4 Script that takes a file (output by wireshark/tshark, in JSON format) and analyze
5 the traffic frequency of a certain device at a certain time.
10 from collections import defaultdict
11 from dateutil import parser
13 JSON_KEY_SOURCE = "_source"
14 JSON_KEY_LAYERS = "layers"
17 JSON_KEY_ETH_DST = "eth.dst"
18 JSON_KEY_ETH_SRC = "eth.src"
19 JSON_KEY_FRAME = "frame"
20 JSON_KEY_FRAME_TIME = "frame.time"
21 TABLE_HEADER_X = "Timestamp (hh:mm:ss)"
22 TABLE_HEADER_Y = "Packet frequency (pps)"
25 def save_to_file(tbl_header, dictionary, filename_out):
26 """ Show summary of statistics of PCAP file
28 tbl_header: header for the saved table
29 dictionary: dictionary to be saved
30 filename_out: file name to save
32 # Appending, not overwriting!
33 f = open(filename_out, 'a')
34 # Write the table header
35 f.write("# " + TABLE_HEADER_X + " " + TABLE_HEADER_Y + "\n");
36 # Iterate over dictionary and write (key, value) pairs
37 for key in sorted(dictionary):
39 #f.write(str(key) + ", " + str(dictionary[key]) + "\n")
41 f.write(str(key) + " " + str(dictionary[key]) + "\n")
44 print "Writing output to file: ", filename_out
51 print "Usage: python", sys.argv[0], "<input_file> <output_file> <device_name> <mac_address>"
53 # Parse the file for the specified MAC address
54 time_freq = parse_json(sys.argv[1], sys.argv[4])
55 # Write statistics into file
56 save_to_file(sys.argv[3], time_freq, sys.argv[2])
57 print "====================================================================="
58 #for time in time_freq.keys():
59 for key in sorted(time_freq):
60 print key, " => ", time_freq[key]
61 print "====================================================================="
64 # Convert JSON file containing DNS traffic to a map in which a hostname points to its set of associated IPs.
65 def parse_json(file_path, mac_address):
66 """ Show summary of statistics of PCAP file
68 file_path: path of the read file
69 mac_address: MAC address of a device to analyze
71 # Maps timestamps to frequencies of packets
73 with open(file_path) as jf:
75 # data becomes reference to root JSON object (or in our case json array)
77 # Loop through json objects in data
78 # Each entry is a pcap entry (request/response (packet) and associated metadata)
80 # p is a JSON object, not an index
81 layers = p[JSON_KEY_SOURCE][JSON_KEY_LAYERS]
83 frame = layers.get(JSON_KEY_FRAME, None)
84 date_time = frame.get(JSON_KEY_FRAME_TIME, None)
85 # Get into the Ethernet address part
86 eth = layers.get(JSON_KEY_ETH, None)
87 # Skip any non DNS traffic
89 print "[ WARNING: Packet has no ethernet address! ]"
91 # Get source and destination MAC addresses
92 src = eth.get(JSON_KEY_ETH_SRC, None)
93 dst = eth.get(JSON_KEY_ETH_DST, None)
94 # Get just the time part
95 date_time_obj = parser.parse(date_time)
96 # Remove the microsecond part
97 time_str = str(date_time_obj.time())[:8]
98 print str(time_str) + " - src:" + str(src) + " - dest:" + str(dst)
99 # Get and count the traffic for the specified MAC address
100 if src == mac_address or dst == mac_address:
101 # Check if timestamp already exists in the map
102 # If yes, then just increment the frequency value...
103 if time_str in time_freq:
104 time_freq[time_str] = time_freq[time_str] + 1
105 else: # If not, then put the value one there
106 time_freq[time_str] = 1
109 if __name__ == '__main__':