![监控远程服务器上的 RAM 和 CPU 使用情况?](https://linux22.com/image/870852/%E7%9B%91%E6%8E%A7%E8%BF%9C%E7%A8%8B%E6%9C%8D%E5%8A%A1%E5%99%A8%E4%B8%8A%E7%9A%84%20RAM%20%E5%92%8C%20CPU%20%E4%BD%BF%E7%94%A8%E6%83%85%E5%86%B5%EF%BC%9F.png)
我想监控远程服务器上的 CPU 和 RAM 使用情况,并设置一个通知系统,当 RAM 使用率过高时通知我并自动终止该进程。
我尝试了 Nagios 插件,但出现了一些错误。
还有其他替代方案吗?
编辑:请不要建议htop
、、free
之top
类的,因为我不会继续通过 ssh 进入远程服务器并反复检查其健康状况!
答案1
您熟悉吗top
?显示每个进程的 CPU 和 RAM 使用情况(以及更多)以及总量。
您还可以运行top -u USER
以仅查看特定用户的进程。只需top
从不带参数的运行开始。q
按以退出。
编辑
并在脚本中使用:尝试top -b -n1
在非交互模式下运行。然后您可以使用来解析数据head
,sed
无论如何grep
您都可以。
答案2
我创建了一些自定义脚本。虽然对你来说不是完整的解决方案,但也许有帮助。
#!/bin/bash
#we use output from "uptime" and "w" on every 5 min
UPTIME=`uptime`;
W=`w`;
RAM=`free -m`;
PS=`ps auxf | sort -nr -k 4 | head -15`
DATUM=`date`;
sleep 3
MYSQL=`mysqladmin -u root -pmz567sQld processlist;`
sleep 3
SYN=`netstat -natp | grep SYN`
sleep 3
HTTPD=`pgrep httpd | wc -l`
sleep 3
echo "######################################" >> /var/log/load-result/load_log
echo "$DATE" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "UPTIME:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$UPTIME" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "W COMMAND:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$W" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "FREE RAM:" >> /var/log/load-result/load_log
echo "$RAM" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "TOP 10 processes:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$PS" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "MYSQL PROCESSLIST:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$MYSQL" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "POTENTIAL ATTACKERS:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$SYN" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "NUMBER OF HTTPD PROCESS (currently maxclients are 64):" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "$HTTPD" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "CURRENTLY OPENED HTTPD CONNECTIONS:" >> /var/log/load-result/load_log
echo "" >> /var/log/load-result/load_log
echo "#####################################" >> /var/log/load-result/load_log
之后还添加 crontab 规则,每 5 分钟将日志与日志文件一起发送到某些邮件。每 5 分钟运行一次脚本
sudo nano /etc/crontab
*/5 * * * * sh /path_to_script/name_of_script
*/5 * * * * mail -s "insert_subject" some_adress@domain < /var/log/load-result/load_log