Strategies for securely applying connection policies via a gateway

STRATEGIES FOR SECURELY APPLYING

CONNECTION POLICIES VIA A GATEWAY

BACKGROUND

0004[0001] A terminal service (TS) system allows a TS client to interact with an application being run on a remote TS server. A user who interacts with the application receives generally the same user experience that would be provided if the application were implemented locally by the TS client. Implementing the application on the TS server device affords a number of benefits. For instance, it may be easier to administer an application that is maintained at a central location. [0002] In a first case, an organization (such as a corporation) may include a TS server that is accessible to a group of clients within the organization. In this scenario, the TS clients can access the TS server without special security provisions because the TS clients are presumed to be trusted entities. In a second case, the organization may wish to make its TS server available to another group of clients that are located outside the organization. For example, the organization may wish to allow employees to access its terminal server from their home via the Internet. In this scenario, the TS clients cannot access the TS server without special security provisions. According to one solution, the organization may use a gateway that sits behind a firewall to administer the interaction between external TS clients and the TS server.

0005[0003] However, there are potential vulnerabilities in above-described type of architecture. Consider the case of device redirection. In a TS system, device redirection allows a user who is interacting with a TS server to utilize devices that are associated with the TS server and devices that are associated with the TS client. For example, a user who is interacting with a word processing program that is being run on the TS server can specify that information is to be saved and retrieved from a storage device that is local with respect to the user's client device. This may allow a malicious user (or other entity) to potentially corrupt the data processing infrastructure of the organization by uploading viruses and the like to the TS server. [0004] There is accordingly an exemplary need to improve the security of TS systems (and the like) that employ a gateway. SUMMARY

0006[0005] A strategy is described for securely applying connection policies in a system that includes a first entity connected to a second entity via a gateway using a remote-operating protocol. The first entity can comprise a terminal service (TS) client, the second entity can comprise a TS server, and the remote-operating protocol can comprise Remote Desktop Protocol (RDP). The strategy involves establishing a first secure channel between the gateway and the TS server and transmitting policy information from the gateway to the TS server. The strategy then involves deactivating the first secure channel and setting up a second secure channel between the TS client and the TS server. The strategy uses the second secure channel to transmit RDP data from the TS client to the TS server. The TS server uses the previously-transmitted policy information to determine whether to enable or disable a feature that affects the TS client, such as device redirection. [0006] The strategy confers a number of benefits. According to one exemplary benefit, the strategy allows a trusted TS server to enforce policy information that affects the TS client, rather than relying on the TS client (which may be malicious). According to another benefit, the strategy provides a centric management of policy; versus group policy, this facilitates each user policy management. According to another benefit, the additional secure channel is deployed between the gateway and the TS server; since these entities are "close," there is little extra delay for building up a connection.

0007[0007] Additional exemplary implementations and attendant benefits are described in the following.

0008BRIEF DESCRIPTION OF THE DRAWINGS [0008] Fig. 1 shows an exemplary terminal service (TS) system that uses a first paradigm to transmit data from a TS client to a TS server; the first paradigm may expose the TS system to malicious activity by the TS client.

0009[0009] Fig. 2 is a flowchart that illustrates an exemplary manner of operation of the TS system of Fig. 1. [0010] Fig. 3 shows an exemplary TS system that uses a second paradigm to transmit data from a TS client to a TS server; the second paradigm provides safeguards against malicious activity by the TS client.

0010[0011] Fig. 4 is a flowchart that illustrates an exemplary manner of operation of the TS system of Fig. 3.

0011[0012] Figs. 5 and 6 show one exemplary implementation of the TS system of Fig. 3.

0012[0013] Fig. 7 shows exemplary processing functionality that can be used to implement any aspect of the systems of Figs. 3 and 5. [0014] The same numbers are used throughout the disclosure and figures to reference like components and features. Series 100 numbers refer to features originally found in Fig. 1, series 200 numbers refer to features originally found in Fig. 2, series 300 numbers refer to features originally found in Fig. 3, and so on.

0013DETAILED DESCRIPTION [0015] This disclosure sets forth a strategy for securely applying connection policies in a system that includes a first entity connected to a second entity via a gateway using a remote-operating protocol. The strategy can be manifested in various systems, apparatuses, modules, procedures, storage mediums, data structures, and other forms. [0016] This disclosure includes the following sections. Section A describes an exemplary system for transmitting data using a first paradigm. The first paradigm may expose the system to certain security threats. Section B describes an exemplary system for transmitting data using a second paradigm. The second paradigm provides enhanced security compared to the first paradigm. Section C describes exemplary processing functionality that can be used to implement any aspect of the systems described in Sections A and B.

0014[0017] As a preliminary note, any of the functions described with reference to the figures can be implemented using software, firmware, hardware (e.g., fixed logic circuitry), manual processing, or a combination of these implementations. The term "logic, "module," "component," "system" or "functionality" as used herein generally represents software, firmware, hardware, or a combination of the elements. For instance, in the case of a software implementation, the term "logic," "module," "component," "system," or "functionality" represents program code that performs specified tasks when executed on a processing device or devices (e.g., CPU or CPUs). The program code can be stored in one or more computer readable memory devices.

0015[0018] More generally, the illustrated separation of logic, modules, components, systems, and functionality into distinct units may reflect an actual physical grouping and allocation of software, firmware, and/or hardware, or can correspond to a conceptual allocation of different tasks performed by a single software program, firmware program, and/or hardware unit. The illustrated logic, modules, components, systems, and functionality can be located at a single site (e.g., as implemented by a processing device), or can be distributed over plural locations. [0019] The terms "machine-readable media" or the like refers to any kind of medium for retaining information in any form, including various kinds of storage devices (magnetic, optical, static, etc.). The term machine-readable media also encompasses transitory forms for representing information, including various hardwired and/or wireless links for transmitting the information from one point to another. [0020] Certain aspects of the technology set forth herein are explained in flow chart form. In these flow charts, certain operations are described as constituting distinct blocks performed in a certain order. Such implementations are exemplary and non-limiting. Certain blocks can be grouped together and performed in a single operation, and certain blocks can be performed in an order that differs from the illustrated order. The blocks shown in the flowcharts can be implemented by software, firmware, hardware, manual processing, any combination of these implementations .

0016[0021] Certain aspects of the technology set forth herein are explained in the context of a terminal service (TS) system that uses a remote operating protocol. For example, without limitation, the technology can be implemented using the Remote Desktop Protocol (RDP). A TS system using RDP allows a TS client to interact with one or more applications maintained on a remote TS server. The user receives the same user experience that would be provided if the applications were being locally run on the TS client. While the following explanation uses the terms "TS client" (instead of "first entity"), "TS server" (instead of "second entity"), and "RDP" (instead of "remote operating protocol"), it should be noted that the principles described herein can be applied to other implementations.

0017A. System and Method Using First Paradigm

0018[0022] Fig. 1 shows a TS system 100 for transmitting RDP data between a TS client 102 and a TS server 104 using a first paradigm. As noted above, the first paradigm may expose the system 100 to certain malicious activity that the TS client 102 may perform. Thus, Section A mainly acts as a vehicle for more effectively communicating the solutions described in the next section.

0019[0023] The TS server 104 is provided within a trusted environment 106. The trusted environment 106 may correspond to a data processing infrastructure provided by any type of organization, such as a corporation, educational institution, governmental institution, and so on. Although not shown, the trusted environment 106 may generally include a collection of server computers, networks (such as intranets), "internal" clients, routers, data storage devices, and so on. The trusted environment 106 is trusted in the sense that there is an expectation that entities that interact with the TS server 104 from within the trusted environment 106 can be generally relied on to not deliberately cause damage to the trusted environment 106.

0020[0024] The TS client 102 is external to the trusted environment 106. For example, the TS client 102 may represent any type of client device that interacts with the TS server 104 via a network 108. For example, the client device 102 can comprise a personal computer, a laptop computer, a personal digital assistant (PDA), a stylus-type input device, a mobile telephone, a game console, a set-top box associated with a television set, and so on. The network 108 can comprise any type of communication mechanism that is not under the control of the trusted environment 106. For example, the network 108 can comprise the Internet. [0025] The TS system 100 uses a firewall 110 and a gateway 112 to allow the external TS client 102 to access the TS server 104. The firewall 110 can comprise a conventional mechanism for restricting the type of data being transmitted to the trusted environment 106 based on predefined rules. The gateway 112 acts as a proxy for forwarding RDP data between the TS client 102 and the TS server 104. The gateway 112 applies various policies which govern what actions are allowed and disallowed when interacting with external TS clients.

0021[0026] The system 102 can use a single secure channel 114 to transmit RDP data between the TS client 102 and the TS server 104. The RDP data that is sent over this channel 114 is encrypted, so it cannot be acted upon by any entity midstream between the TS client 102 and the TS server 104. [0027] TS technology includes a feature referred to in the art as device redirection. First, consider the use of device redirection for clients within the trusted environment 106. When a user is interacting with an application being run on the TS server 104, device redirection allows the user to access devices associated with the TS server 104 as well as devices associated with the internal TS client. More specifically, the TS server 104 may have various devices associated therewith, such as various storage devices (associated with various drives), various printing devices, various scanning devices, various audio input and output devices, and so forth. Likewise, the internal TS client may have various devices associated therewith, including the same kinds of devices mentioned above. Device redirection allows a user to interact with any of these devices in the course of a TS session. For example, consider the case in which the user is using an internal TS client to interact with a word processing program being run on the TS server 104. The user can store and retrieve documents to/from storage locations that are considered local to the TS server 104. But the user can also store and retrieve documents to/from storage locations that are considered local to the internal client. [0028] Device redirection that occurs within the confines of the trusted environment 106 may not pose a significant risk. This is because, as stated above, the users and components within the trusted environment 106 are trusted to not deliberately cause damage to the trusted environment 106. However, device redirection in the context of the external TS client 102 may pose a risk. This is because the TS client 102 is situated outside the trusted environment 106, and therefore may be untrustworthy.

0022[0029] Consider the specific scenario in which the TS client 102 includes local devices (116, ... 118). A malicious user who is operating the TS client 102 can potentially introduce destructive code into the trusted environment 106 through the use of device redirection. For instance, consider the above example in which the user is interacting with a word processing program on the TS server 104 and wishes to retrieve a document from a drive that is local to a TS client. If this document contains destructive code or other harmful content, it can potentially cause damage to the TS server 104 or other parts of the trusted environment 106.

0023[0030] In certain circumstances, it may therefore be desirable to disable device redirection for the external TS client 102. Fig. 1 shows a first paradigm for accomplishing this result. However, the first paradigm has potential shortcomings. As will be explained below, there is a risk that a malicious external TS client can circumvent the provisions of the first paradigm and still cause damage to the trusted environment 106.

0024[0031] According to the first paradigm, the gateway 112 transmits policy information 120 to the TS client 102. The policy information 120 conveys the status of one or more features of a TS session between the TS client 102 and the TS server 104. For example, the policy information 120 can indicate whether these features are enabled or disabled. In the specific context of device redirection, the policy information 120 can be used to indicate that device redirection is disabled for external TS client 102. The policy information 120 can be expressed in various formats, such as a collection of one or more flags. [0032] The first paradigm is based on the expectation that the TS client 102 will forward the policy information 120 (or information derived therefrom) to the TS server 104 at the start of sending RDP data to TS server 104. (The RDP data reflects the normal flow of data that enables the TS client 102 to interact with the TS server 104.) The first paradigm is based on the further expectation that the TS server 104, upon receiving the policy information 120, will take appropriate action based on the status information contained therein. For example, if the policy information 120 informs the TS server 104 that it should disable device redirection, the TS server 104 will prevent the external TS client 102 from using device redirection.

0025[0033] Because the TS client 102 is external to the trusted environment 106, it cannot necessarily be trusted. As such, the TS client 102 may fail to carry out its proper role in forwarding the policy information 120 to the TS server 104. Alternatively, the TS client 102 can send tampered policy information 120 to the TS server 104. For example, the TS client 102 can change the policy information 102 to indicate that device redirection is enabled (when it should, in fact, be disabled). This will cause the TS server 104 to enable device redirection, which, in turn, provides an avenue through which the TS client 102 can upload destructive code to the TS server 104.

0026[0034] Fig. 2 is a procedure 200 which summarizes the above-described operation of the TS system 100. [0035] In block 202, the gateway 112 sends the policy information 120 to the external TS client 102.

0027[0036] In block 204, the TS client 102 connects to the TS server 104 via the gateway 112 to conduct a TS session, e.g., involving the exchange of RDP data. [0037] In block 206, the TS client 102 is expected to transmit the policy information to the TS server 104 when it starts sending RDP data. As described above, a malicious TS client 102 can subvert this operation in various ways. [0038] In block 208, the TS server 104 receives the policy information 120 and the RDP data. The TS server 104 conducts a TS session based the policy information 120. As described above, the policy information 120 may have the effect of enabling or disabling one or more features of a TS session, such as such as device redirection.

0028[0039] Note that, while the gateway 112 performs a role in policing the interaction between external TS clients and the TS server 104, the gateway 112 cannot gain direct access to the RDP data. This is because the RDP data is encrypted en route between the TS client 102 and the TS server 104. This presents various constraints on the use of the gateway 112 to directly disable device redirection.

0029B. System and Method Using Second Paradigm

0030[0040] A TS system 300 of Fig. 3 uses a second paradigm to transfer RDP data from a client 302 to a TS server 304. For frame of reference, the first TS system 100 of Fig. 1 was client-centric in the sense that it relied on the TS client 102 to disable device redirection. The solution provided in Fig. 3 is server-centric in the sense that it relies on the TS server 304 alone to disable device redirection. Since the TS server 304 is situated within a trusted environment 306, the second paradigm can more reliably disable device redirection compared to the first paradigm.

0031[0041] At the outset, it should be noted that the TS system 300 of Fig. 3 is described in the specific context of the disablement of device redirection. However, the TS paradigm can be applied to affect the status of any aspect of a TS session. In one general case, for instance, the second paradigm can allow the TS server 304 to disable any identified type of packet within the RDP data. [0042] The TS system 300 of Fig. 3 includes a similar contextual setting to the TS system 100 of Fig. 100. Namely, the TS client 302 interacts with the TS server 304 in the trusted environment 306. The TS client 302 can access the TS server 304 via a network 308 (such as the Internet), firewall 310, and gateway 312. Without limitation, according to one exemplary implementation, certain aspects of the gateway 312 of Fig. 3 can be implemented using technology described in co- pending U.S. Applications: (1) U.S Serial No. 11/326,992, filed on January 5, 2006, entitled "Providing Consistent Application Aware Firewall Traversal"; and (2) U.S. Serial No. 11/067,125, filed on February 25, 2005, entitled "Enabling Terminal Services through a Firewall." Both of these applications are incorporated by reference herein in their respective entireties.

0032[0043] The TS client 302 can exchange RDP data with the TS server 304 via a second secure channel 314. (This channel is called a "second" channel to distinguish it from a first channel which is established temporally prior to the second channel 314, to be explained in greater detail below). According to one exemplary implementation, to transmit RDP data 316 to the TS server 304, the TS system 300 can first wrap the RDP data 316 in an RPC-based protocol 318 (where RDC denotes Remote Procedure Call). This information, in turn, can then be layered over an HTTPS protocol 320 (where HTTPS denotes Hypertext Transfer Protocol Security). Prior to reaching the TS server 304, the TS system 300 can strip off the various layers to yield the RPC data 316 itself. Generally, the RDP data 316 transmitted via the second channel 314 is secure. For example, the RDP data can be protected using SSL (where SSL denotes Secure Socket Layers). [0044] The TS client 302 can interact with one or more local devices, such as devices (322, ... 324). These devices (322, ... 324) can include local storage devices, printers, audio input and output devices, and so forth. [0045] Now the features of the second paradigm will be set forth in greater detail. First note that the TS system 300 establishes a first secure channel 326 between the gateway 312 and the TS server 304. This channel 326 is used to transfer policy information 328 to the TS server 304 prior to the exchange of RDP data using the second channel 314. The policy information 328 conveys information regarding one or more features that affect the exchange of RDP data between the TS client 302 and the TS server 304. In one non-limiting case, the policy information 328 includes an instruction to disable device redirection for the TS client 302.

0033[0046] Like the second channel 314, the first channel 326 is secure. For instance, the TS system 300 can encrypt the information being transmitted via the first channel 326. In one implementation, the TS system 300 uses SSL for the first channel 326. [0047] The TS system 300 deactivates the first channel 326 before sending the RDP data using the second channel 314. Upon receipt of the RDP data via the second channel 314, the TS server 304 enables or disables certain features in accordance with the instructions conveyed by the policy information 328. Note that the exchange of policy information 328 to the TS server 304 does not rely on the good-faith actions of the TS client 302. Thus, the second paradigm is potentially more secure than the first paradigm used by the first TS system 100. [0048] Each TS session is governed by associated policy information. Thus, the TS system 300 may apply first policy information to a first TS client and second policy information to a second TS client, and so on. Further, the policy information may optionally be specific to each connection, such that the same TS client may receive first policy information when it connects to the TS server 304 at time X and receive second policy information when it connects to the TS server 304 at time Y. [0049] To address the above issue, the gateway 312 can send a token 330 to the TS client 302 when the TS client 302 requests a connection with the TS server 304. The role of the token 330 is to identify the TS client 302. When the gateway 312 forwards the policy information 328 to the TS server 304, it can also forward the token 330 assigned to this particular TS client 302 (and this particular connection). The TS server 304 can subsequently use the token 330 to apply the correct policy information 328 to the TS client 302. [0050] Fig. 4 shows a procedure 400 which summarizes the above-described operation of the TS system 300.

0034[0051] In block 402, the TS client 302 requests a connection to the TS server 304 to conduct an RDP session with the TS server 304.

0035[0052] In block 404, the gateway 312 passes the token 330 to the TS client 302. The token 330 is used to identify the TS client 302. [0053] In block 406, the gateway 312 and the TS server 304 establish the first secure channel 326. Through this channel 326, the gateway 312 sends the policy information 328 and token information 330 to the TS server 304. The policy information 328 identifies features of an RDP session which are enabled or disabled with respect to the particular TS client 302 and the particular connection, as identified by the token 330.

0036[0054] In block 408, the TS system 300 closes the first channel 326 and establishes the second channel 314.

0037[0055] In block 410, the TS client 302 sends RDP data with the token 330 to the TS server 304 via the second channel 314. The TS server 304 can apply the correct policy information 328 to the TS client 302 according to the token 330. [0056] In block 412, the TS server 304 can govern the ensuing exchange of RDP data based on the instructions contained in the policy information 328. In one particular case, this may involve disabling device redirection. This reduces the possibility that the TS client 302 can corrupt the trusted environment 306 through its local devices (322, ... 324).

0038[0057] Figs. 5 and 6 shows one exemplary implementation of the principles set forth in Figs. 3 and 4.

0039[0058] Fig. 5 indicates that the gateway 312 can include a protocol module 502 and a gateway security filter module 504. These components are relevant to the secure transfer of RDP data described herein. The policy module 502 generates the policy information 328 and the token 330. The gateway security filter module 504 establishes secure exchange of information from the perspective of the gateway 312. That is, the gateway security filter module 504 generates a security context and performs encryption and decryption. [0059] The TS server 306 can include a negotiation module 506 and a server security filter module 508. These components are relevant to the secure transfer of RDP data described herein. The negotiation module 506 acts on information sent to the TS server 306 in a manner set forth more fully below in the context of Fig. 6 below. The server security filter module 508 establishes secure exchange of information in the TS system 300 from the perspective of the TS server 304. (Although not shown, the TS client 302 includes a client security filter module for establishing the secure exchange of information from the perspective of the TS client 302.) [0060] Fig. 6 shows how the components set forth in Fig. 5 interact with each other, according to one exemplary implementation. The operations in Fig. 6 are numbered based on an exemplary order in which these operations can be performed. The ordering of these operations can be modified in various ways. [0061] In operation (1), the TS client 302 issues a channel creation request to the gateway 312. [0062] In operation (2), the gateway 312 generates the token 330 and sends the token 330 to the TS client 302. The token 330 can be expressed as a unique GUID. [0063] In operation (3), the gateway 312 activates the gateway security filter module 504.

0040[0064] In operation (4), the gateway security filter module 504 sends a SSL hello to the TS server 304. [0065] In operation (5), the server negotiation module 506 recognizes the SSL hello message and activates the server security filter module 508.

0041[0066] In operation (6), an SSL handshake occurs between the gateway 312 and the TS server 306 to establish the first secure channel 326. Schannel can be used for this security channel 326. [0067] In operation (7), the gateway 312 generates policy information 328. In this example, the policy information 328 takes the form a device redirection data unit. The gateway 312 sends the device redirection data unit to the TS server 304 along with the token 330.

0042[0068] In operation (8), the gateway 312 sends a reset command to the TS server 304.

0043[0069] In operation (9), the TS server 304 instructs the server security filter module 508 to be ready for restarting.

0044[0070] In operation (10), the TS client 302 sends messages to the TS server 304 via the gateway 312. More specifically, a first message sent by the TS client 302 is a protocol negotiation message or a SSL hello. This message will pass through the gateway security filter module 504 and the server security filter module 508. This message can possibly reach the gateway 312 before the gateway/server authentication. If so, the message remains waiting to be handled.

0045[0071] In operation (11), the gateway 312 deactivates the gateway security filter module 504 after sending out the above-described message (in operation 10).

0046[0072] In operation (12), the TS client 302 and the TS server 204 perform authorization to establish the second channel 314.

0047[0073] In operation (13), the TS client 302 sends a first RDP packet along with the stored token 330 to the TS server 304. [0074] In operation (14), the TS server 304 enables or disables device redirection based on the policy information 328 and the token 330. C. Exemplary Processing Functionality

0048[0075] Fig. 7 sets forth exemplary processing functionality 702 that can be used to implement any aspect of systems shown in Figs. 3 and 5. In one non-limiting case, for instance, the processing functionality 702 may represent any TS client, any TS server, any computer used by the gateway, and so on.

0049[0076] The processing functionality 702 can include various volatile and nonvolatile memory, such as RAM 704 and ROM 706, as well as one or more central processing units (CPUs) 708. The processing functionality 702 can perform various operations identified above when the CPU 708 executes instructions that are maintained by memory (e.g., 704, 706, or elsewhere). The processing functionality 702 also optionally includes various media devices 710, such as a hard disk module, an optical disk module, and so forth.

0050[0077] The processing functionality 702 also includes an input/output module 712 for receiving various inputs from the user (via input devices 714), and for providing various outputs to the user (via output devices 716). One particular output device may include a display apparatus and an associated graphical user interface (GUI) 718. The processing functionality 702 can also include one or more network interfaces 720 for exchanging data with other devices via one or more communication conduits 722. One or more communication buses 724 communicatively couple the above-described components together.

0051[0078] The communication conduits 722 can be implemented in different ways to suit different technical and commercial environments. For instance, the communication conduits 722 can include any kind of network (or combination of networks), such as a wide area network (e.g., the Internet), an intranet, Digital Subscriber Line (DSL) network infrastructure, point-to-point coupling infrastructure, and so on. In the case where one or more digital networks are used to exchange information, the communication conduits 722 can include various hardwired and/or wireless links, routers, gateways, name servers, and so on. The communication conduits 722 can be governed by any protocol or combination of protocols. [0079] In closing, a number of features were described herein by first identifying exemplary problems that these features can address. This manner of explication does not constitute an admission that others have appreciated and/or articulated the problems in the manner specified herein. Appreciation and articulation of the problems present in the relevant art(s) is to be understood as part of the present invention.

0052[0080] More generally, although the invention has been described in language specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claimed invention.