US20050050334A1 - Network traffic management by a virus/worm monitor in a distributed network - Google Patents

Abstract

A network level virus monitoring system capable of monitoring a flow of network traffic in any of a number of inspection modes depending upon the particular needs of a system administrator. The system includes a network virus sensor self registration module coupled to a network virus/worm sensor arranged to automatically self register the associated network virus/worm sensor. The monitoring provides an early warning of a virus attack thereby facilitating quarantine procedures directed at containing a virus outbreak. By providing such an early warning, the network virus monitor reduces the number of computers ultimately affected by the virus attack resulting in a concomitant reduction in both the cost of repair to the system and the amount of downtime. In this way, the inventive network virus monitor provides a great improvement in system uptime and reduction in system losses.

Description

CROSS REFERENCE TO RELATED APPLICATIONS
• This application takes priority under 35 U.S.C. §119(e) of U.S. patent application Ser. No. 60/481,313 filed Aug. 29, 2003 (Attorney Docket No.: TRNDP009P) naming Liang et al. as inventor(s) entitled “VIRUS MONITOR AND METHODS OF USE THEREOF” which is also incorporated herein by reference for all purposes. This application is also related to the following co-pending U.S. patent applications, which are filed concurrently with this application and each of which are herein incorporated by reference, (i) U.S. patent application Ser. No. ______ (Attorney Docket No.: TRNDP009), entitled “VIRUS MONITOR AND METHODS OF USE THEREOF” naming Liang et al as inventors; (ii) U.S. patent application Ser. No. ______ (Attorney Docket No.: TRNDP101), entitled “AUTOMATIC REGISTRATION OF A VIRUS/WORM MONITOR IN A DISTRIBUTED NETWORK” naming Liang et al as inventors; (iii) U.S. patent application Ser. No. ______, (Attorney Docket No.: TRNDP014), entitled “NETWORK ISOLATION TECHNIQUES SUITABLE FOR VIRUS PROTECTION”, naming Liang et al as inventors; and (iv) U.S. patent application Ser. No. ______(Attorney Docket No.: TRNDP012), entitled “ANTI-VIRUS SECURITY POLICY ENFORCEMENT”, naming Liang et al as inventors; (v) U.S. patent application Ser. No. ______ (Attorney Docket No.: TRNDP015), entitled “ANTI-COMPUTER VIRAL AGENT SUITABLE FOR INNOCULATION OF COMPUTING DEVICES”, naming Liang et al as inventors; and (vi) U.S. patent application Ser. No. ______ (Attorney Docket No.: TRNDP013), entitled “INNOCULATION OF COMPUTING DEVICES AGAINST A SELECTED COMPUTER VIRUS”, naming Liang et al as inventors.
BACKGROUND OF THE INVENTION
• The present invention relates generally to information analysis and screening using a computer, and, specifically, to configurations and methods for intercepting and removing computer viruses and worms from transmitted media.
• With the rising popularity of the Internet, there are now millions of users connecting to the Internet daily from their host computers to conduct e-commerce transactions, perform searches for information and/or download executable programs to enhance the capability and performance of their own host computers. The interaction between these users and the other host servers on the Internet generally involves the transfer of some amount of data, which may include both static displayable information and executable computer code. Generally speaking, static displayable information refers to static information to be displayed at the host computer while executable code or an “executable” refers to computer instructions configured to be executed at the host computer to perform some task.
• In general, the vast majority of the downloadable data from the Internet represents useful or at least non-harmful content material. However, there exists a class of executable code that, if downloaded and executed at host computers, may wreak havoc with the operating system, the hardware, and/or other software of the host computers. These executables include what are commonly referred to as computer viruses and/or worms.
• A computer virus is a piece of programming code usually disguised as something else that causes some unexpected and usually undesirable event (for the victim). Viruses are often designed so that they automatically spread to other computer users across network connections. For instance, viruses can be transmitted by sending them as attachments to an e-mail message, by downloading infected programming from other web sites, and/or by importing them into a computer from a diskette or CD-ROM. The source application that deals with the e-mail message, downloaded file, or diskette is often unaware of the virus. Some viruses wreak their effect as soon as their code is executed; other viruses lie dormant until circumstances cause their code to be executed by the computer. Some viruses can be quite harmful, causing a hard disk to require reformatting or clogging networks with unnecessary traffic.
• Computer worms are very similar to viruses in that a worm is a small piece of software that uses computer networks and security holes to replicate itself. A copy of the worm scans the network for another machine that has a specific security hole. Once the security hole has been found, the worm copies itself to the new machine using the security hole, and then uses the newly infected computer to start replicating itself in order to infect other computers connected thereto. Although a worm does not alter files but resides in active memory and duplicates itself, the worm uses parts of an operating system that are automatic and usually invisible to the user. Therefore, it is common for worms to be noticed only when their uncontrolled replication consumes system resources, slowing or halting other tasks.
• To combat worms, users and administrators of computer networks (such as corporate local area networks or wide area networks) have long employed a variety of tools designed to detect and block worms from infecting a computer system. In a corporate local area network (LAN), for example, network administrators may employ proxy servers (which are disposed between the host computers of the LAN and the Internet) as well as individual computers to perform any of a number of defense strategies designed to prevent infection by a worm. One such defense strategy relies upon behavioral monitoring of computer actions. In behavioral monitoring, a historical database of actions taken by every computer is maintained that is then used by a monitoring program (heuristic engine) to compare to current actions taken by a particular computer. In those cases where current actions are deemed by the behavior monitoring program to be substantially different from the historical norm, the behavioral monitoring program flags that particular computer as being possibly infected by a worm. Once so flagged, appropriate actions can be taken.
• In day-to-day efforts against computer viruses and other terminal device viruses, an end user is constantly looking for ways to inoculate against such viruses. Even in the case of corporate networks that are closely guarded by an anti-virus firewall and various other virus protection software and protocols, some viruses still manage to penetrate and infect the network resulting in substantial harm since conventional anti-virus technology generally relies on already identified viruses. In particular, conventional anti-virus protection is usually effective against known computer viruses, but may be ineffective in blocking unknown viruses. Therefore, terminal devices such as computers connected to a local area network (LAN) or wide area network (WAN) are generally unable to include effective anti-virus protection against unknown viruses using conventional anti-virus software.
• When the terminal device or computer connected to a network is subject to attack by an unknown virus penetrating into the network, it is the responsibility of network managers to guard against such attacks and the restore the network to normal operating status as quickly as possible. The level of preparedness in a network is dependent upon knowing the probability of a virus to successfully penetrate the corporate network.
• Intrusion Detection System (IDS) products neutralize the network-type attacks by scanning for abnormal network packets at protocols layers, including a method called Application Behavior Monitoring (ABM) at the host base IDS. ABM keeps track of behavioral patterns of target applications and protects the network system by allowing the benign (known) behavior patterns by disallowing or blocking and the unknown or malign ones.
• Conventional anti-virus software sets a particular alert level to early detection of virus outbreaks for system administrators of network systems. The setting of the alert level becomes very important. If the alert level is set too low, it may invite an erroneous determination of a computer virus such that benign applications are deemed viral by mistake. If the alert level is set too high, certain computer viruses will be undetected and allowed into the network.
• Conventional anti-virus software still relies on the support system at the anti-virus service provider to generate cures. Such practice is heavily reliant on the response time at the service provider in procuring the virus sample, implementing the virus analysis, generating the appropriate cures, and deploying them to the end users. Though such support systems may be effective at certain levels, certain end users (such as system administrators of corporate networks) still require solutions that provide better lead time and effectiveness in countering sudden outbreaks of computer viruses.
• There is thus a general need in the art for a network level anti-virus method and system overcoming at least the aforementioned shortcomings in the art. In general, there is a need in the art for an anti-virus method and system having multilevel anti-virus functions for anticipating and detecting computer virus outbreaks. In particular there is a need for a system and method that preserves network bandwidth by minimizing the impact on network traffic.
SUMMARY OF THE INVENTION
• To achieve the foregoing, and in accordance with the purpose of the present invention, a system and method for monitoring a network for computer viruses is described.
• In a first embodiment, in a distributed network of interconnected computing devices, a network virus/worm monitor. The monitor includes a network virus/worm sensor, and a traffic controller in communication with the network virus/monitor sensor and the network operable in a number of modes wherein in a first mode the bandwidth of the network is substantially unaffected by the traffic controller during a virus/worm sensing operation by the network virus/worm sensor wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic, the virus/worm sensor causes the traffic controller to switch to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network.
• In another embodiment of the invention, in a distributed network of interconnected computing devices, a method of managing network traffic by a network virus/worm monitor, having a network virus/worm sensor. During a virus/worm sensing operation, the bandwidth of the network is substantially unaffected by the traffic controller by the network virus/worm sensor in a first mode wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic. Switching when a computer virus or computer worm is detected to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network.
• In yet another embodiment, computer program product executable n a distributed network having a number of server computers and associated client devices, computer program product managing network traffic by a network virus/worm monitor, having a network virus/worm sensor is described. The computer program product includes computer code for performing a computer virus/worm sensing operation, the bandwidth of the network is substantially unaffected by the traffic controller by the network virus/worm sensor in a first mode wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic, computer code for switching when a computer virus or computer worm is detected to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network, and computer readable medium for storing the computer code.
BRIEF DESCRIPTION OF THE DRAWINGS
• The invention, together with further advantages thereof, may best be understood by reference to the following description taken in conjunction with the accompanying drawings.
• FIG. 1 shows a distributed network having a network virus monitor in accordance with an embodiment of the invention.
• FIG. 2 is the distributed network ofFIG. 1 having an active network virus monitor.
• FIG. 3 shows the distributed network ofFIG. 1 whereby the network virus monitor is registering all connected client devices.
• FIG. 4 shows the distributed network ofFIG. 3 whereby the network monitor is operating in standby mode and has flagged a virus event.
• FIG. 5 shows the distributed network ofFIG. 2 whereby the network virus monitor is operating in inline mode.
• FIGS. 6A-6B shows an exemplary distributed network having a segmented portion thereof due to a virus outbreak and virus clean procedure in accordance with an embodiment of the invention.
• FIG. 7 shows an exemplary virus structure and associated anti-virus structure in accordance with an embodiment of the invention.
• FIG. 8 illustrates an exemplary virus monitor in accordance with an embodiment of the invention.
• FIG. 9 illustrates the virus monitor ofFIG. 8 operating in standby mode.
• FIG. 10 illustrates an exemplary security module and file scan module of the virus monitor shown inFIG. 9 operational in standby mode.
• FIG. 11 illustrates an exemplary security module and file scan module of the virus monitor shown inFIG. 9 operational in inline mode.
• FIG. 12 shows a flowchart detailing a process for monitoring a network for a virus in accordance with an embodiment of the invention.
• FIG. 13 shows a flowchart detailing a process for introducing a temporary new client device to the network in accordance with an embodiment of the invention.
• FIG. 14 shows a flowchart detailing a process for introducing a non-temporary new client device to the network in accordance with an embodiment of the invention.
• FIG. 15 shows a flowchart detailing a network segment isolation process in accordance with an embodiment of the invention.
• FIG. 16 shows a flowchart detailing a virus cleaning process in accordance with an embodiment of the invention.
• FIG. 17 shows a flowchart detailing a process for performing an automatic clean/cure of a group of infected computers in accordance with an embodiment of the invention.
• FIG. 18 shows an process for automatically cure and clean in accordance with an embodiment of the invention.
• FIG. 19 shows a system block diagram of a computer system used to execute functions of the present invention including the scanning, deletion, truncation, and quarantine of data packets suspected of harboring computer viruses, worms etc.
DETAILED DESCRIPTION OF THE INVENTION
• Reference will now be made in detail to a preferred embodiment of the invention. An example of the preferred embodiment is illustrated in the accompanying drawings. While the invention will be described in conjunction with a specific embodiment, it will be understood that it is not intended to limit the invention to that particular embodiment. To the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims.
• At the network level, conventional anti-virus software still relies on the support system at the anti-virus service provider to generate cures. Such practice is heavily reliant on the response time at the service provider in procuring the virus sample, implementing the virus analysis, generating the appropriate cures, and deploying them to the end users. Although such systems may be effective at some levels, certain end users, (such as system administrators of corporate networks) still require solutions that provide better lead time and effectiveness in countering sudden outbreaks of computer viruses. This is particularly important as the size of networks increase and the feasibility of incorporating anti-virus software for every component in the network becomes prohibitively expensive. In addition, once a computer has violated the integrity of a network, it is extremely difficult, time consuming, and expensive to both identify and clean the affected computers. This is particularly significant since all infected computers must be identified, cleaned, and inoculated against future infections.
• Accordingly, the invention describes a network level virus monitoring system capable of monitoring a flow of network traffic in any of a number of inspection modes depending upon the particular needs of a system administrator. The monitoring provides an early warning of a virus attack thereby facilitating quarantine procedures directed at containing a virus outbreak. By providing such an early warning, the network virus monitor reduces the number of computers ultimately affected by the virus attack resulting in a concomitant reduction in both the cost of repair to the system and the amount of downtime. In this way, the inventive network virus monitor provides a great improvement in system uptime and reduction in system losses. In order to improve the efficiency of such a system when included in a distributed network having a number of server computers and associated client devices, the system includes a network virus sensor self registration module coupled to the network virus/worm sensor arranged to automatically self register the associated network virus/worm sensor.
• In one embodiment of the invention, the monitoring system includes a virus monitoring device coupled to a distributed network of a number of interconnected computing devices. In the described embodiment, the virus monitoring device includes a virus monitor arranged to detect a network computer virus in a flow of traffic in the network. The monitoring devices also includes a network computer virus outbreak warning unit coupled to the network monitor arranged to provide an early warning of the network computer virus. A network computer virus warning response unit responsive to the network computer virus outbreak warning unit is arranged to isolate a network segment affected by the network computer virus, or to block the network computer virus.
• The invention will now be described in the context of a network of interconnected client devices. Such client devices can include desktop computers, laptop computers, thin client devices such as personal digital assistants (or PDAs), embedded appliances, and so on. Although described using a network of interconnected computers and computing devices, the scope and intent of the invention extends to all those devices for which viruses and/or worms find worthy of attack. Furthermore, for sake of this discussion only, the interconnected devices communicate with each other by way of a packet based communication protocol. Such protocols include TCP(Transmission Control Protocol)/IP(Internet Protocol) which is well known in the art. TCP is a set of rules used along with the IP to send data in the form of message units between computers over the Internet. While IP takes care of handling the actual delivery of the data, TCP takes care of keeping track of the individual units of data (called packets) that a message is divided into for efficient routing through the Internet.
• For example, when an Hyper Text Transfer Protocol (HTTP) file is sent from a Web server, the Transmission Control Protocol (TCP) program layer in that server divides the file into one or more packets, numbers the packets, and then forwards them individually to the IP program layer. Although each packet has the same destination IP address, it may get routed differently through the network. At the other end at the client computer, the TCP layer reassembles the individual packets and waits until they have arrived to forward them as a single file. It should be noted, however, that the invention is well suited for use with other communication protocols such as SMTP etc.
• Accordingly,FIG. 1 shows a virus monitoring system implemented on a distributednetwork100 having a network virus monitor102 in accordance with an embodiment of the invention. As shown,network100 is a distributed computing environment that includes a number of individual client devices104-116. The client devices can take the form of any computing device having on-board memory susceptible to attack by a computer virus or worm. Such devices include but are not limited to computers (both desktop and laptop), and hand held thin client devices such as personal digital assistants (or PDAs).
• Generally, a network is divided into a hierarchy using a geographical classification, a management classification and detailed information. The hierarchy is accordingly displayed in the form of a map having a number of levels. Accordingly,network100 is structured along the lines of a tiered network architecture with a hierarchy of three tiers. In this particular architecture, various multi-service switches are used to provision subscriber services at the first tier of the network (i.e., the Internet backbone, for example).
• Atier1 switch (shown as switch118) can be used to consolidate traffic from many subscribers and may also may perform traffic shaping, depending on the network architecture. In some cases, thetier1switch118 then can be connected to atier2switch120 which, in turn, is connected to atier3switch122, thereby providing further traffic concentration. In this way, the tiered architecture provides a modular way of extending the network's scalability, enabling the carrier to add switching capacity to the network topology as subscriber demand requires. Accordingly,network100 is described in terms of a multiple-layer network, including a first tier, second tier, third tier, etc. For example, the client devices104-116 form a lowest level tier (i.e., the third level) while the virus monitors102 form a next higher order tier (i.e., the second level) and so on.
• In addition to providing scalability, the tiered architecture ofnetwork100 provides for topologically advantageous positioning of thenetwork virus monitor102. For example, in the instant case, virus monitor102 is placed between thetier2switch122 and thelower level tier3 switch124 to which the various client devices104-116 are coupled. In this way, all network traffic between thetier2 switch (which may be coupled directly to the Internet backbone, for example) and any of thetier3 switches can be monitored by the virus monitors102 at a point prior to any of the client devices. By providing a bulwark against a potential virus attack, the virus monitors102 provide a focal point for virus detection, virus outbreak prevention, and, if needed, virus outbreak cleanup and restoration that, in turn, effectively protect the various client devices from the attacking virus. It should be noted, that adocking port125 can be included innetwork100 arranged to accept temporary, or visitor, client devices.
• In the described embodiment, each of the virus monitors102 is coupled to acontroller126 that is, in turn, coupled to a server computer128 (or a number of server computers) each of which can be configured as a stand alone server computer having various user interfacing devices, such as monitor, keyboard and mouse described in some detail below. In the described embodiment, theserver computer128 is a network-connectible computer or a server device, such as a workstation running an UNIX operating system, or a computer running the WindowsNT™ or WindowsXP™ operating system. Thecontroller126 includes arules engine130 used to store and source a plurality of detection rules for detecting computer viruses and an outbreak prevention policy (OPP) distribution andexecution engine132 that provides a set of anti-virus policies, protocols, and procedures suitable for use by a system administrator for both preventing viral outbreaks and repairing any subsequent damage caused by a viral outbreak. It should be noted that the detection rules, policies, and procedures manifest in therules engine130 and in the OPP distribution andexecution engine132 can be periodically updated by way of theserver computer128 as needed.
• Moreover, the controller126 (that in some cases may be located in a separate location) also serves to determine whether the abnormal events observed by thevarious monitors102 are potentially computer viruses based, in part, upon statistical results of the observed abnormal events. In addition, theserver128 can provide virus cleaning agents derived from and based upon those viruses both known and unknown but subsequently analyzed. In this way, even those situations where a previously unknown viral agent attacks various components ofnetwork100, the viral analysis provided by theserver128 can facilitate both quarantine operations (by way of network segmentation protocols) and subsequent viral clean up (by way of viral cleaning agents) and repair (by way of virus repair agents). In addition to providing palliative, or remedial, services, theserver128 is also capable of providing a viral inoculation agent used to prevent future attacks on either those computers affected (and subsequently cleaned) and those computers unaffected but vulnerable to future attack by the viral agent. In the described embodiment, some of the client devices may also include a client rules set (CRS)134 that stores rule information and parameters for detecting computer viruses. It should be noted that the rule information and parameters for detecting computer viruses stored in theCRS134 can be preinstalled in each device, or if not present, can be downloaded to that particular client by theserver128 either directly or by way of thecontroller126 and/orvirus monitor102.
• During an installation phase (or an initialization phase), each of the virus monitors102 self register by collecting certain environment information (such as the IP address of all relevant client devices) as well as self configuration withinnetwork100 by, for example, determining an appropriate IP address forvirus monitor102, itself. In addition to self registering, virus monitor102 will search for an appropriate controller126 (such as a nearest controller, for example) and once found, will register with it accordingly.
• Referring specifically toFIG. 2, once virus monitor102 has completed the installation and/or registration process, thecontroller126 receives an updatedOPP file135 of the most current set of policies and procedures and rules set136 of the most current virus detection rules from the server128 (if needed). Once received by thecontroller126, the OPP file135 and the rules set136 are forwarded to each of the virus monitors102 in order to provide the latest rules and virus filters and patterns as deemed appropriate. For example, the OPP file135 is used by the OPP distribution andexecution engine132 to apply appropriate virus policies (such as particular file types to scan for viruses), while the rules set136 is used by therules engine130 to practice specific virus detection rules. It should be noted that the various monitors, controllers, and servers can be configured in any operating platform. For example, such platforms include embedded Linux, PC based Linux or Windows (as above) and in some cases when higher level resources are required, Sun SPARC™ platforms and the like can be used.
• Once the various virus monitors102 have been updated with the most current rules and policies, virus monitor102 will perform an anti-virus security policy enforcement procedure whereby each of the client devices coupled to virus monitor102 is queried in order to determine if that client device has the appropriate and proper anti-virus software installed. Such appropriate anti-virus software can include any recognized anti-virus software from any number of recognized vendors such as Trend Micro of Cupertino, Calif., and the like. It should be noted, however, that any time a new client device is coupled to virus monitor102, the newly connected client device will also be queried in a similar manner.
• In those situations where a client device is found to not have the appropriate anti-virus software installed, virus monitor102 has any number of options for response. In most cases, virus monitor102 will direct the target client device (i.e., the client device found to not have the appropriate anti-virus software) to an anti-virus installation server138 (which may actually be the server128) and block any traffic to/from the target client device and all other addresses until such time as the appropriate anti-virus software has been properly installed.
• For example, virus monitor102-1 sends aquery140 to each of the client devices110-116 requesting confirmation that each has installed therein the appropriate anti-virus software as determined by the policies contained in theOPP file135. Upon receiving thequery140, each of the client devices checks for confirmation that the appropriate anti-virus software is indeed present. If, say in the case of theclient device116, that it is determined that either no software is present or the installed software is not appropriate (based upon the policies in theOPP file135, for example), theclient device116 is directed only to the anti-virussoftware installation server138 and no other. At this point, optionally, a user interface can be displayed on theclient device116 indicating that until such time as the proper software has been installed, that theclient device116 will be prevented from communication with other systems. It should be noted, however, that some transmission protocols (such as HTTPS) that are essentially immune from viral infections due to the encryption thereof could be used).
• Once the appropriate anti-virus software has been installed in theclient device116, virus monitor102-1 relinquishes the lock on the communication channels for theclient device116. In this way, theclient device116 can communicate with the other devices ofnetwork100.
• In those situations where a temporary user wishes to connect intonetwork100, a determination must be made whether or not the visitor client device is compliant with the current policies and rules. Typically, it is not advisable to grant a temporary user a license to use the anti-virus software since that would be both costly and limit the number of available licenses for other users. From the standpoint of the visitor, installing anti-virus software for only a limited time is also not desirable since the software could possibly interfere with anti-virus software already present on the computer and/or require computing resources that are not readily available. Therefore, in those situations where a visitor client device is to be connected to the network another approach is required.
• Specifically, in a particular embodiment of the invention, when a visitor connects a heretofore unknown (to network100)client device125 as shown inFIG. 3, virus monitor102 will query thevisitor client device125 for the presence of appropriate anti-virus software. If it is determined that thevisitor client device125 does not have the appropriate anti-virus software installed, then access to all addresses other than an anti-virussoftware installation server138 are blocked until a scan is made of the memory of thevisitor client device125 by a virus scan server module142 (which may or may not be part of the anti-virus software installation server138).
• Once thevisitor client device125 has been deemed to be free of computer viruses, thevirus scan server142 passes ause token144 to thevisitor client device125. Theuse token144 is typically valid for a limited amount of time (an hour, for example) after which the token must be re-validated. However, during the period of time that theuse token144 is valid, all channels to/from thevisitor client device125 are open and available for the passage of network traffic. In order to re-validate theuse token144, thevisitor computer125 must request a new token which will be granted based, in part, upon a determination that thevisitor client device125 has remained free of computer viruses (or worms).
• At this point, all client devices (including any visitor client devices) have been confirmed to either have the appropriate anti-virus software (or have a valid use token) and virus monitor102 is ready to begin monitoring network traffic for the presence of computer viruses (and/or worms).
• In general, virus monitor102 monitors activities ofnetwork100 for abnormal events according to both the policies and rules and generates abnormal report if abnormal events are detected which are then transferred to thecontroller126. In some embodiments, thecontroller126 determines an alert level for the detected abnormal events while in other embodiments, thecontroller126 forwards the abnormal events information to theserver128 which will evaluate the data and if determined to be appropriate will send an early virus warning to other virus monitors innetwork100. In some cases, the abnormal reports data is forwarded to a virus attack warning server that decides a course of action to take in order to prevent a spread of the virus. Such courses of action include whether or not to quarantine affected segments ofnetwork100, generate and distribute a virus cleaning agent to the affected segment, inoculate other computers in the network to prevent the spread of the virus, and finally, if possible, repair any damage caused by the virus outbreak.
• In order to protectnetwork100, the virus monitors102 continuously monitor network traffic for potential viral attacks. One of the prime considerations of any network is the available bandwidth in that anything that unnecessarily restricts the bandwidth (i.e., the unimpeded flow of network traffic) must be avoided if at all possible. Therefore, in order to minimize the impact on the flow of network traffic (and therefore preserve bandwidth), the virus monitors102 are initially set to run in what is referred to as stand-by mode. By stand-by mode it is meant that essentially all data packets are allowed to continue to flow innetwork100 with the caveat that virus monitor102 will use a copy of the packet in order to determine whether or not there is a virus present.
• Referring toFIG. 4, virus monitor102 is monitoring the flow of data packets that constitute the network traffic flow (represented as the network traffic flow T1) in standby mode. In this way, the bandwidth ofnetwork100 is minimally affected since the network traffic flow remains substantially constant both before and after virus monitor102. This preservation of network bandwidth is accomplished by the fact that virus monitor102 monitors the network traffic (i.e., the constituent data packets) by copying all of the data packets and using the copied data packet for its analysis. In this way, there is effectively no loss of data packets due to the actions ofvirus monitor102. In some embodiments, a determination is made of the data packet type, and based upon the packet type, only those packet types deemed vulnerable to virus attack are copied. In this way, the resources required to perform the virus monitoring is limited to only that required to adequately monitor the traffic flow.
• In the case where virus monitor102 has detected a possible virus in one or more of the data packets (or in the case where a potential intruder attack is underway), virus monitor102 generates an event flag. This event flag provides information based upon the detected virus using both the rules set136 and the OPP file135 as well as any other data deemed useful. Typically, the event flag is passed directly to thecontroller126 which may, in some cases, forward the event flag to theserver138 for further analysis and/or disposition of any remedial actions, if any. This collaborative nature of the inventive virus monitoring system is well documented and described in co-pending U.S. patent application Ser. No. ______, Attorney Docket No. 87152491-002027 entitled, “MULTILEVEL VIRUS OUTBREAK ALERT BASED ON COLLABORATIVE BEHAVIOR” by Liang et al filed ______ which is incorporated by reference herein in its entirety for all purposes.
• In some cases, the event flag represents a potential threat so severe that the operation mode of virus monitor102 is immediately changed from the standby mode to what is referred to as the inline mode without intervention from thecontroller126 as shown inFIG. 5. In the inline mode, all data packets in the traffic flow T1 are analyzed without copying such that those data packets determined to be (or suspected of being) infected are not allowed to pass back into the traffic flow (in this case T1 is greater than T2). In this the virus is blocked from passing to and throughoutnetwork100. In other instances where the event itself does not trigger virus monitor102 to change operations mode to the inline mode, amode change command506 from either thecontroller126 or theserver128 is used to trigger the mode change. In this way, the inventive anti-virus system has the added advantage of delegating authority to the virus monitors in those situations where speed is of the essence to contain a potential viral outbreak. On the other hand, in those cases where the threat is less clear, or further analysis is required, the onus of determining the threat potential and execution of a defense plan can be focused in higher level analysis engines (such as a system administrator, for example) thereby reducing false alarms and unnecessary system shutdowns.
• It should be noted, that although not explicitly shown in the various figures, the number of virus monitors can be as large a number as necessary to adequately monitor the traffic flow. Therefore, in the case of a nascent virus attack, it is very desirable to determine as quickly as possible both the extent of the virus attack and the probability of the attack becoming a general virus outbreak that threatens the integrity of theentire network100.
• Therefore, each of the virus monitors102 that have detected a virus or viruses in the associated traffic flow will dispatch a corresponding event report to the associatedcontroller126. The various controllers, in turn, will forward the various event reports to theserver128 where they will be collated and analyzed in order to determine if a virus warning508 should be generated. In the case where a virus warning is generated, the virus warning506 is dispatched to thosecontrollers126 that theserver128 has determined to be most likely affected by the virus outbreak. In this way, any system administrator(s) can review the current state ofnetwork100 and be apprised of the potential threat for the system as a whole or for selected segments as might be considered important.
• Once a determination has been completed that a virus outbreak is in progress, the server128 (or in some cases, one or more of the controllers126) will institute an attempt to contain the virus outbreak using a number of tools. One such tool is referred to as network segment isolation, which as the name suggests, physically isolates those segments ofnetwork100 deemed to be affected by the virus from those segments deemed to be most likely unaffected but potentially threatened by the virus. For example, inFIG. 6A, the controller102 (as directed by theserver128 in this example) has instituted a network segment isolation protocol whereby a segment of thenetwork602 has been isolated from the rest ofnetwork100. Thesegment602 includes all client devices104-125 (including any visitor devices that may happen to be connected to network100 at that time). Once isolated, all traffic from the affected client devices can no longer flow freely throughout the network in order to contain the virus. Once a number of client devices have been identified as most likely to be compromised by a virus V, (such asclient devices104 and106 in this example), the affected client devices and restricted in such a way that each of the affected client devices are blocked from communication with even those clients devices in the affected network segment. For example, the affectedclient devices104 and106 can only communicate between each other and not theother client devices108 and125 in thenetwork segment602.
• Once the affected computers have been identified, a virus cleaning agent will be identified that when used has the effect of both cleaning the affected computers, inoculating the cleaned computers from subsequent infections, and inoculating unaffected, but threatened computers, from infection of the virus.
• There are at least two components to every virus and more likely three components.FIG. 7 illustrates a representativecomputer virus structure700. Typically, thevirus structure700 includes a detection module, an infection module, and a payload that is an action it performs on the infected computer. The payload represents those actions that the virus carries out separately from replication. Payloads can vary from the annoying (for example, the WM97/Class-D virus, which repeatedly displays messages such as “I think ‘username’ is a big stupid jerk”), to the disastrous (for example, the CIH virus, which attempts to overwrite the Flash BIOS, which can cause irreparable damage to certain machines).
• However, before a payload has the chance to inflict any damage, the computer virus must be introduced into the computer. This action is accomplished by the actions of thedetection module702 and theinfection module704. Thedetection module702 determines if a particular computer has already been infected by the virus V. If not, then the virus V is introduced into the appropriate portion of the computer where the infection (or sometimes referred to as the replication module or portion) takes over to replicate the virus V as often as possible. Once the virus V has been successfully introduced and replicated, each virus instance will execute its associatedpayload portion706 to the detriment of the computer system
• However, according to one embodiment of the invention, ananti-virus agent710 can be developed using thevirus structure700 that has the effect of cleaning affected computers, inoculating those computers (and others) from subsequent infection, and if necessary, repairing any damage caused by an executedvirus payload portion706. In order to accomplish these goals, the anti-virus structure V1 uses a modified (albeit recognizable from the standpoint of the virus V)virus structure710. For example, theanti-viral detection module712 still identifies those computers affected by the virus V but unlike thevirus detection module702, theanti-virus detection module712 continues the infection process by introducing theanti-virus infection module714. In this way, the original virus V is overwritten by the anti-virus V1 thereby setting the stage of eventual clean up and repair. Once the anti-virus V1 is introduced, an anti-virusV1 payload portion716 repairs any damage caused by the original virus V. It should be noted, that the derivation of the anti-virus V1 is directly related to the structure of the original virus V (much like an antibody related to an associated biological virus) and therefore is effective against a particular virus, class, or group of viruses.
• Returning out attention toFIG. 6A, once the affected client devices have been identified and isolated, theserver128 releases and directs the anti-virus agent V1 to the affected computers (which in this example areclient devices104 and106). The anti-virus agent V1 proceeds to identify all computers in thenetwork segment602 and begins to systematically “infect” all computers in thenetwork segment602. For theinfected client devices104 and106, thedetection module712 of the anti-virus V1 ignores the fact that each device is already infected with the computer virus V and proceeds to “infect” these devices with the anti-virus agent V1. The anti-virus V1 then proceeds to overwrite the original virus V in thecomputers104 and106 and executes therepair payload portion716. The effects of therepair payload716 again depends upon the specific damage caused by the original virus V and therefore is specifically linked to the virus V and any related viruses. But in any case, the cleaned and repaired computer is then inoculated by the anti-virus V1 in such as way that no subsequent infection by the virus V or related viruses is likely.
• For those computers uninfected by the virus V, the anti-virus agent V1 is used to inoculate (or “lock the door” so to speak) those computers against subsequent infection by the computer virus V. Once it has been determined that all computers in thenetwork segment602 have been either cleaned, repaired and inoculated or merely inoculated, the quarantine of the network segment602 (and more importantly the formerly infected with the computer virusV client devices104 and106) is ended. At some point, however, a decision is made whether or not to inoculate all the client devices innetwork100 against the virus V. The decision must take into account the virulence of the computer virus V, the effects of the computer virus V, and any potential for disruption ofnetwork100 caused, at least in part, by the inoculation process. This decision can be made at the system administrator level, or in some cases, can be based upon criteria set in the OPP.
• For example, referring toFIG. 6b,shows the anti-virus agent V1 being directed at a number of infected client devices having been infected by the virus V (i.e.,client devices104 and106). In addition, a number of heretofore uninfected client devices (i.e.,108,125, and110-114) have been inoculated by the anti-virus agent V1 against future infections by the virus V.
• Virus Monitor
• Turning now to specific implementations ofvirus monitor102, it is well to note that the described embodiments are merely exemplary and do not limit either the scope or intent of the invention.
• Accordingly,FIG. 8 illustrates avirus monitor800 as one possible implementation ofvirus monitor102. Accordingly, thevirus monitor800 includes atraffic controller802 coupled tonetwork100 by way of anetwork interface804 that includes an intruder detection system (IDS)module806 for evaluation of potential intruder attacks described in co-pending U.S. patent application Ser. No. ______, Attorney Docket No. 87152491-002027 entitled, “MULTILEVEL VIRUS OUTBREAK ALERT BASED ON COLLABORATIVE BEHAVIOR” by Liang et al filed ______ which is incorporated by reference herein in its entirety for all purposes. Such intruder based attacks include a Denial of Service (DoS) attack whereby a large number of requests are made to a particular server computer within a small period of time resulting in the attacked server computer being unable to provide access to other, legitimate, requesters. TheIDS module806 determines an associated alert level based on the volume of the data traffic flow at the virus monitor800 in a unit time interval which is designated as being abnormal if the volume of the data traffic flow is larger than a predetermined value in a predetermined time period.
• Typically, a host base IDS (not shown) sets an alert threshold very high in order to reduce the rate of false alarms in detecting viruses, which may cause inefficiencies and inflexibilities in dealing with virus outbreaks. In contrast, the collaborative anti-virus system adopts multilevel alert thresholds, with the highest alert thresholds being comparable to those of a host base IDS. Below the highest threshold, at least two lower thresholds are maintained in grouping activities at different levels of potential virus outbreak. It should be noted at this point, that in some embodiments, thetraffic controller802 can be a distributed type controller and can be located in a remote location. By remote location, it is meant that thetraffic controller802 can be implemented as discrete components each of which are in communication with each other but not within the same physical container. However, in the described embodiment, thetraffic controller802 is included as a separate component in a larger device (such as virus monitor102) suitably disposed to monitornetwork100.
• Coupled to thetraffic controller802 is asecurity module808 arranged to apply all relevant policies and rules (as provided by thecontroller126, for example, by way of acontroller interface810 and a controller link812). Thesecurity module808 also provides a determination whether or not a particular data packet is of a type likely to be infected by a computer virus. For example, it is unlikely that encrypted type data packets (such as those following the HTTPS protocols) are likely to be infected by a virus and are thus not passed on for analysis by afile scan module814.
• Operationally, a data packet will be directed to thesecurity module808 by thetraffic controller802. The data packet can be either a copy of an original data packet when thevirus monitor800 is in the standby mode or the data packet can be the original data packet when the virus monitor is in the inline mode (both described in more detail below). Once at thesecurity module808, thesecurity module808 makes a determination whether the data packet is one most likely to be affected by a particular virus based upon a combination policies included in the OPP file135 and the rules set136. If thesecurity module808 determines that the data packet in question requires a virus scan, the data packet is passed to thefile scan module814 for further analysis. It should be noted that thefile scan module814 is used to scan any particular file for any virus and/or worm as determined by thesecurity module808 using those characteristics of the virus or worm known to thesecurity module808. At any point, any of the modules (i.e.,traffic controller802,security module808, or the file scan module814) can file a report in real time or as a log, or both, providing details of its operations and results. In most cases, thevirus monitor800 is operationally set in what is referred to as standby mode in response to a selection signal S.
• FIG. 9 shows a particular example of the virus monitor800 in standby mode in accordance with an embodiment of the invention. While in standby mode, a datapacket copier unit902 included in or coupled to thetraffic controller802 copies every data packet transferred fromnetwork100 by way of thenetwork interface804. It should be noted that in some implementations, a data packet type analysis is performed prior to the intended arrival at the datapacket copier unit902 at a data packet type analyzer (not shown). In some cases, if a particular data packet type is deemed unlikely to be affected by a particular computer virus, that data packet is not copied but is merely sent on its way in the network traffic flow thereby preserving computational resources for those data packets deemed more problematic.
• Once the data packet in question is copied by the datapacket copier unit902, only the copied data packet is passed to thesecurity module808 for further analysis. It should be noted, however, that even in those situations where a copied data packet is determined to be infected with a virus, the original data packet still remains in the network traffic flow presenting a potential infection threat. It is for this reason that optionally the virus monitor800 can be switched immediately to the inline mode internally without the need for thecontroller126 to intervene based upon analysis performed by thefile scan unit814.
• FIG. 10 shows a portion of virus monitor1000 that is a particular embodiment of the virus monitor900 ofFIG. 9. In the described embodiment, thesecurity module808 includes apacket protocol determinator1006 arrange to receive a copied data packet from the datapacket copier unit902. It should be noted, that in some embodiments, the copied data packet can be stored in a data buffer prior to acquisition by the data packet determinator1002. In this way, the virus monitor900 can take advantage of any advantages due to pipelining the flow of data packets. The purpose of thepacket protocol determinator1006 is to evaluate the received data packet for the data packet protocol type. For example, the data packet protocol type can be any number of protocols, such as HTTP, HTTPS, IMAP, POP3, SMTP, etc. not all of which are susceptible to being infected by a computer virus. Therefore, once a data packet protocol type has been identified, acomparator unit1008 determines if the data packet should be processed further by the file scan module1004 or, in most cases, trashed at agarbage collection unit1010. The basis for the determination is typically found in theOPP FILE135 or rules set136 which can be updated as needed whenever necessary.
• In the case where thecomparator unit1008 has determined that the data packet should be scanned by thefile scan module1006, the data packet is passed to avirus scan unit1012 that determines if the data packet has been infected by a virus. This determination is based upon, for example, comparing the data packet to known patterns for a particular virus or group of viruses. In the case where the data packet is found to be clean of any virus, the data packet is discarded at thegarbage collection unit1010 whereas, if the data packet is found to be infected with a virus, the virus is analyzed at avirus analyzer unit1014. In the described embodiment, the virus analysis can includes parsing the virus into its constituent components thereby exposing a payload portion of the virus. In this way, an anti-virus agent can ultimately be produced that is useful for both cleaning and repair. Optionally, in some cases, avirus warning module1016 sends avirus warning1018 to thecontroller126. The virus warning has the effect of notifying other monitors and their associated controllers of a potential virus attack in an attempt to thwart a potential virus outbreak. In this way, any effects of a virus outbreak can be mitigated. The analysis from thevirus analyzer1014 is, in turn, presented to thecontroller126 for further collation, correlations, and collaboration by theserver128, if necessary.
• In addition to operating in the standby mode (in order to preserve network bandwidth), virus monitor102 can also be selected to operate in the inline mode when a full analysis and screening of data packets is required. Although, the inline mode somewhat reduces bandwidth by forcing almost all data packets throughvirus monitor102, a virus outbreak can be effectively contained since any infected data packet is not returned to network100 (as is in the case of standby mode), but is rather isolated fromnetwork100 for analysis and ultimate discarding at thegarbage collection unit1010. Accordingly,FIG. 11 shows a virus monitor1100 arranged to operate in the inline mode in accordance with an embodiment of the invention. When operating in the inline mode, the virus monitor1100 has disabled the datapacket copier unit902 in such a way that all data packets are sent directly to thesecurity module808. In those cases where thesecurity module808 has determined that a particular data packet is not one that would likely be affected by a particular virus, that data packet is passed back tonetwork100 by way of thenetwork interface804. On the other hand, those data packets deemed more likely to be affected are passed directly to thefile scan module814. Those data packets found to be virus free are sent back to the network traffic flow by way of thenetwork interface804, whereas those data packets found to be infected by a virus are retained for further analysis, garbage collection, quarantine, etc. In some cases, the file scan unit1104 will issue avirus report1106 indicative of the virus analysis performed. Typically, thereport1106 will be forwarded to thecontroller126 for further analysis and correlation with other reports generated by other virus monitors throughoutnetwork100.
FLOWCHART EMBODIMENTS
• The methodology of the invention will now be described in terms of a number of flowcharts each describing a particular process for enabling the invention. Specifically,FIGS. 12-18 describe a number of interrelated processes that when used singly or in any combination described aspects of the invention.
• FIG. 12 shows a flowchart detailing a process1200 for monitoring a network for a virus in accordance with an embodiment of the invention. The process1200 begins at1202 by connecting a virus monitor(s) to the network at any number of appropriate locations. Typically, these locations are those that provide the virus monitor with greatest visibility to any computing devices coupled thereto. For example, coupling the virus monitor between atier2 and atier3 switch typically provides a good vantage point for subsequent virus monitoring. Once the virus monitor has been connected to the network, the virus monitor self-registers at1204. Such self registration includes identifying a location in the network (such as an IP address) associated with the virus monitor and/or identifying a location or locations of nearest controllers. Once a controller has been located, the virus monitor links itself to the located controller at1206 at which point the virus monitor invokes an initialization procedure at1208. Typically, the initialization procedure includes downloading a rules set and outbreak protection policy (OPP) file from a server computer by way of the controller.
• Once the initialization procedure has completed, the virus monitor monitors the network traffic (i.e., typically a flow of data packets) for any of a number and kind of network viruses at1210. If a virus is detected at1212, then at1214, the detected virus is identified or otherwise analyzed to provide an identity (if heretofore unknown virus). Once the virus has been analyzed and identified, the controller is notified of the identity of the virus at1216. At this point, the controller performs any of a number of operations substantially simultaneously. Such operations include resetting the virus monitor at1218 to provide a filtering type monitoring such that no data packet determined to be infected by the virus is passed back to the network traffic flow. In the described embodiment, this particular mode is referred to as inline mode as opposed to the as originally initialized standby mode in which data packets are copied for analysis thereby allowing the original data packets to be returned to the network traffic flow.
• In addition to setting the virus monitor from the stand by mode to the inline mode, the controller identifies those client devices affected or most likely to be affected by the virus at1220 and in those cases where a high level of security is necessary, the controller correlates all virus notifications for all virus monitors at1222. Based upon the identification of the affected or possibly affected client devices and the correlation of the received virus notices, the controller quarantines those client devices determined to be affected and likely to be threatened at1224. At1226, a determination is made whether or not the virus can be auto cured. By auto cured it is meant that the virus is eliminated from the affected computers and any damage is repaired automatically. In some cases, all client devices are inoculated against further infection by the virus or related viruses. If it is determined that the virus can not be auto cured, then the affected client devices are manually cleaned at1228, by for example, rebooting the computer system in the case of a computer worm, or in a worst case scenario, re-formatting the hard drives of the affected computers.
• If on the other hand, it is determined that the affected computers can be auto cured, then at1230 the affected computers are automatically cleaned thereby ending the process1200. It should be noted that in the unlikely event that the computer(s) could not be cured of the virus in a timely and cost effective manner, the computers so affected will be disconnected from the network in order to preserver the integrity of the remaining interconnected devices.
• In those situations where a visitor client device is connected to the monitored network, the visitor client device is introduced to the network using aprocess1300 detailed in a flowchart shown inFIG. 13 in accordance with an embodiment of the invention. Accordingly, theprocess1300 begins at1302 by the visitor client device being connected to a visitor port included in a portion of the network being monitored. At1304, a determination is made whether or not the visitor client device complies with the latest network anti-virus policies including acceptable anti-virus software. If it is determined that the visitor client device does comply, then the compliant visitor client device is granted a temporary use token at1306 that provides network access for a limited amount of time as spelled out in the terms and conditions of the use token. Such terms and conditions can include various authority levels, security levels, and the like. Typically, the use token is for a continuous period of time and not cumulative thereby limiting the loss of availability for the visitor port to only that period of time.
• If, on the other hand, the visitor client device is determined to be non-compliant, then at1308 the visitor client device is scanned for any active or latent virus infections. If the scanned visitor client device passes the virus scan at1310, then the visitor client device is granted the use token at1306, otherwise, a determination is made at1312 whether or not connecting the visitor client device to the network is to continue. If the connection is to end, then theprocess1300 is ended without the connecting, however, if the connection is to continue, then at1314, the infected visitor client device is cleaned and any damage repaired after which control is passed to1306 where the use token is granted to the visitor client device.
• Once the use token has been granted, access to the network is granted at1316 during which periodic checks of the validity of the use token are made at1318. In those cases where the use token has been determined to not be valid, then at1320 a determination is made whether or not a new use token is requested. If a new use token is requested, then control is passed back to1308 for scanning of the visitor client device to assure that no viruses have infected the requesting device, otherwise, theprocess1300 ends normally.
• In some cases, a new client device is permanently added to the network (as opposed to the temporary addition of a visitor client device) in which case aprocess1400 shown inFIG. 14 is followed to assure compliance to anti-virus security policy. Accordingly, theprocess1400 begins at1402 by identifying a new client device to be added to the network. Identification typically includes identifying the type of system, resident operating system, installed anti-virus software (if any), network address (such as IP address), and the like. Once the new client device has been properly identified, then at1404 a determination is made whether or not a proper set of anti-virus policies and protocols are in place. Such anti-virus policies and protocols include proper anti-virus software, filters, etc. Such appropriate anti-virus software can include any recognized anti-virus software from any number of recognized vendors such as Trend Micro of Cupertino, Calif., and the like.
• If it is determined that the new client device does have the proper anti-virus policies and protocols in place, then at1406, selected device information is retrieved. Such information can be related to the identification of the client device as well as any other appropriate information pertinent to the network connection and the new client device is connected to the network at1408.
• If, on the other hand, it has been determined that the proper anti-virus policies and protocols are not in place, then at1410, all access to all addresses except to that of an anti-virus software installation server are blocked. Once access to the anti-virus software installation server is granted at1412, then a determination at1414 is made whether or not the appropriate anti-virus software is to be downloaded to the new visitor device. If it is determined that the anti-virus software is not to be downloaded, then theprocess1400 ends without the new client device being connected to the network. Otherwise, control is passed to1406 and finally to1408 where the new client device is connected to the network.
• Once all client devices are connected to the network, any virus attacks can be limited in scope by instituting a defensive measure referred to a network segment isolation whereby the affected portion of the network is logically isolated from the unaffected portion of the network. More specifically, aprocess1500 shown inFIG. 15 describes in detail the network isolation process in accordance with an embodiment of the invention. Accordingly, theprocess1500 begins at1502 where a controller (or a system administrator) correlates various reports of virus attacks. When a number of reports have been correlated, the controller determines at1504 whether or not a virus outbreak is confirmed. If a virus outbreak is confirmed, then the controller isolates the affected client devices (typically by way of network nodes to which the client devices are connected either physically or logically) at1506. Once the outbreak has been confirmed, then the controller signals a virus monitor to switch operating mode to inline mode such that all data packets that constitute the network traffic are checked for the virus and related viruses at1508. Once the virus is identified at1510, the controller instructs the virus monitor to block only those data packets infected by that particular virus and related viruses at1512.
• Once the virus has been identified, an anti-virus (or curing agent) is created by way of an anti-virus creatingprocess1600 in accordance with an embodiment of the invention. Accordingly, theprocess1600 shown inFIG. 16 begins at1602 by the identified virus being parsed. By parsed it is meant that the various components of the various are identified. Such components include a infection module, a detection module, and a payload module. Once the various components of the virus have been identified, the components are analyzed. More specifically, the detection module is analyzed at1604, the infection module at1606, and the payload module at1608. By analyzed, it is meant that the various virus components are studied to determine the method of infection in the case of the infection module and the deleterious effects of the virus payload portion can have on the infected systems.
• It should be noted that the analysis can be performed concurrently or in any appropriate order. Once the various components have been analyzed, the infection module is modified to infect all computers already infected (regardless of the fact that a particular computer may already be infected by the parent virus) by the virus at1610 and the payload module is modified to “lock the door” (i.e., prevent the virus from re-infecting the computer, or inoculating it from future virus attacks) at1612. Once the various modules have been modified, the detection module (unmodified), the modified infection module, and the modified payload module are combined at1614 to create the anti-virus at1616. It should be noted that in some cases, the modified payload module provides a damage repair protocol that repairs the damage caused by the original virus.
• FIG. 17 shows a flowchart detailing aprocess1700 for performing an automatic clean/cure of a group of infected computers in accordance with an embodiment of the invention. Accordingly, theprocess1700 begins at1702 by the acquiring of the identity of a particular virus found to be infecting a number of computers in a network. Once the identity has been acquired, at1704, an associated anti-virus for the identified virus is received after which at1706 the infected computers are identified. By identified it is meant that not only a physical location is known, but a logical location as well as any other pertinent information related to the computer. Once the location has been identified, a determination is made whether or not a cleaning tool has been installed in the identified computers at1708. By cleaning tool it is meant a tool that uses the anti-virus generated for the particular virus to automatically clean/cure the infected computer. If a cleaning tool is not found, then all traffic to/from the infected computers are re-routed to a cleaning tool server at1710 where at1712 the appropriate cleaning tool is downloaded.
• Once the appropriate cleaning tool has been downloaded, the infected computers are cleaned at1714. Returning to1708, if the appropriate cleaning tools are found, then the cleaning tool is sent to all infected computers in the network at1716 such that at the cleaning commences at1714. After the cleaning is completed, selected ones of the heretofore unaffected computers are inoculated against a future virus attack at1718.
• Once the anti-virus has been created, the anti-virus can be used to automatically clean any affected computers as well as inoculate unaffected computers thereby preventing any future virus outbreaks. The automatic cure/clean process1800 is detailed in a flowchart illustrated inFIG. 18 where at1802 the anti-virus detects a computer infected with the associated virus. Once the anti-virus has detected the affected computer, the anti-virus infects the computer already infected by the parent virus a1804 while at1806 the anti-virus overwrites the parent virus in the affected computer with the anti-virus payload. In some embodiments, the anti-virus payload results in inoculating the affected computer from future attacks by “locking the door” to the parent virus. Once the anti-virus payload has been downloaded, the anti-virus payload repairs any damage caused by the parent virus at1808. Once the affected computers have been both inoculated and any damage repaired, the anti-virus inoculates unaffected computers at1810. It should be noted that in the case of unaffected computers, since there is no damage caused by the parent virus, the anti-virus payload in not active and thereby remains dormant.
• FIG. 19 shows a system block diagram of acomputer system801 which may be to implement the computing system300 used to execute functions of the present invention including the scanning, deletion, modification, and quarantine of data packets suspected of harboring computer worms. Thecomputer system801 includes display monitor803 andkeyboard809, andmouse811.Computer system801 further includes subsystems such as acentral processor851,system memory853, fixed storage855 (e.g., hard drive), removable storage857 (e.g., CD-ROM drive),display adapter859,sound card861,speakers863, andnetwork interface865. Thecentral processor851, may execute computer program code (e.g., an operating system) to implement the various aspects of the scan engine of the present invention as described herein. The operating system is normally, but not necessarily, resident in thesystem memory853 during its execution. Other computer systems suitable for use with the invention may include additional subsystems or fewer subsystems. For example, another computer system could include more than one processor851 (i.e., a multi-processor system) or one or more levels of cache memory.
• The system bus architecture ofcomputer system801 is represented byarrows867. However, these arrows are illustrative of any interconnection scheme serving to link the subsystems. For example, a local bus could be utilized to connect the central processor to the system memory and display adapter.Computer system801 shown inFIG. 6 is but an example of a computer system suitable for use with the present invention. Other computer architectures having different configurations of subsystems may also be utilized.
• Although a virus filter has been described, it can be appreciated that those skilled in the are often interchange terminology with regard to malicious code. Thus, the term “computer worm” can include any type of malicious or otherwise undesirable or unexpected computer code that presents itself at a computer, whether the code is actually referred to as a “virus”, “worm”, “Trojan horse”, and the like. Further, although the foregoing invention has been described in some detail for purposes of clarity of understanding, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. Therefore, the described embodiments should be taken as illustrative and not restrictive, and the invention should not be limited to the details given herein but should be defined by the following claims and their full scope of equivalents.

Claims (33)

1. In a distributed network of interconnected computing devices, a network virus/worm monitor, comprising:

a network virus/worm sensor; and

a traffic controller in communication with the network virus/monitor sensor and the network operable in a number of modes wherein in a first mode the bandwidth of the network is substantially unaffected by the traffic controller during a virus/worm sensing operation by the network virus/worm sensor wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic, the virus/worm sensor causes the traffic controller to switch to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network.

2. A monitor as recited inclaim 1, wherein those data packets deemed to be infected by the identified computer virus or computer worm are forwarded to a virus/worm analyzer unit coupled to the network computer virus/worm sensor.

3. A monitor as recited inclaim 2, wherein in the first mode, substantially all data packets included in the network traffic are copied by the traffic controller.

4. A monitor as recited inclaim 3, wherein substantially all of the copied data packets are forwarded to the virus/worm analyzer unit.

5. A monitor as recited inclaim 4, at the virus/worm analyzer unit, the copied data packet is forwarded to a packet protocol determinator that determines if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

6. A monitor as recited inclaim 5, further comprising:

a trash collector arranged to receive those copied data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm;

a filescan unit arranged to receive and analyze those copied data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm.

7. A monitor as recited inclaim 5, further comprising:

a virus/worm analyzer unit arranged to determine if those copied data packets received at the filescan unit are infected by the detected computer virus or computer worm wherein those packets determined not to be infected are forwarded to the trash collector;

a virus analysis unit arranged to analyze the infected copied data packets; and

a virus report module arranged to generate a virus report based upon the analysis.

8. A monitor as recited inclaim 2, wherein in the second mode, only those original data packets included in the network traffic suspected of being infected by the detected computer virus or computer worm are forwarded by the traffic controller to the virus/worm analyzer unit.

9. A monitor as recited inclaim 8, at the virus/worm analyzer unit, the original data packet is forwarded to a packet protocol determinator that determines if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

10. A monitor as recited inclaim 9, further comprising:

a network interface arranged to return to the network traffic only those original data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm, wherein the filescan unit receives and analyzes those original data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm.

11. A monitor as recited inclaim 10, wherein the virus/worm analyzer unit determines if those original data packets received at the filescan unit are infected by the detected computer virus or computer worm wherein those packets determined not to be infected are forwarded to the network interface for return to the network traffic and wherein the virus analysis unit analyzes the infected original data packets and the virus report module generates a virus report based upon the analysis.

12. A monitor as recited inclaim 1, wherein the first mode is an inline mode and wherein the second mode is a standby mode.

13. In a distributed network of interconnected computing devices, a method of managing network traffic by a network virus/worm monitor, having a network virus/worm sensor, comprising:

during a virus/worm sensing operation, the bandwidth of the network is substantially unaffected by the traffic controller by the network virus/worm sensor in a first mode wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic; and

switching when a computer virus or computer worm is detected to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network.

14. A method as recited inclaim 13, further comprising:

copying substantially all data packets.

15. A method as recited inclaim 14, wherein in the first mode, copying by the traffic controller substantially all data packets included in the network traffic; and

forwarding the copied data packets to the virus/worm analyzer unit.

16. A method as recited inclaim 15, comprising:

forwarding the copied data packet to a packet protocol determinator; and

determining if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

17. A method as recited inclaim 16, further comprising:

receiving at a trash collector those copied data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm; and

receiving and analyzing those copied data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm at a filescan unit.

18. A method as recited inclaim 17, further comprising:

determining by a virus/worm analyzer unit if those copied data packets received at the filescan unit are infected by the detected computer virus or computer worm;

forwarding those packets determined not to be infected to the trash collector;

analyzing the infected copied data packets; and

generating a virus report based upon the analysis.

19. A method as recited inclaim 14 comprising:

wherein in the second mode,

forwarding only those original data packets included in the network traffic suspected of being infected by the detected computer virus or computer worm are to the virus/worm analyzer unit.

20. A method as recited inclaim 19, at the virus/worm analyzer unit, forwarding the original data packet to a packet protocol determinator; and

determining if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

21. A method as recited inclaim 20, further comprising:

returning to the network traffic only those original data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm; and

analyzing those original data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm.

22. A method as recited inclaim 21 comprising:

determining if those original data packets received at the filescan unit are infected by the detected computer virus or computer worm forwarding those packets determined not to be infected are to the network interface for return to the network traffic; and

generating a virus report based upon the analysis.

23. A monitor as recited inclaim 13, wherein the first mode is an inline mode and wherein the second mode is a standby mode.

24. Computer program product for managing network traffic by a network virus/worm monitor, having a network virus/worm sensor, comprising:

computer code for performing a computer virus/worm sensing operation, the bandwidth of the network is substantially unaffected by the traffic controller by the network virus/worm sensor in a first mode wherein when the network virus/worm sensor detects a computer virus or a computer worm in network traffic;

computer code for switching when a computer virus or computer worm is detected to a second mode such that only those data packets infected by the detected computer virus or computer worm are not returned to the network; and

computer readable medium for storing the computer code.

25. Computer program product as recited inclaim 24, further comprising:

computer code for copying by the traffic controller substantially all data packets included in the network traffic.

26. Computer program product as recited inclaim 25, wherein in the first mode,

computer code for forwarding the copied data packets to the virus/worm analyzer unit.

27. Computer program product as recited inclaim 26, comprising:

computer code for forwarding the copied data packet to a packet protocol determinator; and

computer code for determining if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

28. Computer program product as recited inclaim 27, further comprising:

computer code for receiving at a trash collector those copied data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm; and

computer code for receiving and analyzing those copied data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm at a filescan unit.

29. Computer program product as recited inclaim 28, further comprising:

computer code for determining by a virus/worm analyzer unit if those copied data packets received at the filescan unit are infected by the detected computer virus or computer worm;

computer code for forwarding those packets determined not to be infected to the trash collector;

computer code for analyzing the infected copied data packets; and

computer code for generating a virus report based upon the analysis.

30. Computer program product as recited inclaim 29 comprising:

wherein in the second mode,

computer code for forwarding only those original data packets included in the network traffic suspected of being infected by the detected computer virus or computer worm are to the virus/worm analyzer unit.

31. Computer program product as recited inclaim 30, at the virus/worm analyzer unit,

computer code for forwarding the original data packet to a packet protocol determinator; and

computer code for determining if the packet protocol of the copied data packet is one likely to be infected by the detected computer virus or computer worm.

32. Computer program product as recited inclaim 31, further comprising:

computer code for returning to the network traffic only those original data packets determined to be of a protocol not likely to be infected by the detected computer virus or computer worm; and

computer code for analyzing those original data packets determined to be of a protocol likely to be infected by the detected computer virus or computer worm.

33. Computer program product as recited inclaim 32 comprising:

computer code for determining if those original data packets received at the filescan unit are infected by the detected computer virus or computer worm forwarding those packets determined not to be infected are to the network interface for return to the network traffic; and

computer code for generating a virus report based upon the analysis.

Images