How to parse in linux sniffer results on the fly? - linux

Question

I want to sort and calculate how much clients downloaded files (3 types) from my server.
I installed tshark and ran followed command that should capture GET requests:
`./tshark 'tcp port 80 and (((ip[2:2] - ((ip[0]&0xf)<<2)) - ((tcp[12]&0xf0)>>2)) != 0)' -R'http.request.method == "GET"'`
so sniffer starts to work and every second I get new row, here is a result:
0.000000 144.137.136.253 -> 192.168.4.7 HTTP GET /pids/QE13_593706_0.bin HTTP/1.1
8.330354 1.1.1.1 -> 2.2.2.2 HTTP GET /pids/QE13_302506_0.bin HTTP/1.1
17.231572 1.1.1.2 -> 2.2.2.2 HTTP GET /pids/QE13_382506_0.bin HTTP/1.0
18.906712 1.1.1.3 -> 2.2.2.2 HTTP GET /pids/QE13_182406_0.bin HTTP/1.1
19.485199 1.1.1.4 -> 2.2.2.2 HTTP GET /pids/QE13_302006_0.bin HTTP/1.1
21.618113 1.1.1.5 -> 2.2.2.2 HTTP GET /pids/QE13_312106_0.bin HTTP/1.1
30.951197 1.1.1.6 -> 2.2.2.2 HTTP GET /nginx_status HTTP/1.1
31.056364 1.1.1.7 -> 2.2.2.2 HTTP GET /nginx_status HTTP/1.1
37.578005 1.1.1.8 -> 2.2.2.2 HTTP GET /pids/QE13_332006_0.bin HTTP/1.1
40.132006 1.1.1.9 -> 2.2.2.2 HTTP GET /pids/PE_332006.bin HTTP/1.1
40.407742 1.1.2.1 -> 2.2.2.2 HTTP GET /pids/QE13_452906_0.bin HTTP/1.1
what I need to do to store results type and count like /pids/*****.bin in to other file.
Im not strong in linux but sure it can be done with 1-3 rows of script.
Maybe with awk but I don't know what is the technique to read result of sniffer.
Thank you,

Answer 1

Can't you just grep the log file of your webserver?
Anyway, to extract the lines of captured http traffic relative to your server files, just try with
./tshark 'tcp port 80 and \
(((ip[2:2] - ((ip[0]&0xf)<<2)) - ((tcp[12]&0xf0)>>2)) != 0)' \
-R'http.request.method == "GET"' | \
egrep "HTTP GET /pids/.*.bin"