Created
October 5, 2013 06:20
-
-
Save oinume/6837358 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
# -*- coding: utf-8 -*- | |
# Usage: tail -10000 <access_log> | ./parse-nginx-log.py | |
import os | |
import re | |
import sys | |
class LogParser(object): | |
#10.32.202.174 - gm [27/Sep/2013:16:19:59 +0900] "GET /giftbox/index HTTP/1.1" 200 8012 "-" "Mozilla/5.0 (Linux; U; Android 2.1-update1; ja- jp; SonyEricssonSO-01B Build/2.0.1.B.0.19) AppleWebKit/530.17 (KHTML, like Gecko) Version/ 4.0 Mobile Safari/530.17" "-" 4.010 4.010 | |
# '$remote_addr - $remote_user [$time_local] "$request" ' | |
# '$status $body_bytes_sent "$http_referer" ' | |
# '"$http_user_agent" "$http_x_forwarded_for" $request_time $upstream_response_time'; | |
LOG_PATTERN = re.compile(r"""^(?P<remote_host>[0-9]{,3}\.[0-9]{,3}\.[0-9]{,3}\.[0-9]{,3}) (?P<ident>[^ ]{1,}) (?P<remote_user>[^ ]{1,}|\-) \[(?P<datetime>[0-9]{2}\/[A-Za-z]{3}\/[0-9]{1,4}:[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2} [+\-][0-9]{4})\] "(?P<method>[A-Z ]+) (?P<uri>[^"]*) (?P<protocol>[^"]*)" (?P<status>[0-9]{3}) (?P<bytes>[0-9]{1,}|\-) "(?P<referer>[^"]*|\-)" "(?P<user_agent>[^"]+)" "(?P<forwarded_for>[^"]*|\-)" (?P<elapsed>[\d\.]+)$""") | |
def __init__(self): | |
self.summary = { | |
'total_requests': 0, | |
'slow_requests': 0, | |
} | |
self.slow_request_summary = {} | |
def parse(self, file): | |
for line in file: | |
m = self.LOG_PATTERN.match(line) | |
if not m: | |
continue | |
request_info = m.groupdict() | |
#print str(request_info) | |
self.summarize(request_info) | |
print """\ | |
Total requests: %d | |
Slow requests : %d (>= 1000ms) | |
""" % (self.summary['total_requests'], self.summary['slow_requests']) | |
print "%-40s %-10s %-10s %8s" % ('URL', 'count', 'average(ms)', '%') | |
def s(x): | |
return x['total_time'] / x['count'] | |
total_time = 0 | |
for data in (self.slow_request_summary.values()): | |
total_time += data['total_time'] | |
for data in sorted(self.slow_request_summary.values(), key=s, reverse=True): | |
print "%-40s %10d %10d %10.2f" % ( | |
data['uri'], data['count'], data['total_time'] / data['count'] / 1000, | |
1.0 * data['total_time'] / total_time * 100, | |
) | |
def summarize(self, req): | |
s = self.summary | |
s['total_requests'] = s['total_requests'] + 1 | |
uri = req['uri'] | |
normalized_uri = re.sub(r'[\d]+', '', uri) | |
normalized_uri = re.sub(r'\?(.+)', '', normalized_uri) | |
elapsed = float(req['elapsed']) | |
elapsed = int(elapsed * 1000 * 1000) | |
if elapsed < 1000 * 1000: | |
return | |
s['slow_requests'] = s['slow_requests'] + 1 | |
summary = self.slow_request_summary.get(normalized_uri) | |
if summary: | |
summary['count'] = summary['count'] + 1 | |
summary['total_time'] = summary['total_time'] + elapsed | |
else: | |
self.slow_request_summary[normalized_uri] = { | |
'uri': normalized_uri, | |
'count': 1, | |
'total_time': elapsed | |
} | |
if __name__ == '__main__': | |
file = None | |
if len(sys.argv) == 1: | |
file = sys.stdin | |
else: | |
if os.path.exists(sys.argv[1]): | |
file = open(sys.argv[1], 'r') | |
else: | |
print >>sys.stderr, "error" | |
sys.exit(1) | |
LogParser().parse(file) | |
file.close() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment