556 packets from this application are always routed over this bound 557 interface using the source address of the interface. 559 o Applications may rely on a separate connection manager to control 560 (e.g. start/stop) the network interface. In this model, 561 applications are not necessarily bound to any one interface. All 562 outbound packets from such applications are routed on one of the 563 interfaces that match its netpolicy. The routing decision is made 564 individually for each packet and selects the best interface based 565 on the criteria described above and the destination address. 566 Source address is always that assigned to the interface used to 567 transmit the packet. 569 All of the routing/interface selection decisions are based on the 570 netpolicy and not just on the destination address to avoid 571 overlapping private IPv4 address issue. This also allows multiple 572 interfaces to be configured with the same IP address, for example, to 573 handle certain tunnelling scenarios. Applications that do not 574 specify a netpolicy are routed by AMSS to the best possible interface 575 using the default netpolicy. Default netpolicy could be pre-defined 576 or provisioned by the administrator or operator. Hence default 577 interface could vary from device to device and also depends upon the 578 available networks at any given time. 580 AMSS allows each interface to be configured with its own set of DNS 581 configuration parameters (e.g. list of DNS servers, domain names 582 etc.). Interface selected to make a DNS resolution is the one to 583 which application making the DNS query is bound. Applications can 584 also specify a different netpolicy as part of DNS request to select 585 another interface for DNS resolution. Regardless, all the DNS 586 queries are sent only over this selected interface using the DNS 587 configuration from the interface. DNS resolution is first attempted 588 with the primary server configured in the interface. If a response 589 is not received, the queries are sent to all the other servers 590 configured in the interface in a sequential manner using a backoff 591 mechanism. 593 3.1.6. Leadcore Tech. Arena 595 Arena, a mobile OS based on Linux, provides a Connection Manager, 596 which is described in [I-D.zhang-mif-connection-manager-arena] and 597 [I-D.yang-mif-connection-manager-impl-req]. The arena connection 598 manager provides a means for applications to register their 599 connectivity requirement. The Connection Manager can then choose an 600 interface that matches the application's needs while considering 601 other factors such as availability, cost and stability. Also, the 602 Connection Manager can handle multiple-interfaces issues such as 603 connection sharing. 605 3.2. Desktop Operating Systems 607 Multi-interface issues also occur in desktop environments in those 608 cases where a desktop host has multiple (logical or physical) 609 interfaces connected to networks with different reachability 610 properties, such as one interface connected to the global Internet, 611 while another interface is connected to a corporate VPN. 613 3.2.1. Microsoft Windows 615 The multi-interface functionality currently implemented in Microsoft 616 Windows operation systems is described in more detail in 617 [I-D.montenegro-mif-multihoming]. 619 3.2.1.1. First hop selection 621 It is possible, although not often desirable, to configure default 622 routers on more than one Windows interface. In this configuration, 623 Windows will use the default route on the interface with the lowest 624 routing metric (i.e. the fastest interface). If multiple interfaces 625 share the same metric, the behavior will differ based on the version 626 of Windows in use. Prior to Windows Vista, the packet would be 627 routed out of the first interface that was bound to the TCP/IP stack, 628 the preferred interface. In Windows vista, host-to-router load 629 sharing [RFC4311] is used for both IPv4 and IPv6. 631 3.2.1.2. Outbound and Inbound Addresses 633 If the source address of the outgoing packet has not been determined 634 by the application, Windows will choose from the addresses assigned 635 to its interfaces. Windows implements [RFC3484] for source address 636 selection in IPv6 and, in Windows Vista, for IPv4. Prior to Windows 637 Vista, IPv4 simply chose the first address on the outgoing interface. 639 For incoming packets, Windows will check if the destination address 640 matches one of the addresses assigned to its interfaces. Windows has 641 implemented the weak host model [RFC1122] on IPv4 in Windows 2000, 642 Windows XP and Windows Server 2003. The strong host model became the 643 default for IPv4 in Windows Vista and Windows server 2008, however 644 the weak host model is available via per-interface configuration. 645 IPv6 has always implemented the strong host model. 647 3.2.1.3. DNS Configuration 649 Windows largely relies on suffixes to solve DNS resolution issues. 650 Suffixes are used for four different purposes that are reminded 651 hereafter: 653 1. DNS Suffix Search List (aka domain search list): suffix is added 654 to non-FQDN names. 656 2. Interface-specific suffix list, which allows sending different 657 DNS queries to different DNS servers. 659 3. Suffix to control Dynamic DNS Updates: determine which DNS server 660 will receive a dynamic update for a name with a certain suffix. 662 4. Suffix in the Name Resolution Policy Table [NRPT] to aid in 663 identifying a Namespace that requires special handling (feature 664 available only after Windows 7 and its server counterpart, 665 Windows Server 2008 R2). 667 However, this section focuses on the interface-specific suffix list 668 since it is the only suffix usage in the scope of this document. 670 DNS configuration information can be host-wide or interface specific. 671 Host-wide DNS configuration is input via static configuration or, in 672 sites that use Active Directory, Microsoft's Group Policy. Interface 673 specific DNS configuration can be input via static configuration or 674 via DHCP. 676 The host-wide configuration consists of a primary DNS suffix to be 677 used for the local host, as well as a list of suffix that can be 678 appended to names being queried. Before Windows Vista and Windows 679 Server 2008, there was also a host-wide DNS server list that took 680 precedent over per-interface DNS configuration. 682 The interface-specific DNS configuration comprises an interface- 683 specific suffix list and a list of DNS server IP addresses. 685 Windows uses a host-wide "effective" server list for an actual query, 686 where the effective server list may be different for different names. 687 In the list of DNS server addresses, the first server is considered 688 the "primary" server, with all other servers being secondary. 690 When a DNS query is performed in Windows, the query is first sent to 691 the primary DNS server on the preferred interface. If no response is 692 received in one second, the query is sent to the primary DNS servers 693 on all interfaces under consideration. If no response is received 694 for 2 more seconds, the DNS server sends the query to all of the DNS 695 servers on the DNS server lists for all interfaces under 696 consideration. If the host still doesn't receive a response after 4 697 seconds, it will send to all of the servers again and wait 8 seconds 698 for a response. 700 3.2.2. Linux and BSD-based Operating Systems 701 3.2.2.1. First hop selection 703 In addition to the two commonly used routing tables (the local and 704 main routing tables), the kernel can support up to 252 additional 705 routing tables which can be added in the file /etc/iproute2/ 706 rt_tables. A routing table can contain an arbitrary number of 707 routes, the selection of route is classically made according to the 708 destination address of the packet. Linux also provides more flexible 709 routing selection based on the Type of Service, scope, output 710 interface. In addition, since kernel version 2.2, Linux supports 711 policy based routing using the multiple routing tables capability and 712 a routing policy database. This database contains routing rules used 713 by the kernel. Using policy based routing, the source address, the 714 ToS flags, the interface name and an "fwmark" (a mark carried through 715 added in the data structure representing the packet) can be used as 716 route selectors. 718 Policy based routing can be used in addition to Linux packet 719 filtering capabilities, e.g provided by the "iptables" tool. In a 720 multiple interfaces context, this tool can be used to mark the 721 packets, i.e assign a number to fwmark, in order to select the 722 routing rule according to the type of traffic. This mark can be 723 assigned according to parameters like protocol, source and/or 724 destination addresses, port number and so on. 726 Such a routing management framework allows to deal with complex 727 situation such as address space overlapping. In this situation, the 728 administrator can use packet marking and policy based routing to 729 select the correct interface. 731 3.2.2.2. Outbound and Inbound Addresses 733 By default, source address selection follows the following basics 734 rules: the initial source address for an outbound packet can be 735 chosen by the application using the bind() call. Without information 736 from the application, the kernel chooses the first address configured 737 on the interface which belongs to the same subnet than the 738 destination address or the nexthop router. 740 Linux also implements [RFC3484] for source address selection for IPv6 741 and dual-stack configurations. However, the address sorting rules 742 from [RFC3484] are not always adequate. For this reason, Linux 743 allows the system administrator to dynamically change the sorting. 744 This can be achieved with the /etc/gai.conf file. 746 For incoming packets, Linux checks if the destination address matches 747 one of the addresses assigned to its interfaces then, processes the 748 packet according the configured host model. By default, Linux 749 implements the weak host model [RFC1122] on both IPv4 and IPv6. 750 However, Linux can also be configured to support the strong host 751 model. 753 3.2.2.3. DNS Configuration 755 Most BSD and Linux distributions rely on their DHCP client to handle 756 the configuration of interface-specific information (such as an IP 757 address and netmask), and a set of system-wide configuration 758 information, (such a DNS server list, an NTP server list and default 759 routes). Users of these operating systems have the choice of using 760 any DHCP client available for their platform, with an operating 761 system default. This section discusses the behavior of several DHCP 762 clients that may be used with Linux and BSD distributions. 764 The Internet Systems Consortium (ISC) DHCP Client [ISCDHCP] and its 765 derivative for OpenBSD [OPENBSDDHCLIENT] can be configured with 766 specific instructions for each interface. However, each time new 767 configuration data is received by the host from a DHCP server, 768 regardless of which interface it is received on, the DHCP client 769 rewrites the global configuration data, such as the default routes 770 and the DNS server list (in /etc/resolv.conf) with the most recent 771 information received. Therefore, the last configured interface 772 always become the primary one. The ISC DHCPv6 client behaves 773 similarly. 775 The Phystech dhcpcd client [PHYSTECHDHCPC] behaves similarly to the 776 ISC client. It replaces the DNS server list in /etc/resolv.conf and 777 the default routes each time new DHCP information is received on any 778 interface. However, the -R flag can be used to instruct the client 779 to not replace the DNS servers in /etc/resolv.conf. However, this 780 flag is a global flag for the DHCP server, and is therefore 781 applicable to all interfaces. When dhcpd is called with the -R flag, 782 the DNS servers are never replaced. 784 The pump client [PUMP] also behaves similarly to the ISC client. It 785 replaces the DNS servers in /etc/resolv.conf and the default routes 786 each time new DHCP information is received on any interface. 787 However, the nodns and nogateway options can be specified on a per 788 interface basis, enabling the user to define which interface should 789 be used to obtain the global configuration information. 791 The udhcp client [UDHCP] is often used in embedded platforms based on 792 busybox. The udhcp client behaves similarly to the ISC client. It 793 rewrites default routes and the DNS server list each time new DHCP 794 information is received. 796 Redhat-based distributions, such as Redhat, Centos and Fedora have a 797 per-interface configuration option (PEERDNS) that indicates that the 798 DNS server list should not be updated based on configuration received 799 on that interface. 801 The most configurable DHCP clients can be set to define a primary 802 interface to use only that interface for the global configuration 803 data. However, this is limited, since a mobile host might not always 804 have the same set of interfaces available. Connection managers may 805 help in this situation. 807 Some distributions also have a connection manager. However, most 808 connection managers serve as a GUI to the DHCP client, therefore not 809 changing the functionality described above. 811 3.3. Focus on access network selection 813 This section describes behaviors of connection managers in presence 814 of multiple points of attachment for a same interface. The section 815 focuses on WLAN access technology, it is described how does the 816 connection manager deal with the list of preferred SSID and how does 817 it select the access point for attachment. Desktops are not covered 818 since many different connection managers can be easily installed, 819 thus making hard to report a common behaviour. This section only 820 focuses on a specific use-case and some of the current 821 implementations; however further considerations on network discovery 822 and selection can be found in [RFC5113]. [RFC5113] describes the 823 network discovery and selection and discusses limitations and 824 constraints on potential solutions. 826 3.3.1. Android/HTC magic 828 When the terminal is under coverage of different WLAN networks with 829 different SSIDs: 831 The connection manager selects the first SSID of the the list of 832 preferred access network. The connection manager constructs the 833 list of preferred SSID giving priority to the last SSID on which 834 it has managed to attach. The user is not allowed to define its 835 preferred access. So, if the terminal discovers and manages to 836 attach to SSID1, SSID1 becomes the preferred access for future 837 attachment. If the terminal moves out of SSID1 coverage and 838 attaches to a new SSID, e.g. SSID2. SSID2 will then be the 839 preferred access of the connection manager. Then, if the terminal 840 processes to WLAN attachment within both SSID1 and SSID2 coverage, 841 the connection manager will select SSID2 for attachment. 843 When the layer 2 fails to attach to the selected SSID, the 844 connection manager automatically selects the next SSID in the list 845 of preferred SSID. Fallback come into play at expiration of a 846 timeout which can be configured. 848 When the IP stack fails to obtain an IP address after successful 849 WLAN attachment, the handset tries to connect to the next SSID in 850 the list. 852 When the terminal receives signals from different point of attachment 853 with same SSID: 855 The connection manager selects the point of attachment with best 856 signal strength; no other criteria (e.g. MAC address) is taken 857 into account. If the handset fails to attach to the selected 858 point of attachment (e.g. due to L2 authentication failure), the 859 connection manager selects the next point of attachment with 860 higher signal strength. If no more points of attachment, with 861 this SSID, is available, the connection manager selects the next 862 SSID in the list of preferred SSID. 864 If the terminal is unable to set up the IP connectivity on one 865 WLAN access, the connection manager will not try to attach to an 866 alternative point of attachment, or to an alternative SSID, as 867 long as the signal strength of the first radio link is the most 868 powerful. In other words, fallback on L3 attachment failure is 869 not supported if the terminal remains under coverage of the same 870 WLAN access point. 872 3.3.2. RIM BlackBerry 874 When the terminal is under coverage of different WLAN networks with 875 different SSIDs: 877 The device scans the WLAN frequency band. Then, if some SSIDs are 878 in the list of preferred SSID, the connection manager selects the 879 top of this list. The user can define its list of preferred 880 accesses. This list can be also enriched after successful 881 attachment to a new SSID. If so, the SSID is added at the end of 882 the list. 884 When the layer 2 fails to attach to the selected SSID, the 885 connection manager automatically selects the next SSID in the list 886 of preferred SSID. Fallback come into play at expiration of a 887 timeout which can be configured. 889 When the IP stack fails to obtain an IP address after successful 890 WLAN attachment, the BlackBerry tries to connect to the next SSID 891 in the list. 893 When the terminal receives signals from different point of attachment 894 with same SSID: 896 The connection manager selects the point of attachment with best 897 signal strength; no other criteria (e.g. MAC address) is taken 898 into account. If the handset fails to attach to the selected 899 point of attachment (e.g. due to L2 authentication failure), the 900 connection manager selects the point of attachment with lower 901 signal strength. If no more points of attachment (corresponding 902 to the preferred SSID) are available, the connection manager 903 selects the next SSID in the list of preferred SSID. 905 The connection manager always selects the most powerful signal 906 strength without considering IP configuration results. If the 907 terminal is unable to set up the IP connectivity on one WLAN 908 access, the connection manager will not try to attach to an 909 alternative point of attachment, or alternative SSID, as long as 910 the signal strength of the current access point is the most 911 powerful. If the user is moving and if the signal strength of an 912 alternative point of attachment become better, the terminal 913 automatically restarts layer 2 and IP connectivity processes on 914 that alternative access point. 916 4. Acknowledgements 918 Authors of the document would like to thank following people for 919 their input and feedback: Dan Wing, Hui Deng, Jari Arkko, Julien 920 Laganier and Steinar H. Gunderson. 922 5. IANA Considerations 924 This memo includes no request to IANA. 926 6. Security Considerations 928 This document describes current operating system implementations and 929 how they handle the issues raised in the MIF problem statement. 930 While it is possible that the currently implemented mechanisms 931 described in this document may affect the security of the systems 932 described, this document merely reports on current practice. It does 933 not attempt to analyze the security properties (or any other 934 architectural properties) of the currently implemented mechanisms. 936 7. Contributors 938 The following people contributed most of the per-Operating System 939 information found in this document: 941 o Marc Blanchet, Viagenie 943 o Hua Chen, Leadcoretech, Ltd. 945 o Yan Zhang, Leadcoretech Ltd. 947 o Shunan Fan, Huawei Technology 949 o Jian Yang, Huawei Technology 951 o Gabriel Montenegro, Microsoft Corporation 953 o Shyam Seshadri, Microsoft Corporation 955 o Dave Thaler, Microsoft Corporation 957 o Kevin Chin, Microsoft Corporation 959 o Teemu Savolainen, Nokia 961 o Tao Sun, China Mobile 963 o George Tsirtsis, Qualcomm. 965 o David Freyermuth, France telecom. 967 o Aurelien Collet, Altran. 969 o Giyeong Son, RIM. 971 8. References 973 8.1. Normative References 975 [I-D.ietf-mif-problem-statement] 976 Blanchet, M. and P. Seite, "Multiple Interfaces and 977 Provisioning Domains Problem Statement", 978 draft-ietf-mif-problem-statement-13 (work in progress), 979 April 2011. 981 8.2. Informative References 983 [ANDROID] Google Inc., "Android developers: package android.net", 984 2009, . 987 [ANDROID-RFC3484] 988 Gunderson, S., "RFC 3484 support for Android", 2010, . 992 [BLACKBERRY] 993 Research In Motion Limited, "BlackBerry Java Development 994 Environment - Fundamentals Guide: Wireless gateways", 995 2009, . 998 [I-D.montenegro-mif-multihoming] 999 Montenegro, G., Thaler, D., and S. Seshadri, "Multiple 1000 Interfaces on Windows", 1001 draft-montenegro-mif-multihoming-00 (work in progress), 1002 March 2009. 1004 [I-D.yang-mif-connection-manager-impl-req] 1005 Yang, J., Sun, T., and S. Fan, "Multi-interface Connection 1006 Manager Implementation and Requirements", 1007 draft-yang-mif-connection-manager-impl-req-00 (work in 1008 progress), March 2009. 1010 [I-D.zhang-mif-connection-manager-arena] 1011 Zhang, Y., Sun, T., and H. Chen, "Multi-interface Network 1012 Connection Manager in Arena Platform", 1013 draft-zhang-mif-connection-manager-arena-00 (work in 1014 progress), February 2009. 1016 [ISCDHCP] Internet Software Consortium, "ISC DHCP", 2009, 1017 . 1019 [NRPT] Windows, "Name Resolution Policy Table", February 2010, < 1020 http://technet.microsoft.com/en-us/magazine/ 1021 ff394369.aspx>. 1023 [OPENBSDDHCLIENT] 1024 OpenBSD, "OpenBSD dhclient", 2009, 1025 . 1027 [PHYSTECHDHCPC] 1028 Phystech, "dhcpcd", 2009, 1029 . 1031 [PUMP] RedHat, "PUMP", 2009, . 1033 [RFC1122] Braden, R., "Requirements for Internet Hosts - 1034 Communication Layers", STD 3, RFC 1122, October 1989. 1036 [RFC3484] Draves, R., "Default Address Selection for Internet 1037 Protocol version 6 (IPv6)", RFC 3484, February 2003. 1039 [RFC4311] Hinden, R. and D. Thaler, "IPv6 Host-to-Router Load 1040 Sharing", RFC 4311, November 2005. 1042 [RFC5113] Arkko, J., Aboba, B., Korhonen, J., and F. Bari, "Network 1043 Discovery and Selection Problem", RFC 5113, January 2008. 1045 [S60] Nokia Corporation, "S60 Platform: IP Bearer Management", 1046 2007, . 1050 [UDHCP] Busybox, "uDHCP", 2009, . 1053 [WINDOWSMOBILE] 1054 Microsoft Corporation, "SDK Documentation for Windows 1055 Mobile-Based Smartphones: Connection Manager", 2005, 1056 . 1058 [WNDS-RFC3484] 1059 Microsoft Corporation, "SDK Documentation for Windows 1060 Mobile-Based Smartphones: Default Address Selection for 1061 IPv6", 2005, 1062 . 1064 Authors' Addresses 1066 Margaret Wasserman 1067 Painless Security, LLC 1068 356 Abbott Street 1069 North Andover, MA 01845 1070 USA 1072 Phone: +1 781 405-7464 1073 Email: mrw@painless-security.com 1074 URI: http://www.painless-security.com 1075 Pierrick Seite 1076 France Telecom - Orange 1077 4, rue du clos courtel BP 91226 1078 Cesson-Sevigne 35512 1079 France 1081 Email: pierrick.seite@orange-ftgroup.com