Real-time prevention of malicious content via dynamic analysis
This disclosure is related to methods and apparatus used to for preventing malicious content from reaching a destination via a dynamic analysis engine may operate in real-time when packetized data is received. Data packets sent from a source computer may be received and be forwarded to an analysis computer that may monitor actions performed by executable program code included within the set of data packets when making determinations regarding whether the data packet set should be classified as malware. In certain instances all but a last data packet of the data packet set may also be sent to the destination computer while the analysis computer executes and monitors the program code included in the data packet set. In instances when the analysis computer identifies that the data packet set does include malware, the malware may be blocked from reaching the destination computer by not sending the last data packet to the destination computer.
1. A method for detecting malicious content, the method comprising:
receiving data at a sandbox device, wherein the data is received from a separate firewall device after the separate firewall device receives the data from a sender device, a first portion of the data being sent to a destination device while holding a second portion of the data at the firewall device until at least after an observation of the data at the sandbox device;
observing that a first action is performed when instructions included in the received data are executed at the sandbox device;
identifying that the first action is suspicious and includes reorganizing at least a portion of the received data;
observing an additional action performed when instructions included in the reorganized data portion are executed at the sandbox device, wherein the additional action includes accessing an inappropriate data storage location;
identifying that the additional action is classified as malicious; and
performing a second action based on the additional action being classified as malicious, wherein the second action occurs after the observation of the data at the sandbox device.
2. The method of claim 1 , further comprising sending a message to the firewall device identifying that the received data is malicious, wherein the firewall device drops the second portion of the data based on the message indicating that the received data is malicious.
3. The method of claim 1 , further comprising:
identifying an attribute associated with the received data; and
storing the attribute associated with the received data.
4. The method of claim 1 , further comprising:
generating a signature from the received data; and
storing the signature at a deep packet inspection data store.
5. The method of claim 1 , wherein the first action includes intercepting a basic input/output (BIOS) instruction.
6. The method of claim 1 , wherein the first action includes accessing an inappropriate data storage location.
7. The method of claim 1 , wherein the first action includes de-obfuscating the additional instructions included in the received data.
8. The method of claim 1 , wherein the additional action or the second action includes preparing to transmit data from the sandbox device.
9. The method of claim 1 , further comprising:
receiving a second set of data at the sandbox device, wherein the second set of data is received from the separate firewall device after the separate firewall device receives the second set of data;
observing one or more actions performed when instructions included in the second set of received data are executed; and
performing a deep packet inspection (DPI) scan on the second set of data while observing the one or more actions performed when the instructions included in the second set of received data are executed.
10. The method of claim 9 , wherein the one or more actions are observed while the DPI scan is performed based on the sandbox device being a multi-processor platform.
11. The method of claim 9 , further comprising:
identifying based on the observing of the one or more performed actions that the second set of data includes malicious instructions;
storing a signature generated from the second set of data;
generating a second signature from a third set of received data;
identifying that the signature matches the second signature; and
identifying that the third set of data includes the malicious instructions based on the signature matching the second signature.
12. A non-transitory computer-readable storage medium having embodied thereon a program executable by a processor for implementing a method for detecting malicious content, the method comprising:
receiving data at a sandbox device, wherein the data is received from a separate firewall device after the separate firewall device receives the data from a sender device, a first portion of the data being sent to a destination device while holding a second portion of the data at the firewall device until at least after an observation of the data at the sandbox device;
observing that a first action is performed when instructions included in the received data are executed at the sandbox device;
identifying that the first action is suspicious and includes reorganizing at least a portion of the received data;
observing an additional action performed when instructions included in the reorganized data portion are executed at the sandbox device, wherein the additional action includes accessing an inappropriate data storage location;
identifying that the additional action is classified as malicious; and
performing a second action based on the additional action being classified as malicious, wherein the second action occurs after the observation of the data at the sandbox device.
13. The non-transitory computer-readable storage medium of claim 12 , the program is further executable to send a message to the firewall device identifying that the received data is malicious, wherein the firewall device drops the second portion of the data based on the message indicating that the received data is malicious.
14. The method of claim 12 , further comprising:
identifying an attribute associated with the received data; and
storing the attribute associated with the received data.
15. The method of claim 12 , further comprising:
generating a signature from the received data; and
storing the signature at a deep packet inspection data store.
16. The non-transitory computer-readable storage medium of claim 12 , wherein the first action includes intercepting a basic input/output (BIOS) instruction.
17. The non-transitory computer-readable storage medium of claim 12 , wherein the first action includes accessing an inappropriate data storage location.
18. A system for detecting malicious content, the system comprising:
a firewall device that:
receives a data set based on information received from a destination device;
sends the data set for analysis,
sends a first portion of the data set to the destination device, and
holds a second portion of the data set at the firewall device without immediately sending the second portion of the data set to the destination device; and
a sandbox device that is separate from the firewall device, wherein the sandbox device:
receives the data set from the firewall device,
performs the analysis,
observes that a first action is performed when instructions included in the first data set are executed based on the analysis,
identifies that the first action is suspicious and includes reorganizing at least a portion of the received data,
observes that an additional action is performed when instructions included in the reorganized data portion are executed at the sandbox device, wherein the additional action includes accessing an inappropriate data storage location,
identifies that the additional action is classified as malicious, and
performs a second action based on the additional action being classified as malicious, wherein the second action occurs after the observation of the data at the sandbox device.
19. The system of claim 18 , wherein the sandbox device further:
identifies an attribute associated with the received data; and
stores the attribute associated with the received data.
20. The system of claim 18 , wherein the sandbox device further:
generates a signature from the received data; and
stores the signature at a deep packet inspection data store.
21. The system of claim 18 , wherein the first action includes intercepting a basic input/output (BIOS) instruction.