Secure digital traffic analysis
A network monitoring “sensor” is built on initial startup by checking the integrity of the bootstrap system and, if it passes, downloading information from which it builds the full system including an encrypted and an unencrypted portion. Later, the sensor sends hashes of files, configurations, and other local information to a data center, which compares the hashes to hashes of known-good versions. If they match, the data center returns information (e.g., a key) that the sensor can use to access the encrypted storage. If they don't, the data center returns information to help remediate the problem, a command to restore some or all of the sensor's programming and data, or a command to wipe the encrypted storage. The encrypted storage stores algorithms and other data for processing information captured from a network, plus the captured/processed data itself.
1. A network monitoring system comprising a processor, a network interface, and a memory having an encrypted portion and an unencrypted portion, the unencrypted portion being encoded with programming instructions executable by the processor to:
create one or more hashes of at least some information stored in the unencrypted portion of the memory;
send the one or more hashes to a remote computer;
receive a response from the remote computer; and
if the response is of a first type, then
use information in the response to decrypt the encrypted portion of the memory; and
use information from the encrypted portion of the memory to capture and process data using the network interface.
2. The network monitoring system of claim 1 , wherein the programming instructions are further executable by the processor to, if the response is of a second type, erase the encrypted portion of the memory.
3. The network monitoring system of claim 1 , wherein the programming instructions are further executable by the processor to flag dangerous content using filtering information in the encrypted portion of the memory.
4. The network monitoring system of claim 3 , wherein the filtering information used to flag the dangerous content comprises at least one profile selected from a profile set consisting of:
a client profile comprising information that characterizes at least one of expected content and unexpected content for a user of the network monitoring system; and
an industry profile comprising information that characterizes at least one of expected content and unexpected content for a generic user who operates in a particular industry, where the user of the network monitoring system operates in the particular industry.
5. The network monitoring system of claim 3 , wherein the programming instructions are further executable by the processor to:
open flagged content in a sandbox;
collect intelligence regarding behavior of the flagged content as it is opened in the sandbox; and
store the intelligence in the encrypted portion of the memory.