1. 日志分析工具
1.1. 提取日志中错误信息
1def extract_errors(log_file, output_file):
2 with open(log_file, 'r') as f:
3 lines = f.readlines()
4
5 errors = [line for line in lines if "ERROR" in line]
6
7 with open(output_file, 'w') as f:
8 f.writelines(errors)
9
10# 使用示例
11extract_errors("app.log", "errors.log")
1.2. 统计日志中关键词出现次数
1from collections import defaultdict
2
3def count_keywords(log_file, keywords):
4 keyword_count = defaultdict(int)
5
6 with open(log_file, 'r') as f:
7 for line in f:
8 for keyword in keywords:
9 if keyword in line:
10 keyword_count[keyword] += 1
11
12 return keyword_count
13
14# 使用示例
15keywords = ["ERROR", "WARNING", "INFO"]
16result = count_keywords("app.log", keywords)
17print(result)
1.3. 按时间范围过滤日志
1from datetime import datetime
2
3def filter_logs_by_time(log_file, start_time, end_time, output_file):
4 start = datetime.strptime(start_time, "%Y-%m-%d %H:%M:%S")
5 end = datetime.strptime(end_time, "%Y-%m-%d %H:%M:%S")
6
7 withopen(log_file, 'r') as f:
8 logs = f.readlines()
9
10 filtered_logs = []
11 forloginlogs:
12 log_time_str = log.split()[0] + " " + log.split()[1] # 假设时间戳在日志的前两部分
13 log_time = datetime.strptime(log_time_str, "%Y-%m-%d %H:%M:%S")
14 ifstart <= log_time <= end:
15 filtered_logs.append(log)
16
17 withopen(output_file, 'w') as f:
18 f.writelines(filtered_logs)
19
20# 使用示例
21filter_logs_by_time("app.log", "2025-02-26 12:00:00", "2025-02-06 14:00:00", "filtered_logs.log")
1.4. 统计日志中高频错误
1from collections import Counter
2import re
3
4def top_n_errors(log_file, n=5):
5 error_pattern = re.compile(r"ERROR: (.+)")
6 errors = []
7
8 with open(log_file, 'r') as f:
9 for line in f:
10 match = error_pattern.search(line)
11 if match:
12 errors.append(match.group(1))
13
14 return Counter(errors).most_common(n)
15
16# 使用示例
17top_errors = top_n_errors("app.log", n=3)
18print(top_errors)
1.5. 日志文件合并
1def merge_log_files(log_files, output_file):
2 with open(output_file, 'w') as outfile:
3 for log_file in log_files:
4 with open(log_file, 'r') as infile:
5 outfile.write(infile.read())
6
7# 使用示例
8merge_log_files(["log1.log", "log2.log", "log3.log"], "merged_logs.log")
1.6. 日志文件实时监控
1import time
2
3def tail_log_file(log_file):
4 with open(log_file, 'r') as f:
5 f.seek(0, 2) # 移动到文件末尾
6 while True:
7 line = f.readline()
8 if line:
9 print(line.strip())
10 else:
11 time.sleep(0.1)
12
13# 使用示例
14tail_log_file("app.log")
2. 性能监控工具
2.1. 监控cpu和内存使用情况
1import psutil
2import time
3
4def monitor_system(interval=1):
5 while True:
6 cpu_usage = psutil.cpu_percent(interval=interval)
7 memory_usage = psutil.virtual_memory().percent
8 print(f"CPU Usage: {cpu_usage}% | Memory Usage: {memory_usage}%")
9 time.sleep(interval)
10
11# 使用示例
12monitor_system(interval=2)
2.2. 监控GPU使用情况
1import pynvml
2
3def monitor_gpu_usage():
4 pynvml.nvmlInit()
5 device_count = pynvml.nvmlDeviceGetCount()
6
7 for i in range(device_count):
8 handle = pynvml.nvmlDeviceGetHandleByIndex(i)
9 util = pynvml.nvmlDeviceGetUtilizationRates(handle)
10 memory_info = pynvml.nvmlDeviceGetMemoryInfo(handle)
11 print(f"GPU {i}: Usage={util.gpu}%, Memory Used={memory_info.used / 1024 ** 2} MB")
12
13# 使用示例
14monitor_gpu_usage()
2.3. 监控网络带宽
1import psutil
2import time
3
4def monitor_network_usage(interval=1):
5 old_value = psutil.net_io_counters().bytes_sent + psutil.net_io_counters().bytes_recv
6
7 whileTrue:
8 new_value = psutil.net_io_counters().bytes_sent + psutil.net_io_counters().bytes_recv
9 bandwidth = (new_value - old_value) / interval # 计算带宽(字节/秒)
10 print(f"Network Bandwidth: {bandwidth} B/s")
11
12 old_value = new_value
13 time.sleep(interval)
14
15# 使用示例
16monitor_network_usage(interval=2)
2.4. 监控磁盘IO
1import psutil
2import time
3
4def monitor_disk_io(interval=1):
5 old_read = psutil.disk_io_counters().read_bytes
6 old_write = psutil.disk_io_counters().write_bytes
7
8 whileTrue:
9 new_read = psutil.disk_io_counters().read_bytes
10 new_write = psutil.disk_io_counters().write_bytes
11
12 read_speed = (new_read - old_read) / interval
13 write_speed = (new_write - old_write) / interval
14
15 print(f"Read Speed: {read_speed / 1024} KB/s | Write Speed: {write_speed / 1024} KB/s")
16
17 old_read = new_read
18 old_write = new_write
19 time.sleep(interval)
20
21# 使用示例
22monitor_disk_io(interval=2)
2.5. 监控进程资源占用
1import psutil
2
3def monitor_process(pid):
4 process = psutil.Process(pid)
5
6 while True:
7 cpu_usage = process.cpu_percent(interval=1)
8 memory_usage = process.memory_info().rss / 1024 ** 2 # 转换为MB
9 print(f"PID {pid}: CPU={cpu_usage}%, Memory={memory_usage} MB")
10
11# 使用示例
12monitor_process(1234) # 替换为目标进程的PID
2.6. 监控系统温度
1import psutil
2
3def monitor_temperature():
4 temps = psutil.sensors_temperatures()
5 for name, entries in temps.items():
6 for entry in entries:
7 print(f"{name}: {entry.label} = {entry.current}°C")
8
9# 使用示例
10monitor_temperature()
3. 数据处理工具
3.1. 数据清洗
1import pandas as pd
2
3def clean_data(input_file, output_file):
4 df = pd.read_csv(input_file)
5 df.dropna(inplace=True) # 删除空值
6 df.drop_duplicates(inplace=True) # 删除重复值
7 df.to_csv(output_file, index=False)
8
9# 使用示例
10clean_data("data.csv", "cleaned_data.csv")
3.2. 数据对比
1import pandas as pd
2
3def compare_data(file1, file2):
4 df1 = pd.read_csv(file1)
5 df2 = pd.read_csv(file2)
6 diff = df1.compare(df2)
7 return diff
8
9# 使用示例
10result = compare_data("file1.csv", "file2.csv")
11print(result)
4. 文件处理工具
4.1. 批量重命名文件
1import os
2
3def batch_rename(directory, prefix):
4 for count, filename in enumerate(os.listdir(directory)):
5 new_name = f"{prefix}_{count}.txt"
6 os.rename(os.path.join(directory, filename), os.path.join(directory, new_name))
7
8# 使用示例
9batch_rename("/path/to/files", "file")
4.2. 查找大文件
1import os
2
3def find_large_files(directory, size_limit_mb):
4 size_limit = size_limit_mb * 1024 * 1024# 转换为字节
5 large_files = []
6
7 for root, dirs, files in os.walk(directory):
8 for file in files:
9 file_path = os.path.join(root, file)
10 if os.path.getsize(file_path) > size_limit:
11 large_files.append(file_path)
12
13 return large_files
14
15# 使用示例
16large_files = find_large_files("/path/to/directory", 100) # 查找大于100MB的文件
17print(large_files)
5. 网络检测工具
5.1. 检测端口是否开放
1import socket
2
3def check_port(host, port):
4 sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
5 result = sock.connect_ex((host, port))
6 sock.close()
7 return result == 0
8
9# 使用示例
10if check_port("example.com", 80):
11 print("Port 80 is open")
12else:
13 print("Port 80 is closed")
5.2. 批量ping测试
1import os
2
3def ping_hosts(hosts):
4 for host in hosts:
5 response = os.system(f"ping -c 1 {host}")
6 if response == 0:
7 print(f"{host} is up")
8 else:
9 print(f"{host} is down")
10
11# 使用示例
12hosts = ["google.com", "example.com", "localhost"]
13ping_hosts(hosts)
6. 定时任务工具
6.1. 定时执行任务
1import time
2from datetime import datetime
3
4def scheduled_task(interval):
5 while True:
6 print(f"Task executed at {datetime.now()}")
7 time.sleep(interval)
8
9# 使用示例
10scheduled_task(60) # 每60秒执行一次