EA025082B1 - System for providing access to independently operating servers using the same network address - Google Patents

Abstract

A system for control of virtual servers includes several virtual servers functioning on a base computer. An address and a unique identifier may be associated with each virtual server. A control procedure provided inside of each of respective virtual servers can be used for controlling the respective server in accordance with control commands. A control interface located outside the virtual servers can be used for controlling the virtual servers in accordance with control commands. A control command can be fulfilled both by the control procedure and by the control interface, or it can be fulfilled by the control interface if the procedure is inaccessible, or can be fulfilled by the control interface if the virtual server is damaged, or can be fulfilled even in case the virtual server is inaccessible. The control interface can be accessed using several addresses and identifiers.

Description

The present invention relates to virtual servers and, in particular, to control systems for isolated virtual servers, which are presented to the user in the form of an unresponsive or failed virtual server.

Description of the prior art

Today, with the growing popularity and success of the Internet, server technologies have gained great commercial importance. Separate server applications typically run on a single physical host computer that serves client requests. However, providing one basic physical computer for each server application is an expensive and inefficient solution.

For example, often, when providing network services at a commercial Internet service provider (ISP) (data center, IT department, hosting company, etc.), as a rule, it provides a separate basic physical computer for each client that runs server application.

However, a client who has paid for hosting services (services for providing computing power) often not only does not require, but also does not intend to use the entire base computer. In general, individual customers require only part of the computing power, part of the data storage facilities and other computer resources.

Thus, it is desirable to use the hosting of several server applications on the same physical computer. To ensure commercial acceptability, each serving (or server) application must be isolated from all other server applications on the same physical base computer (or host). In this context, a single computing system (usually referred to as the host or host) supports several virtual servers (for example, virtual private servers or VPNs), and sometimes more than hundreds and thousands of virtual private servers.

Many applications that use virtualization concepts such as virtual private servers (VPNs) relate to the use of virtual private servers in the context of web servers. For this, such VPNs are provided with assignable ΙΡ addresses (Internet protocol addresses), as well as a number of other parameters associated with a network server or web server, for example, IP (universal resource search tool) with the name ΌΝ8 (domain name server), etc. d.

A typical situation in which a VPN can be used is a single box (in other words, one computer) located in a data center. The box has an operating system and many virtual private servers (VPN1, VPN2, VPN3, etc.), each of which is presented to the user as a single dedicated web server. Therefore, users have no idea that the VPS is something other than the box allocated to this user. As a rule, each VPN has a control panel, or some other similar mechanism by which the user can control the behavior of their VPN. In other words, as the administrator of this VPN, the user has significant control over the behavior of this VPN, as long as the underlying operating system allows the actions in question.

In addition, the operating system has its own control panel, which is available for the system administrator to manage various functions of the stay, as well as to regulate the activities of various tasks. As a rule, the operating system control panel is accessible to the administrator of the data center, where the VPN is physically located. In fact, a typical data center can have hundreds or even thousands of such boxes.

Each VPN will usually contain many processes associated with the VPN, as well as an address, such as a host name or ΙΡ address, with which the user can communicate with processes associated with a specific VPN. A VPN can have control procedures in the form of a common set of procedures. Management procedures are typically required for things such as monitoring (or defining parameters and operating modes) of the VPN, configuration and management of the VPN. Each management procedure can be accessed using the address corresponding to this procedure.

Since the control procedures typically performed in the VPN are not available if the VPN is unsuitable for use, the management procedure (procedures) performed outside the VPN can be used to control the operation and performance of the VPN.

The problem with the conventional approach is that the user must use different addresses to establish a connection with the control procedures, while the addresses, in turn, depend on the state of the VPN. Such control procedures cannot be used to determine the activity of the HFV until a connection is established. In addition, the user must remember the exact address and configuration parameters in order to use such control procedures, which often differ from those required for administering the VPN itself.

In addition, certain management or monitoring actions can be performed using an internal or external control interface, graphical user interfaces

- 1 025082 (GUI) may differ, and users who manage the VPN will often need special training, for example, to select the correct address not only corresponding to the status of the VPN, but also the required action.

As a rule, VPNs communicate with the outside world using the input / output ports of a computer system (computer), which are connected to any network, such as the Internet, local area network (LAC), etc. Each such VPN is associated with a ΙΡ address, which, as noted above, is used by the outside world to access this VPN. But for some reason, for example, due to an accident or taking the VPN to a break, etc., a problem may arise. In the event of problems with physical equipment, software for solving this problem is usually not used, except for performing a reboot of the VPN on another machine. However, many accidents occur due to manifestations of the internal activity of the VPS itself, for example, the impact of a computer virus on the VPS, or the user launching a code that leads to a VMS accident, etc. In this regard, the considered VPN is becoming inaccessible to the user who is trying to access the VPN, and does not receive feedback from the VPN. Thus, the user does not have a mechanism to provide responses from the emergency VPN, and, as a rule, the user should contact the data center personnel, who can manually restart the VPN. This, often, becomes another problem, as it requires time, in addition, the data center personnel are not always available, if necessary, and the data center may not be adequately staffed, in particular, an accident at the VPN may occur during the period when the data center has few staff, etc. Thus, the usual approach is not acceptable for a number of critical tasks, or commercially important applications, since an emergency and non-responsive VPN can lead to a drop in sales, decrease in productivity, etc.

Accordingly, there is a need for a system and method for remotely controlling and verifying the parameters of a failed VPN, without the manual intervention of a data center administrator, and in an acceptable way of managing the failed virtual servers.

SUMMARY OF THE INVENTION

Thus, the present invention relates to a system and method for providing services for virtual servers using the same network address, which substantially eliminates one or more of the disadvantages associated with the prior art.

A system for managing virtual servers comprises a plurality of virtual servers operating on one main computer system and sharing one main operating system, each server may contain an associated address and a unique identifier; each virtual server can be executed as responding to customer requests, contain an internal control procedure for managing the specified virtual server, and the system can contain an external control interface operating on the main computer system and operable outside the virtual server, configured to control the virtual server, in accordance with commands sent by a remote user using an address and a unique identifier, moreover, an external control interface can be configured to administer the specified virtual server using the address and unique identifier of the specified virtual server if the specified virtual server does not respond to requests. The system can be executed with the ability to execute commands, both by the control procedure and the external control interface, with the possibility of executing commands by the control interface, if the control procedure is not available. Moreover, the management interface can be made accessible using multiple addresses and with the ability to manage multiple virtual servers. It can be configured to execute at least two of the following commands:

start / stop the virtual server;

Change virtual server passwords

reinstall virtual server;

repair a virtual server;

make a backup copy of the virtual server;

change the characteristics of the current level of resource use;

process the virtual server management command;

determine the status of the virtual server;

set up a virtual server;

Install a custom application

uninstall the user application and configure the user application.

In addition, the management interface can be executed activated in case of failure of any of the virtual servers, the control procedure can be performed activated in response to commands sent by the remote user using the address and a unique identifier, at least one server can be executed as a virtual one cars or at

- 025082, at least one server can be made in the form of a virtual software execution environment, and the inoperability of the virtual server, if the virtual server does not respond to requests, is hidden from the user by using the management interface, and the address can be used to generate, using the specified address, information transfer channel between the remote user of the specified server and either the management interface or the control procedure.

In an exemplary embodiment of the invention, a system, method and computer program product for managing virtual private servers, including many virtual private servers, is provided. Each virtual private server has its own address and unique identifier. Each virtual private server has internal management procedures available on the virtual private server to manage the specified virtual private server in response to commands sent by the remote user using an address and a unique identifier. Outside the virtual private servers, the external management interface is in operable state for managing virtual private servers in response to commands sent by the remote user using an address and a unique identifier. If some virtual private server does not respond to requests, the external management interface can be used to manage this virtual private server using the addresses and identifiers associated with this virtual private server. A system for managing virtual servers can contain many virtual servers operating on one main computer system and sharing one instance of the main operating system, each virtual server having an address and a unique identifier associated with it. In this case, each of the virtual servers responds to customer requests, and each virtual server may contain an internal management procedure for managing the specified virtual server. In addition, the system may include an external control interface running on the main computer system, operating outside the virtual server to control the specified virtual server in response to commands sent by the remote user using the specified address and the specified unique identifier. If any virtual server does not respond to requests, the external management interface can be used to administer the specified virtual server using the address and unique identifier associated with the specified virtual server. At the same time, the commands can be executed both by the control procedure and the external control interface, and the commands can be executed by the control interface if there is no access to the control procedure or in the event of a virtual server failure. In addition, the management interface can be accessed using multiple addresses. Teams can be used to manage user applications running inside at least one of the virtual servers. In the system, the address can be the host name, and virtual servers can be supported by one core of the operating system. The management interface can be activated when virtual servers fail. The control procedure may be activated in response to a command sent by the remote user using the specified address or the specified unique identifier.

A virtual server administration method can be implemented in the system, including launching many virtual servers operating on the same main computer system and sharing one instance of the main operating system, each of the virtual servers responding to client requests, assigning an address and a unique identifier to each virtual server, starting , outside of many virtual servers, an external control interface that runs on the main computer system and is accessible user, launching, in each virtual server, an internal management program to manage the specified virtual server, setting a policy for executing management commands that determines the choice from (a) and (b), where (a) an internal management program is used to administer the virtual server, and (b) an external management interface is used to administer a virtual server, receiving a command from a remote user; and in accordance with the policy for executing management commands, administering a virtual server by executing either (a) an internal management program of a specified virtual server using an address and a unique identifier associated with the specified virtual server or (b) an external management interface using an address and a unique identifier associated with the specified virtual server, and the choice between (a) and (b) is hidden from the remote user. In this case, the command can be executed if the virtual server is unavailable. The control interface can be accessed using several addresses, moreover, if the control interface and the control program use the same identifiers, the external control interface is available when the internal control program is not available. In addition, the execution of the control command may depend on the state of the virtual servers, the control interface may be activated in case of failure of any of the virtual servers, and the control program may be activated upon receipt of a request from the user.

Additional features and advantages of the invention will be set forth in the description which follows, and, in part, will be readily apparent in the description, and in part may be revealed.

- 3 025082 when using the invention in practice. The advantages of the invention will be realized and achieved in the structure, which is reflected in the description, claims, as well as in the attached graphic materials.

It should be understood that both the following general description and a detailed description are made for illustrative and explanatory purposes and are intended to clarify the claimed invention.

A brief description of the graphic materials

The accompanying graphic materials, which are used to better understand the invention, are part of the present description illustrating embodiments of the present invention, together with a text description that serves to explain the principles of the present invention.

In the drawings:

FIG. 1-3 show examples of the implementation of the systems of VPN

FIG. 4 illustrates an example of access of various client controls to the contents of several VPNs;

in FIG. 5 is a schematic diagram of an example computer or server that may be used in the invention.

Detailed Description of Preferred Embodiments

The following are details of embodiments of the present invention, illustrated by the accompanying graphic materials.

The present invention relates to a system, method and computer program product for managing virtual private servers, such as virtual runtimes or virtual machines operating as servers. Since various types of virtual private servers behave similarly, from the point of view of the objectives of the present invention, the following description will be primarily disclosed in terms of virtual private servers.

A virtual private server (VPN) is a closed collection or set of processes, system resources, users, user groups, objects and data structures. Each VPN has its own identifier, which distinguishes it from other VPNs. VPN offers its users services that are essentially functionally equivalent to a stand-alone server with remote access. From the point of view of the administrator of the VPN, the VPN operates predominantly as a separate data center computer. For example, it is desirable that the administrator of the VPN will have the same remote access to the server via the Internet, the same as for a regular server, the ability to restart the server, load the system and application software, the ability to allow connection of the VPN users, the ability to set disk space quotas for users , and user groups, provide support for storage networks (8ΆΝ), establish and configure network connections and web servers, etc. In other words, it is desirable to have a full set of system administrator functions, as if the VPNs were dedicated remote servers, and the existence of the VPN should be hidden from the user of the VPN and the administrator of the VPN.

The approach described in this document has the advantage that the operating system has a number of mechanisms for determining the parameters of the incoming and outgoing traffic of virtual private servers. Thus, a user of the VPN can be provided with tools located in the main operating system (in addition to the tools that are in the VPN) to manage the VPN. In addition, the approach disclosed in the present description has the advantage that usually the main (main) operating system is notified of a failure or failure of the VPN, or does not respond to access requests at the address assigned to the VPN. A user of a failed VPN, using the tools provided by the operating system, can restart the VPN itself, without interference from the data center administrator.

The main (basic) operating system monitors network traffic for virtual private servers, but usually does not take any action, since it is assumed that the vast majority of virtual private servers are functioning normally. However, if one VPS does not show signs of life or is in emergency mode, the network traffic of the corresponding VMS can be intercepted, recorded, analyzed, etc. It should also be noted that, although the description above mentions one unique address (for example, an ΙΡ address) assigned to one VPN, the proposed approach also applies to the situation when the address is common to several VPNs.

Another way to develop the proposed approach is to consider it from the point of view of providing the VPN user with an additional control panel in the base operating system, which acts as a layer for intercepting incoming traffic. In the event of a VPN incident, the interceptor layer may still respond in some way to access requests directed to address Ч of the emergency VPN, ensuring that there is something interacting with the user at the VPN address. The interception layer can be configured for each VPN, either by the administrator of the data center or by the user of the VPN, depending on the minimum required functionality and rights that the administrator of the data center is ready to grant the user in these circumstances. Without- 4 025082 conditionally, the basic right is the ability to restart the VPN (in its essence, possible only from outside the VPN). In addition, stopping the VPN can be done using an intercept layer, where the user knows that he can restart the VPN without the intervention of the data center administrator.

For example, a VPN control panel can give a user access to such a VPN or to address-specific information such as an access and statistics log, resource usage statistics, management of various applications running on the VPN, passwords, etc. The interception layer can control the traffic at the protocol level, for example, determine the ICI for which the traffic is intended, etc. The interception layer, in the process of work, can redirect the request for access to the damaged ΙΡ address to its own server, the functionality of which is guaranteed in some way.

In addition, the properties of the control panel of the main operating system (or the interception layer), similar to those described above, can be constantly implemented for all VPNs, or can be provided in any form (for example, as additional services provided for a certain fee by a hosting provider in including in the form of different service levels for different sets of commands). In the case of arbitrary (optional) provision of the control panel in the main operating system, it can be activated, as shown above, if there is reason to believe that the VPN is more than usual prone to accidents or malfunctions, or, for example, when the specified VPN is tested with new program code.

It should be noted that an important aspect is the definition of specific VPNs that should be controlled using management tools, such as a control procedure (s) or control interface. The control procedure (s) must be set up in such a way that it can be operational inside the VPN, for which it is intended to be controlled. The control interface must be installed in such a way as to ensure the operability of the interface outside the VPN, for which it is intended to be controlled. For this, the control interface can be installed in the main operating system, or in another VPN, which can be directly supported by the hardware of the system. Without this, the operating system is usually not able to determine whether or not a specific action is valid for a given VPN. You should pay attention to the fact that the control procedure and / or the control interface can be constantly executed, or can be started or stopped on demand, for example, when establishing a transmission channel. For example, control procedures and the control interface can be loaded on demand (for example, start and stop) to execute control commands.

After the user closes the control session, the corresponding processes or computational tasks may become inactive. In other words, the user can exit the management procedures and the management interface, or even kill these processes in the task manager.

In one implementation of a virtual server, the operating system creates a data structure that stores information corresponding to the VPN, which usually occurs when the VPN itself is created. Some information may be stored permanently. For example, the file system may contain a configuration file for each VPN. Data structures may also contain a number of parameters used to ensure the health of the VPN itself. In the preferred case, there is the possibility of quick naming of virtual private servers, so that a quick search of data structures is provided. The name may be the identifier (or ΙΌ) of the virtual private server, which may be, for example, a number, an alphanumeric string, or a similar structure. Those objects that can be unambiguously identified as belonging to a specific VPN, as a rule, cannot be re-connected with another VPN, since this possibility usually involves the simultaneous appearance of security vulnerabilities. In other words, a process that was born in the context of one VPN can not live in another VPN.

Running several VPNs on the same main computing system, thus, allows for a high level of use of the hardware or server hardware. Due to the effective isolation of VPN resources, it is possible to offer users a guaranteed level of security and availability of resources specified in the service level agreement (8L).

Thus, a typical commercial application includes a physical computing system or a stand-alone computer (which is usually called the host or host), which is configured to run several isolated virtual private servers (VPNs). The host contains the kernel of the operating system (OS). The kernel performs a number of processes and threads to perform various system processes associated with the system. The host also contains a number of processes running on server hardware that correspond to the virtual private servers of at least one user connected to at least one VPN.

The host may contain physical memory, such as dynamic random access memory (RAM), used by various processes and computing threads. A hardware complex may contain various storage elements, such as a high-speed buffer memory or cache, as well as buffer elements of the operating system. A computer system also contains a central processing unit (SRI), or, in the particular case, several processors, whose time can be shared between virtual private servers. Peripheral devices may include, for example, disk storage devices, input / output interfaces, information network interfaces, etc. Storage devices, in particular, disk drives, may contain data files, databases, metadata storages, and other permanently and temporarily stored digital objects.

Usually, when used in the present description, it is understood that the kernel of the operating system has one or more processes executed in kernel mode. The operating system itself can be seen as a process. The OS process has its own objects intended for its own internal use, as well as many objects that represent the virtual private servers supported in it.

Each VPS is, therefore, a set of processes running in user mode, with support from the core of the operating system. Each such VPN, as a rule, has a set of users and computing threads that correspond, for example, to application software launched by a particular user.

Each VPN can have its own identifier, which allows someone, for example, an administrator, to find VPNs through a network interface. Examples of such identifiers may be names or addresses or similar labels. VPS also contains a set of interfaces that allow users, including the VPS administrator, to access various services or services available on the server and in the kernel. A special case of such services are remote control services, which are sometimes called server configuration tools. Such interfaces contain system calls, shared memory interfaces, control of input / output drivers (IUST), and other similar mechanisms.

Each VPN as a rule contains many processes, threads, lightweight processes and other primitives for serving users of this VPN. Each VPS also typically has its own objects and data structures that are associated with the processes and flows of this particular VPS. Each VPN can also contain several objects and data structures used by the operating system to manage this VPN.

It should be noted that in the general case, the processes of different virtual private servers are completely isolated from each other and cannot affect the processes of other VPNs, although in some cases one VPN can support other VPNs within itself. In addition, high-level VPNs can provide control and other service functions. In addition, the isolation level between virtual private servers can be reduced either by the main OS or by using additional settings.

Virtual private servers can also be implemented in the form of software expansion of capabilities based on servers \ UshDo \ U5 based (for example, in the form of add-ons), for example, servers Μίετοκοίΐ ^ ίηάθ№5 ΝΤ, servers Myutokoy \ UshDo \ U5 2000, M1 quick \ UtDo \ U5 8sgusg 2003, as well as their various derivatives, by adding the missing isolation mechanisms and VPN resource management mechanisms to the kernel of the operating systems of the above operating systems.

Each VPN offers a set of services for users of the specified VPN, which are implemented as processes within the specified VPN. From the user's point of view, the presence of the VPN is not obvious, so that for the user, the VPN will look like a completely remote server. The present invention makes it possible to maintain the activity of some functions of the VPN, even if the VPN itself is out of order or seems inaccessible to the user.

One example of a system in which the present invention may be implemented will be further discussed with reference to FIG. 1, 2 and 3.

FIG. 1 illustrates a generalized embodiment of a system with isolation of the RF components, and FIG. 2-3, illustrate system examples in the context of a network environment where virtual private servers can be servers that share some resources among themselves, or virtual servers with resources specially allocated for them, for example, those that run with the support of the virtual machine monitor (VMM).

As shown in FIG. 1, the host computer system (host) 102 runs the host operating system with core 108. The VPS 104 (only one VPS is shown in the drawing) usually has a unique address, for example, its own 1P address (although, in some cases, the same 1P the address may be shared by several VPNs). The interface 114 of the operating system (for example, typically including system calls, input / output control (usOSGsShK) drivers, etc.), is used to communicate between processes 120 of user space (user mode) and processes 108 of kernel space (kernel mode ) In this context, the kernel 108 of the operating system is a protected part of the operating system, usually such a part that is important for the functioning of the host 102. Note also that, although only one host 102 is shown, the invention can also be implemented with virtual environments. working on a computer cluster.

- 6 025082

Numbers 122A and 122B show applications and automatic services executed in user space 120 of the main operating system. A server can include a number of such programs, applications, system services, automatic services, and other similar structures commonly used by users (for example, word processors, database management systems, network page servers, remote procedure calls, support for automatic procedures, Te1ie1 servers etc.).

Position 124A shows a collection of objects associated with a particular VPN. Position 124Ό denotes a set of objects and data structures associated with the operation of system software processes. Position 124C denotes a set of objects and structures that are used by the kernel 108 of the operating system for its own needs. Position 126 denotes operating system drivers, such as hardware drivers. Position 128 denotes buffers, cache, and other similar structures of the kernel of the OS, which are used for storage and input / output operations. Positions 130A, 130B show the processes of system software and automatic procedures operating in user mode. Positions 132A, 132B are applications and services that exist in user space 120. Objects and data structures 124A, in turn, can be created in user mode 120 by user mode processes 132A, 132B. Other examples include automatic procedures (daemons), for example, automatic procedures of a web server or network server that interact with an Internet browser, servers with data conversion of the input port of a remote procedure call (РРС Roy Tarreg, for example, servers that do not interact directly with by user) etc.

Positions 134A, 134B characterize the relationship between user space processes 120 and kernel space processes 108. In one embodiment of a system for using the present invention, isolation between virtual private servers 104 is implemented at the core level, in addition, isolation between some of the VPNs 104 can be implemented in user space 120.

In addition, each VPN can have its own set of resources, for example, its own disk space, file space, its share of the total bandwidth of the network adapter, etc. It should be understood that, in fact, any number of VPNs 104 can be run on host 102, and restrictions are imposed only by physical system parameters, such as processor speed, system bus bandwidth, and other resources.

Each VPN 104 has a share of system resources allocated to it. The resources allocated to VPNs 104 (for example, memory, file space, I / O resources, etc.) are isolated from each other using a virtual private server GO and / or using a special kind of addressing.

In general, in order to implement the present invention, the address space is divided into two parts: kernel space 108, which is used in the kernel mode in the operating system, and is generally not available for application software launched by users 112, and a set of virtual address spaces, allocated to user processes, which is commonly referred to as user space 120, in which virtual private servers 104 and user applications 132. is its own set of addresses used by user processes to access different data both locally and over the network. Such an implementation prevents security breaches of the VPN 104. It should be noted that some functions of the VPN can exist in the core space 108. However, some functions may exist in the user space 120. In user space 120, software 130 may support the necessary core functions of the operating system. An example of such a function may be an automatic procedure (daemon), which collects statistics about the use of VPNs 104. Another example may be an automatic procedure that monitors the processes of VPNs 104, etc.

As noted above, each VPS 104 contains many associated objects. Such objects represent users 112, corresponding user processes and / or applications running within this VPN.

As shown in FIG. 2, the base computing system 102 may support or execute multiple VPS 104A, 104B, and 104C (note that 104C is not shown in FIG. 1B), each with its own 1P address and / or other identifier (e.g., IRB or host name) that allow the user to find the VPN or establish a connection with the procedures of the VPN, for example, via the Internet or using the network software interface 106A, 106, b, 106C. It should also be noted that dashed line 160 indicates that user applications, as well as system utilities, can exist in user space 120, in addition to system utilities that can exist in kernel space 108.

In addition, user 112 can connect to the host network interface, for example, operating in user mode (see 106C).

Some VPNs, such as 104C, with the support of host 102, may be inoperative due to software errors, or may be completely inactive and exist only as records on

- 7 025082 data storage device. Failures can occur by accident, due to an error in a particular application, or intentionally when a user conducts a denial of service attack (ΌΟ8). In addition, application crashes, web server crashes, and other crashes and corruptions can occur. A poorly written script can easily capture a significant part of the processor resources (sometimes, up to 100% of SRI resources), as a result of which, other users do not get the computing power. An example of a poorly written script CC1 (common gateway interface) is an infinite loop, namely:

line 1, go to line 1.

In an exemplary embodiment, the user 112 can connect to the web interface 106 of the host using the identifiers of one or more VPNs, where in order to create the proper connection, a unique address is assigned to the virtual private server. This unique address and identifier is used to create a transmission channel in the network, for example, in a packet switching network, such as TCP / 1P network. This transmission channel is set to deliver control commands to an available control procedure (s).

To illustrate, in FIG. 1C, a single control interface 107C located outside the virtual private servers 104 is shown. This interface is external to the virtual private servers 104 (although the invention is not limited to one such control interface, but may use several interfaces). It should be noted that the control interface 107C is located both in the user space and in the space of the operating system (kernel). The control interface 107C has the ability to check and manage at least some operating modes and settings of the VPN, as if the specified VPN were running on one hardware server corresponding to this VPN. It should be noted that the operating system may contain more than one control interface and operating controls in it.

Each VPS 104 may also contain its own internal control procedure (s) operating inside VPS 104 for monitoring and control functions, which, in itself, can be executed as user object 109 or kernel object 124.

Virtual private servers 106A and 106B are connected to users (clients) 112A and 112B, moreover, in this case, one client 112 is connected to each VPN, 104. In this case, client 112 is an illustrative example of a user. Another example, user 112 may be a VPN manager. The user (client) 112C is connected to the host network or network interface 106C.

Clients 112A and 112C can manage the virtual private server objects 104A and 104B, respectively, from within the VPNs 104A and 104B, and the user 112C can manage any VPNs based on the established connection or access rights granted to the client 112C.

FIG. 4 illustrates an example of access to various client-side control mechanisms in the context of multiple VPNs 104 running on the same server 102. FIG. 4, reference numeral 201 denotes a communication channel, for example, Internet link 202. Several VPS 104 are executed on the host 102. Each VPS 104 has its own application software 106, its own system software 210A, its own system libraries 212, and its own internal control procedure (s) 214. Each VPS 104 typically has its own 1P address (although 1P addresses can be used simultaneously by several VPS). Clients 112 (which may be VPN administrators, authorized users, or unauthorized users, such as anonymous Internet users) use communication channel 201 to access virtual private servers 104. The external control interface 107 (shown here without an alphabetic designation that reflects it the ability to work with several VPNs simultaneously), which is located outside the virtual private servers 104, in the core space 108, can also be used to organize and manage VPNs 104. External The control interface 107 can interact with various elements of the RFS 104 using the interface 220. It should be noted that the external control interface 107 may be freely accessible to one user 112 (e.g., 112A and 112b), but not to other users (e.g., 112C). We also note that the various rights granted to different users 112 may differ, respectively, the interfaces 220a and 220B are connected to different elements of their respective RFs.

The external control interface 107 is designed to solve various problems associated with the operation of the VHS 104. Such an implementation is more flexible than the internal control procedure (s) 214 due to the ability to work with the malfunctioning VPS 104. It should be noted that the external control interface 107 may be a set of procedures and programs, where, for example, each such procedure or program is responsible for one control parameter. In addition, the external control interface 107 may be unique to each of the VPNs 104, or may be common to some VPNs 104, and not used for others, or may be common to all VPNs 104.

The functionality of the control interface 107, as well as the internal control procedure (s) 214, provides the user 112, such as the administrator of the VPN, the ability to administer the user VPN 10 104 using any standard network browser on any platform. It should also be noted that although most often this situation occurs when the control interface 107 is used when specific VPNs are not responding, while the internal control procedure (s) 214 is used when the VPN is responding. It is also possible to use the control interface 107 to control the VPN, which responds to requests (instead of its own internal control procedure (s) 214 of the specified VPN.

User 112 has the ability to control the VPN user 104, usually through some form of graphical user interface (GUI), which is part of the control interface 107, in any of the following ways:

(a) starting, stopping or restarting the VPN;

(b) mounting and unmounting the VPN in recovery mode;

(c) reinstallation of the VPN;

(4) backup and recovery of the VPN;

(f) changing the root password (administrator password) of the VPN;

(ί) starting, stopping, or restarting certain services within the VPN;

(e) view the list of processes of the HFS; transmitting them signals;

(B) connection to the VPN through 88Н (the secure shell protocol is packet-based, a binary protocol that supports an encrypted connection to remote hosts or servers);

(ί) view current resource consumption and warnings about excessive use of resources;

(]) viewing logs (protocols);

(k) setting up a virtual server;

(l) installing a user application;

(t) removal of the user application;

(o) customizing a user application;

(o) GUI setup.

There are at least two possible ways to control access to VPNs 104.

1. Using the ΙΡ address (or host name) of the virtual private server 104 and the TCP (data transfer control protocol) of the port of the external control interface 107.

2. Using the ΙΡ address (or host name) and TCP port of the private VPN 104.

These two methods can be almost identical with respect to the services available for the control procedure (s) 214 in the VPN, 104. For example, if the VPN is not available due to an internal error or inoperability, the management procedure (procedures) 214 is also not available.

In any case, the user 112 must connect to the internal control procedure (s) 214 using the VCh GO, as well as, in the particular case, the user GO and password necessary to obtain access rights to the control procedure. For example, user 112 can use the address (or host name) and the TCP port of his VPN (104) by entering the specified address in the address bar of the user's browser (browser), and also use his identification data (username and password) to enter the corresponding procedure ( procedures) 214 management. After that, the user, in the General case, can administer and manage the state of the VPS 112, for example, using the graphical user interface or simply sending a command line to control the procedure.

The external control interface or GUI of the internal control procedure (s) may be included in the interface 107 of the external control or internal control procedure (s) 214, respectively, and may be implemented, for example, using a web browser, so as to enable the administrator of the VPN Quickly perform all possible operations using an intuitive navigation system. In addition, control commands can be generated by entering, for example, a command line containing control commands and, in particular, some additional options and keys, for example, a VPN address and a unique identifier for a given VPN.

GUI pages can contain menus and status bars, and can contain links to other pages on which user 112 can perform various tasks. In one implementation option, all links are grouped into four groups: VPN management, VPN services, protocols and more, and the names of these groups are links themselves. When performing a specific task, user 112 may first click on the name of the group to open the corresponding panel, and then select the task, or user 112 can immediately select the task in the menu under the corresponding group. The links in the menu do the following:

VPN management: opens the VPN management panel, where user 112 can learn more about the various tasks of administering VPNs and complete any of them.

Start / Stop VPNs: opens a page where user 112 can start, stop, or restart the current VPNs 104.

Password Change: opens the password change page for the VPN administrator.

- 9 025082

Reinstallation opens the VPS reinstallation page where user 112 can restore the original VPS 104 template files if something goes wrong.

Correction mode: opens a page on which user 112 can start VPS 104 in the so-called repair mode, in order to check and correct the command line.

Backup: opens a page where user 112 can administer the backup of VPN 10 104.

Resources: opens a page where user 112 can obtain information about the current consumption of hardware resources of the user VPN 104.

VPN services: opens the Services Administration panel, in which user 112 can receive more detailed information about various tasks of administering services to perform any of them.

System processes: opens a list of processes running in VPS 104 with the ability to transmit various signals to any of them.

Connecting to the VPN due to §§Н: opens the input / output window for §§H for connecting to the VPN due to the connection via 88Н.

Logs: opens a panel of logs or protocols, on which user 112 can obtain more detailed information about the various protocols of a particular VPS 104 and can view any of them.

State Changes: Opens a list of recent changes in the state of VHS 104.

Quality of Service Alerts: Opens a list of warnings on the VChS 104 about excessive use of any hardware or modes in which the degree of hardware utilization is close to the limit.

Event log: opens a list of user 112 actions performed with the considered VPN 104 recently.

Help: opens the help system.

Exit: following this link provides the user 112 with the exit from the procedure (s) for managing certain VPNs.

The status bar may contain the following elements:

(a) GO or host name of the current VPN 104;

(b) the current state of the VPN 45;

c) the percentage of available disk space used by the VPN 104.

The status modes that can be administered and displayed by controls can be as follows.

In operation: VPS 104 is operational, so it can only be rebooted or stopped.

Out of operation: VPS 104 is stopped, so it can only be started.

Under repair: VChS 104 is under repair. User 112 cannot perform any action on VPS 104 until user 112 clicks the Finish Repair button on the Correction Mode page.

In addition to these states, VPS 104 can be in one of the transition states: mounted, mounted, started, stopped, etc. When the VPS is in a transition state, the user 112 cannot, in a safe manner, perform any actions on the VPS 104 until the transition state is completed. A description of all the possible status of the VPN is given in the section Status of the VPN.

To view detailed information on the use of resources, user 112 can follow the corresponding link that displays the current resource consumption by a specific user of the VPN 104. The System Processes Page may also be of interest. This page can display which process (s) are running on the VPS 104 so that the user 112 can control the processes, in particular those processes that cause problems during the operation of the VPS 104. For example, the user 112 may need to stop or kill a dangerous process. Otherwise, system performance may remain the same for a long time. System processes can be displayed, for example, in the form of a table indicating all processes running in the VPS 104. The table may provide the following information:

(a) GO process;

(b) the absolute priority of the process assigned to it by the process planner. The range can be, for example, from 0 (highest priority) to 39 (lowest priority). The normal priority of the process is 30;

(c) the relative priority of the process assigned to it by user 112. A negative value means that user 112 manually increased the priority, a positive value means that he reduced it;

(ά) the size of the physical memory actually used by the process (in kilobytes);

- 10 025082 (e) state of the process. Possible states can be, for example: K - readiness, or in line for execution, δ - rest state; T - damage is detected or stopped, And - at rest, without the possibility of activation; Ζ - not functional, zombies. If two letters are shown, the second letter means the following: V - does not have resident (permanently present in memory) pages; <- high priority process; N is a low priority task; B - the process has a locked page in memory;

(ί) the percentage of total processor time currently used by the process;

(e) the percentage of total RAM currently used by the process;

(B) the total amount of processor time used by the process to date;

(ί) The command that is used to start the process.

As shown in FIG. 4, a user application or client 112 calls an external interface 107 via a physical network 202 to execute a control command (for example, to terminate or kill a process within VPS 104). Since VPS 104 is a fully functional OS, such a mechanism, such as a task manager, can be called remotely using ΙΡ VPS addresses and the corresponding username and password, which provide administrator rights to user 112.

As noted above, each VPS 104 has a share of the system resources allocated to it. Resources allocated to VPNs 104 (e.g., memory, file space, I / O devices, etc.) can be isolated from each other using VPNs and other specific types of addressing.

Thus, the process can be killed safely, for other considered VPNs, since only the process related to the implementation of a specific VPN is killed within the specified VPN 104. If the existence of the process leads to instability or inaccessibility of the operating system of the VPN, easy access to the main operating system 112 does not solve the problem of the impact on this process, since different VPNs can use different instances of this process. In this case, a control mechanism is created outside the VPS 104 to provide access to the VPS applications and libraries using a common set of identifiers with the VPS managed.

Thus, in a preferred embodiment, the control interface 107 outside of the VHS 104 (which itself can be formed from several procedures or modules) can manage or check internal objects and data structures in a safe manner. An external control interface 107 may be unique to a particular VPN, 104, or may be associated with many different VPNs 104 at the same time. (For example, using different identifiers or ΙΡ addresses provides access to various settings of the VPN.)

It should be noted that the core 108 of the main operating system, when servicing a user’s call, should be able to work with the representation of the object within the allocated VPS 104 defined by a unique VPS identifier.

In one embodiment, a unique identifier, such as a PEC or a host name, is used to assign an address to VPN 104. A typical example of such an address is a web server name address (or name ΌΝδ) or a network name for small and medium-sized businesses in the context of Μδ No. (\ wogk (Microsoft network) used to access the shared resources Μδ νίηάοΥδ, or the server address Te1пе1. used to interactively log into the session. When choosing a specific address, user 112 must be sure that he will get access to the server associated with this address , and not another server, and that anyone using addresses that do not belong to this VPN 104 will not get access to this VPN 104.

Another example of VPN identification is the credentials required for logging in and authentication. In this case, even after entering the user name using the ΙΡ address common to different VPNs 104, user 112 will be authenticated in the VPNs 104 in which he is registered as a user.

The functions of the VPN in each VPN 104 typically include the creation of the corresponding files and file structures that the VPN will use 104. These files and file structures may contain, for example, administrative files, user files, etc. Information relating to specific VPNs 104 is registered in the registration database and other similar structures supported by the operating system. This information should ensure the continuous operation of the VPS 104. In addition, each VPS 104 can be assigned a unique address or group of addresses and other network resources.

The virtual private server GO discussed above can be similar to a marker or descriptor, or it can be similar to an identifier or data structure that can be used to track VPNs 104. Thus, access to VPNs 104 and related objects through management tools can be implemented by using the GO of a virtual private server and or ΙΡ the address of the VPN.

Although the above description is set forth in terms of the operation of several VPNs 104 on a single host 102, the invention equally relates to a server cluster where several computers, 102

- 11 025082 are interconnected. In addition, the present invention is applicable to any type of server, for example, a web server, a LAN server, a wide access network server, an internal network server, etc.

An example of a host 102 is shown in FIG. 5. The host 102 comprises one or more processors, such as a processor 301. The processor 301 is connected to a data communications infrastructure 306, for example, a bus or network. Various options for implementing software (software) are disclosed in terms of this example computer system. After reading the proposed description, the specialist in the field of technology will be obvious ways of implementing the invention using other computer systems and / or computer architectures.

The host 102 also contains a main memory 308, in a preferred embodiment, a random access memory (KAM), and may also include an auxiliary memory 310. The auxiliary memory 310 may include, for example, a hard disk 312 and / or a removable disk 314, which is magnetic tape drive, optical disk, etc. The removable storage device 314 reads / or writes data to the removable storage device 318 in the usual manner. The removable storage unit 318 is a magnetic tape, optical disk, or other medium that is read or written using the removable storage device 314. As follows from the description, the removable storage unit 318 may contain a storage medium suitable for use in computer technology with on it with computer programs and / or data.

In alternative embodiments, auxiliary memory 310 may include other means that enable programs and other instructions to be loaded into host 102. Such means may comprise, for example, a removable storage device 322 and an interface 320. An example of such a tool may be a removable memory chip (eg, an erasable programmable read-only memory, or programmable read-only memory), and a corresponding socket, or other removable storage units 322 and interfaces 320, which allow software and data to be transferred from removable storage device 322 to host 102.

The host 102 may also contain one or more data transfer interfaces, such as data transfer interface 324. A data interface 324 (communication interface) allows software and data to be transferred between the host 102 and external devices. Examples of the communication interface 324 can be a modem, a network interface (for example, an E1eigie1 card), a communication port, a slot and a PCMS1A card (specification of expansion modules), etc. Software and data are transmitted through the communication interface 324 in the form of signals 328, which may be electronic, electromagnetic, optical or other signals that can be received by the communication interface 324. These signals 328 are transmitted to the communication interface 324 via a path (e.g., channel) 326 communication. This channel 326 supports 328 signal transmission and can be implemented using wire or cable, fiber optics, radio frequency connections, and other communication channels. In an example embodiment, the signals 328 comprise data packets sent to the processor 301. Information representing the processed packets may also be sent as signals 328 by the processor via path 326.

The terms computer program media and computer-used media are used to generically describe data carriers, such as removable storage devices 318 and 322, a hard disk installed in the hard disk drive 312, and signals 328 carrying software to the host 102.

Computer programs (programs for electronic computers) are stored in the main memory 308 and / or auxiliary memory 310. The programs can also be received through the communication interface 324. Such computer programs, when executed, allow the host 102 to implement the invention described in the present description. In particular, computer programs, when executed, allow processor 301 to implement the present invention. If the invention is implemented using software, the software may be stored as a computer program product and downloaded to the host 102 using a removable drive 314, a hard disk 312, or a communication interface 324.

Given the present description of a preferred embodiment, it should be apparent to a person skilled in the art that certain advantages are achieved in the disclosed method and apparatus. It should also be appreciated that various modifications, adaptations, and alternative embodiments of the invention can be made while maintaining the meaning and essence of the present invention. The invention is further defined by the following claims.

Claims (7)

CLAIM 1. A system for managing virtual servers, comprising many virtual servers configured to operate on one main computer system and sharing one main operating system, each server containing an associated address and a unique identifier; each virtual server is configured to respond to customer requests; each virtual server contains an internal control procedure for managing the specified virtual server, the control procedure being activated in response to commands sent by the remote user using an address and a unique identifier; an external management interface configured to operate on the main computer system and operable outside the virtual server, and managing the virtual server, in accordance with the commands sent by the remote user, using the address and a unique identifier, the external management interface being configured to administer the specified virtual server if the specified virtual server is not responding. 2. The system according to claim 1, characterized in that it is configured to execute commands by both the control procedure and the external control interface, with the possibility of executing commands by the control interface, if the control procedure is unavailable. 3. The system according to claim 1, characterized in that the control interface is made accessible using multiple addresses. 4. The system according to claim 1, characterized in that the control interface is configured to control multiple virtual servers. 5. The system according to claim 1, characterized in that it is configured to perform the following commands: start / stop the virtual server; Change virtual server passwords reinstall virtual server; repair a virtual server; make a backup copy of the virtual server; change the characteristics of the current level of resource use; process the virtual server management command; determine the status of the virtual server; set up a virtual server; Install a custom application uninstall the user application and configure the user application. 6. The system according to claim 1, characterized in that the control interface is activated upon failure of any of the virtual servers. 7. The system according to claim 1, characterized in that at least one server is made in the form of a virtual machine or at least one server is made in the form of a virtual software execution environment.