Improve dummy log generation script

gen_log.sh

@@ -1,9 +1,11 @@
 #!/bin/bash
+LANG=en_US.UTF-8
+now=$(date "+%d/%b/%Y:%H:%M:%S %z")
 
-
-for i in {1..34000}; do
-	n=$( echo "$i % 256" | bc)
-	echo "10.10.100.$n - - [10/Sep/2021:23:17:55 +0200] \"GET /mentions-legales/ HTTP/2.0\" 301 178 \"https://saxodwarf.fr/tags/h%C3%A9bergement/\" \"Mozilla/5.0 (X11; Linux aarch64; rv:91.0) Gecko/20100101 Firefox/91.0 $i\""
+max=$(( 200000 - "$(wc -l 'mucho_log_today.log' | cut -d ' ' -f 1)"))
+for (( i=0; i < $max; i++)); do
+	n=$(( $i % 256))
+	echo "10.10.100.$n - - [$now] \"GET /mentions-legales/ HTTP/2.0\" 301 178 \"https://saxodwarf.fr/tags/h%C3%A9bergement/\" \"Mozilla/5.0 (X11; Linux aarch64; rv:91.0) Gecko/20100101 Firefox/91.0 $i\""
 	if (( $i % 1000 == 0 )); then
 		echo $i 1>&2
 	fi

get_page_stats.py

@@ -13,7 +13,7 @@ import argparse
 import subprocess
 import configparser
 from datetime import datetime
-from collections import defaultdict
+from collections import defaultdict, Counter
 from itertools import repeat
 from urllib.parse import urlparse
 from typing import Dict, List, Tuple, Set
@@ -22,7 +22,7 @@ import xml.etree.ElementTree as ET
 import requests
 
 VisitDict = Dict[str, Set[str]]
-
+MAX_UA_NB = 1000
 
 def parse_args()-> argparse.Namespace:
     """ Parse arguments of the script
@@ -88,7 +88,10 @@ class TelegrafExporter():
 
 
     def export_result_to_telegraf(self, page_hits: VisitDict,
-                                  bot_hits: VisitDict, timestamp: int) -> None:
+                                  bot_hits: VisitDict,
+                                  user_agents: VisitDict,
+                                  methods: Counter,
+                                  timestamp: int) -> None:
         """ Export the bot_hits and page_hits dictionnaries to telegraf
         """
         # export standard hits
@@ -120,6 +123,32 @@ class TelegrafExporter():
             except requests.exceptions.RequestException as excpt:
                 print(excpt)
                 sys.exit(1)
+        # export user agent variety
+        name="user_agent_variety"
+        for ua_type, uas in user_agents.items():
+            try:
+                response = self.telegraf_post(timestamp,
+                                              create_time,
+                                              name,
+                                              ua_type,
+                                              uas)
+                response.raise_for_status()
+            except requests.exceptions.RequestException as excpt:
+                print(excpt)
+                sys.exit(1)
+        # export method variety
+        name="method_variety"
+        for method, count in methods.items():
+            try:
+                response = self.telegraf_post(timestamp,
+                                              create_time,
+                                              name,
+                                              method,
+                                              count)
+                response.raise_for_status()
+            except requests.exceptions.RequestException as excpt:
+                print(excpt)
+                sys.exit(1)
 
 def get_crawler_patterns(exclude_crawler: bool) -> List[str]:
     """ Parse the crawler-user-agent file, and returns a list
@@ -174,7 +203,7 @@ def parse_logfile(logfile_path: str, locations: List[str],
     # pylint: disable=line-too-long
     # $remote_addr - $remote_user [$time_local] "$request" $status $body_bytes_sent "$http_referer" "$http_user_agent"
     log_line_template = (r'^(?P<ip_address>[0-9a-f.:]+) \- .+? \[(?P<time_local>.*)\] '
-                          '"GET (?P<location>{locations}) .*?" [0-9]{{3}} [0-9]+ ".+?" '
+                          '"(?P<method>[A-Z]+) (?P<location>{locations}) .*?" [0-9]{{3}} [0-9]+ ".+?" '
                           '"(?P<user_agent>.+)"$')
     #known_page_regex = re.compile(log_line_template.format(locations='|'.join(map(re.escape,
     #                                                                              locations))))
@@ -186,6 +215,7 @@ def parse_logfile(logfile_path: str, locations: List[str],
     other_visit_dict: VisitDict = defaultdict(set)
     bot_user_agents = set()
     client_user_agents = set()
+    method_counter = Counter()
     # The way to get the timezone data here is not great (not taking into account DST and such)
     # but it is a fallback default date that should hardly ever be used.
     last_log_date = datetime.now(datetime.now().astimezone().tzinfo).strftime(time_local_fmt)
@@ -202,12 +232,14 @@ def parse_logfile(logfile_path: str, locations: List[str],
                 location = match_obj.group("location")
                 last_log_date = match_obj.group("time_local")
                 user_agent = match_obj.group("user_agent")
-                if location in locations:
+                method = match_obj.group("method")
+                if method == "GET" and location in locations:
                     # For each line, if it is a GET on a known page, count it
-                    if (not user_agent in bot_user_agents and
-                        user_agent in client_user_agents or
+                    if ((not user_agent in bot_user_agents and
+                        (len(client_user_agents) + len(bot_user_agents)) < MAX_UA_NB) and
+                        (user_agent in client_user_agents or
                         not any(map(re.search, crawler_patterns,
-                                    repeat(user_agent)))):
+                                    repeat(user_agent))))):
                         visit_dict[location].add(client_ip)
                         client_user_agents.add(user_agent)
                     else:
@@ -217,6 +249,7 @@ def parse_logfile(logfile_path: str, locations: List[str],
                     # Also count lines that are NOT "GET on a known page" in a different dict.
                     # Those other hits can be static site ressources loaded,
                     # in which case we group the hits
+                    method_counter[method] += 1
                     if location.startswith("/isso/"):
                         other_visit_dict["/isso/*"].add(client_ip)
                     elif location.startswith("/assets/css/"):
@@ -234,7 +267,8 @@ def parse_logfile(logfile_path: str, locations: List[str],
                                                                           microsecond=0)
     additional_infos = {"last_log_timestamp": int(today_date.timestamp()),
                         "bot_user_agents_nb": len(bot_user_agents),
-                        "client_user_agents_nb": len(client_user_agents)}
+                        "client_user_agents_nb": len(client_user_agents),
+                        "methods": method_counter}
 
 
     return visit_dict, bot_visit_dict, other_visit_dict, additional_infos
@@ -280,6 +314,8 @@ def main() -> None:
         print_visit_dict("Bot visits", bot_visit_dict)
         print(f"There were {additional_infos['bot_user_agents_nb']} unique bot user agent(s)")
     print_visit_dict("Other visits", other_visit_dict)
+    for method, count in additional_infos["methods"].items():
+        print(f"{method}: {count}")
 
     if telegraf_url:
         exporter = TelegrafExporter(telegraf_url=telegraf_url,
@@ -288,6 +324,9 @@ def main() -> None:
                                     source=socket.gethostname())
         exporter.export_result_to_telegraf(visit_dict,
                                            bot_visit_dict,
+                                           {"bot_user_agents":additional_infos['bot_user_agents_nb'],
+                                            "client_user_agents": additional_infos['client_user_agents_nb']},
+                                           additional_infos["methods"],
                                            additional_infos["last_log_timestamp"])
 
 if __name__ == "__main__":