This invention relates generally to diagnosis of network problems, and more particularly, for example, to systems and methods for remote troubleshooting of network devices having private Internet Protocol (IP) addressing.
Internet-based service providers (ISPs) are beginning to offer more services to subscribers utilizing network devices deployed behind gateways, such as a home gateway, in subscriber's networks. Voice Over Internet Protocol (VoIP) services, in particular, are gaining momentum, while video-over-IP services such as video conferencing and video-on-demand are expected to follow. These new services are often delivered using the subscriber's existing broadband connection. This deployment option opened up an opportunity for a new breed of independent service providers to enter the market and compete with established access providers. The competition is expected to heat up and is forcing service providers to look for very cost efficient options for rolling out and managing these services.
The process of configuring a network device and establishing service is known as provisioning. Some provisioning systems are used to automatically configure devices using a variety of configuration protocols such as Data Over Cable Service Interface Specification (DOCSIS), the PacketCable interface specification, the Customer Premises Equipment (CPE) Wide Area Network (WAN) management protocol according to Technical Report TR-069, and eXtensible Markup Language (XML) Configuration Access Protocol (XCAP), for example. However, once the device is configured, a user may wish to customize certain settings for their environment. For example, some VoIP devices come with a built-in wireless gateway and consumers may wish to setup local wireless network security. If the user encounters difficulties in the customization of their device and/or Local Area Network (LAN) settings, they may seek assistance from their service provider. Instead of building sophisticated automated subscriber support systems, some service providers opt for a more direct solution that relies on remotely accessing a subscriber device's configuration user interface (UI) by support personnel. Many devices provide a web-based UI intended for local access by the subscriber and remote access by the service provider.
Troubleshooting using remote access to subscriber device's UI is not the only mechanism for remote troubleshooting and assistance, but it has been utilized by some service providers that find its simplicity appealing. In the past and presently, this model is used for remote troubleshooting of home gateways with public Internet Protocol (IP) connectivity. More recently, these service providers have expressed desire to extend this support model to devices deployed behind the home gateway which often have private local IP address assigned by the DHCP server in the home gateway.
Devices behind a home gateway are typically shielded from remote entities initiating connections to them by virtue of the home gateway functioning as a firewall and network address translation (NAT) router. In this environment, it becomes impossible for service providers to use the device UI remotely because (a) the device IP address is not known and (b) there is no standard mechanism to initiate a remote HTTP connection to a NAT'ed device. Therefore, there remains a need in the art for systems and methods that address the problems of accessing a client device that has a private address.
Embodiments of the present invention and their advantages are best understood by referring to the detailed description that follows. It should be appreciated that like reference numerals are used to identify like elements illustrated in one or more of the figures.
Devices, methods and systems are disclosed herein, in accordance with one or more embodiments of the present invention, that provide remote administration or device management for client devices with private or local IP addresses, including those connected to a network behind a firewall, Network Address Translation (NAT) router, or a gateway, where a gateway may be considered generally as any network traffic router which connects two networks. According to one or more embodiments of the present invention, the first part of the present invention involves establishing a communication connection between a client device with a private address and a proxy server, while the second part of the present invention involves establishing relay communications from an administrator device, or admin device, to the client device through the proxy server. The client device is separated from the admin device by at least one intermediate NAT so that only the client device can initiate the communication connection. Once the relay communications of the remote access system are established, an administrator may perform remote administration or troubleshooting of the client device and diagnosis of network problems. An administrator is enabled to access a client device troubleshooting User Interface (UI) using a proxy in spite of the client device having a private IP address due to the presence of an intermediate NAT. The client device may maintain a path through the NAT where the path is used by the proxy server to instruct the client device to establish a connection when it is needed. Once the connection is established, the proxy server may then relay message traffic, such as HyperText Transfer Protocol (HTTP) traffic, between an admin device user interface and the client device.
Private networks, or networks having devices with private addresses, are common and may be used where an organization does not require globally unique addresses for every network device or where there is a shortage of available addresses. In addition, private addressing provides a basic form of security since it is not possible for an external device outside the private network to initiate a direct connection to a device on the private network, since the private network address is not known to the external device. In this manner, the client device on a private network may have an address that is either not known to or is not directly accessible from an outside network. This is one example of a type of barrier that a NAT gateway, router, or firewall can create. NATs may be used anywhere within a network hierarchy, and not merely on the boundary between public and private networks.
First client device 102, henceforth device 102, and second client device 104, henceforth device 104, may each be any network device including a computer, a user terminal, a router, a gateway, a hub, an access point, a Voice Over Internet Protocol (VoIP) telephone, a television set-top decoder box, or other device for use with a service provider. Device 102 and device 104 may each be considered Customer Premises Equipment (CPE) since they typically are present at a customer work-site, business, or home and are leased or owned by the customer. Device 102 communicates with gateway 106 through a channel 118 that can be a wired or a wireless connection for passing messages according to a network protocol such as the Transfer Control Protocol/Internet Protocol (TCP/IP), where messages are routed on the network based on sender and receiver addresses. Similarly, device 104 communicates with gateway 106 through a channel 120 that can also be a wired or a wireless connection for passing messages according to a network protocol. Although in this exemplary embodiment LAN 108 includes only three devices in a star network topology, this is not considered limiting since a number of different network topologies and a larger or smaller number of network devices may comprise LAN 108. Gateway 106 communicates with second network 110 through a channel 122 that can be a wired or a wireless connection for passing messages according to a network protocol. Admin device 114 communicates with second network 110 through a channel 124 that can be a wired or a wireless connection for passing messages according to a network protocol. Similarly, proxy server 116 communicates with the second network 110 through a channel 126
Device 102 may include a suitably programmed client processor 130 configured to execute computer instructions, a client memory 132 configured to store and retrieve information including a client device identifier 134, a client user interface 136 configured to receive data from and present data to a client user and provide an administrative user interface for device 102, and/or a client communication unit 138 configured to send and receive messages on channel 118. Client memory 132 may include a Random Access Memory (RAM), a disc memory, an optical memory, a magnetic memory device, a register file, and/or any technology for storing and retrieving information for use by client processor 130 or any device 102 resource. Further, client memory 132 may be removable from device 102 to provide for safekeeping of the stored information, ease of maintenance, and/or re-configurability. Client device identifier 134 may be a serial or identification number, comprising a text string of alpha and/or numeric characters, a user selected text string, an administrator selected text string, or some other information used to identify client device 102. Similarly, device 104 may also include a client processor, a client memory, a client device identifier, a client user interface, and/or a client communication unit configured to send and receive messages on channel 120.
Client user interface 136 can be a web browser or other communication program running on client processor 130 or another processor. Alternatively, client user interface 136 may be a device or method used to communicate over LAN 108 from device 102, and/or provide an administrator user interface from which a user or an administrator may change, configure, and/or provision a service associated with a service provider for device 102 or a connected resource, including the service through which device 102 operates. Although an administrator may be a person such as a customer support representative from a service provider, the administrator may alternatively be an autonomous client program or diagnostic program executing on admin device 114 such as a troubleshooting script tailored for or adapted for use with device 102. Client communication unit 138 can include hardware and/or software for use in sending and receiving messages on channel 118, where the messages may be routed to or from any network resource.
Gateway 106 may include a suitably programmed gateway processor 150 configured to execute computer instructions, a gateway memory 152 configured to store and retrieve information, a network address translator (NAT) 154 configured to allow the use of one set of external network address and one set of local network addresses which are not necessarily globally unique, a dynamic host configuration protocol (DHCP) server 156 configured to dynamically assign local network address, a gateway communication unit 158 configured to send and receive messages on LAN 108 over channels 118 and 120, and/or send and receive messages on WAN 112 over channel 122. NAT 154 provides a translation or mapping of address on LAN 108 to addresses on WAN 112 enabling local traffic on LAN 108 to use one set of network addresses and external traffic on WAN 112 to use another set of network addresses, where the network addresses on LAN 108 are not necessarily globally unique. Finally, DHCP server 156 dynamically assigns network addresses to network devices connected to LAN 108. These addresses may be assigned upon request and expire automatically if address lease is not renewed after a predetermined period of time has expired. Gateway memory 152 may include a Random Access Memory (RAM), a disc memory, an optical memory, a magnetic memory device, a register file, and/or any technology for storing and retrieving information for use by gateway processor 150 or any gateway 106 resource, including NAT 154, DHCP server 156, and/or gateway communication unit 158. Further, gateway memory 152 may be removable from device 102 to provide for safekeeping of the stored information, ease of maintenance, and/or re-configurability.
Client communication unit 138 may be dynamically or statically assigned a network client transport address 160 comprising a network address and/or a port address, for example. The network address can be an Internet Protocol (IP) address comprising the familiar series of address field octets, or may be a name resolved into octets. Gateway memory 152 may store and retrieve the client device transport address 160 and port comprising the network address of the client 102 on LAN 108. To avoid collisions, each network address is unique for devices on a particular LAN, such as LAN 108. A port is an endpoint to a logical connection and is usually specified by a number to designate the kind of server or protocol to which the network message traffic applies. Some port numbers are well known by convention where, for example, port 80 designates Hypertext Transfer Protocol (HTTP) traffic, and port 443 designates HTTP Secure (HTTPS) traffic as described in the Internet Engineering Task Force (IETF) Request for Comments (RFC) document 1700, also referred to as IETF-RFC1700. The LAN 108 client device transport address 160 for device 102 is locally assigned by a DHCP server 156 in gateway 106 and is not typically visible on second network 110. The NAT 154 assigns a mapping between a local address:port combination to an external address:port combination for a period of time which may be updated if message traffic flows through the mapped address:port combination pair. System 100 may include an extensive hierarchy of networks where client device 102 and proxy server 116 may be separated by a plurality of NATs for translating addresses at the interconnection point between adjacent networks.
Admin device 114 may include a suitably programmed admin processor 170 configured to execute computer instructions, an admin memory 172 configured to store and retrieve information including network addresses and client device identifier 134-1, an admin user interface 174, such as a web browser, configured to receive data from and present data to an admin user, and/or an admin communication unit 176 configured to send and receive messages on channel 124 onto WAN 112. Admin memory 172 may include a Random Access Memory (RAM), a disc memory, an optical memory, a magnetic memory device, a register file, and/or any technology for storing and retrieving information for use by client processor 170 or any admin device 114 resource. Further, admin memory 172 may be removable from admin device 114 to provide for safekeeping of the stored information, ease of maintenance, and/or re-configurability.
Proxy server 116 may include a suitably programmed proxy processor 180 configured to execute computer instructions and a proxy memory 182 configured to store and retrieve information including a client device identifier 134-2 and a client device network address 160-1. Proxy server 116 may also include a proxy communication unit 184 configured to send and receive messages on channel 126 onto WAN 112. According to one or more embodiments of the present invention, proxy server 154 may be configured to (a) maintain network connectivity with devices using a light-weight protocol, (b) coordinate the establishment of TCP connections, and (c) relay traffic between an administrator and client device. It is not necessary that the router function in gateway 106 cooperate with proxy server 116 to accomplish these capabilities since they operate independently. Proxy memory 182 may include a Random Access Memory (RAM), a disc memory, an optical memory, a magnetic memory device, a register file, and/or any technology for storing and retrieving information for use by proxy processor 180 or any proxy server 116 resource. Further, proxy memory 182 may be removable from proxy server 116 to provide for safekeeping of the stored information, ease of maintenance, and/or re-configurability.
As will be more fully described below, one or more embodiments of the present invention provide remote access by an administrator on an external network to the administrative user interface (UI) of a client device on a private network through a smart HTTP proxy and private device functionality that work together in unison to provide access to the client device. A private network may be a local area network or network hierarchy located behind a firewall or a NAT router such as a home gateway or a restricted business network.
Flow 400 continues with operation 404 where gateway 106 optionally responds to device 102 with a UDP authentication challenge. This message may also contain the source transport address as seen by the proxy server. This allows the Client Device to learn its current public transport address assigned by the gateway. Flow 400 continues with operation 406 where device 102 responds to the optional authentication challenge by asserting a UDP Keep-Alive message with authentication, including device identifier 134. Once proxy server 116 receives the authentication response, flow 400 continues with operation 408 where proxy server 116 validates the authentication and updates proxy memory 182 with the associated transport address 160-1 and device identifier 134-2. Thus, client device 102 establishes a network path to proxy server 116 that can optionally be secured (e.g. authenticated and encrypted) using a variety of mechanisms. Once proxy server memory 182 is updated, flow 400 continues with device 102 periodically sending UDP Keep-Alive messages (410, 412). The appropriate period between sending UDP Keep-Alive messages depends on the behavior of the NAT function in gateway 106. The appropriate period may be configured on client device 102, for example by an administrator, or client device 102 may dynamically discover the appropriate period by observing NAT binding expiration time of the NAT 154 in gateway 106. Flow 400 concludes when the NAT communication path is no longer needed and is disabled.
Flow 500 continues with proxy server 116 uses client device transport address 160-1 to instruct client device 102 to initiate a connection with proxy server 116 in operation 506.
Client device 102 initiates the connection with proxy server 116 in operation 508. This connection between client device 102 and proxy server 116 may conform to the Transport Control Protocol (TCP) and be capable of transporting messages comprising HyperText Transport Protocol (HTTP) traffic. Once the TCP communication paths are established with proxy server 116, flow 500 continues with proxy server 116 relaying the request from admin device 114 to client device 102 in operation 510. Flow 500 concludes with proxy server 116 relaying a response from client device 102 to admin device 114.
Once proxy server 154 receives the request to open the TCP connection in operation 602, flow 600 optionally continues with proxy server 116 and admin device 114 mutually establishing the authenticity of each other and/or establishing encryption to guarantee privacy or data security during the requested TCP session in operation 604. A Transport Layer Security (TLS) protocol, for example, may be used to mutually authenticate the identity of admin 114 and proxy server 116 and/or encrypt the data portion of one or more message packets to prevent an unauthorized network device or user from accessing the message data. In this manner, operation 604 can allow authentication between proxy server 116 and client device 102 as well as allow the negotiation of an encryption algorithm and cryptographic keys before exchanging sensitive data.
Once the optional authentication and/or encryption is completed, flow 600 continues with admin 114 asserting a first HTTP request to proxy server 116 which may including a Uniform Resource Locator (URL) along with device identifier 134-1 for device 102 in operation 606. When proxy server 116 receives the first HTTP request, flow 600 continues with proxy server 116 searching proxy server memory 182 for device transport address 160-1 corresponding to the device identifier 134-2 of device 102 in operation 608. In one example of searching, the client device identifier 134-1 supplied by admin device 114 is compared with client device identifier 134-2 stored in proxy memory 182 to determine a match. If they match, then the associated client device transport address 160-1 allocated to client device 102 is retrieved from proxy memory 182 and used to access client device 102. A different device transport address is assigned each network device on LAN 108. Once proxy server 116 has located the record in proxy server memory 182 corresponding to device 102, proxy server 116 sends a UDP request to device 102 at the stored transport address with a request to establish a TCP session with proxy server 116 in operation 610. Once device 102 receives the UDP request for TCP session, device 102 optionally responds with a UDP authentication challenge in operation 612. Upon receiving the challenge, proxy server 116 responds to device 102 with an authenticated request in operation 614. Device 102 examines the authenticated request and responds by creating a client TCP connection or session with proxy server 116 in operation 616.
Once the private TCP connection is created in operation 616, flow 600 optionally continues with proxy server 116 and client device 102 mutually establishing the authenticity of each other and/or establishing encryption to guarantee privacy or data security during the requested TCP session in operation 618. Once both the admin TCP connection and the client TCP connections are established, proxy server 116 relays a version of the first HTTP request to device 102 in operation 620. In this case, the relayed first HTTP request of operation 620 does not need to include device identifier 134. Device 102 receives the first HTTP request and provides a first HTTP response over TCP in operation 622 conveying the response from device 102 to the first HTTP request from proxy server 116 in operation 606. Proxy server 116 may augment the response as necessary before relaying. For example, proxy server 116 may insert an HTTP instruction to Admin Device 114 web browser 174 to keep the TCP connection open or it may insert an HTTP cookie to identify the session. Proxy server 154 receives the first HTTP response over TCP from device 102 and relays that response in an operation 624 to admin device 114 as a response to the first HTTP request in operation 606. In this manner, a relay communication path through gateway 106 and proxy server 116 is established between client device 102 having a private address and admin device 114 on a separate network, where admin device 114 is not aware of the private address for client device 102. From this point on, proxy server 116 maintains two, separate TCP connections: one to client device 102 and one to admin device 114. Subsequent requests from the client may be relayed to the device and responses may be relayed back. Both TCP connections may be terminated for various reasons including an explicit indication from admin device 114 that the session is over, or after a period of inactivity.
Once all messages are passed between client device 102 and admin device 114, flow 700 concludes in operation 704 that closes the relay link through proxy server 116 between client device 102 and admin device 114. Specifically, operation 704 includes an operation 720 that closes the client TCP connection between client device 102 and proxy server 116 and an operation 722 that closes the admin TCP connection between proxy server 116 and admin device 114, both established in operation 600 as described in reference to
Although the invention has been described with respect to particular embodiments, this description is only an example of the invention's application and should not be taken as a limitation. Consequently, the scope of the invention is set forth in the following claims.