Linux系统硬件老化测试脚本:自动化负载与监控

发布于:2024-11-28 ⋅ 阅读:(43) ⋅ 点赞:(0)

简介:

  这篇文章介绍了一款用于Linux系统的自动化硬件老化测试脚本。该脚本能够通过对CPU、内存、硬盘和GPU进行高强度负载测试,持续运行设定的时长(如1小时),以模拟长时间高负荷运行的环境,从而验证硬件的稳定性与可靠性。脚本还包括了系统资源监控,实时显示CPU温度、频率、内存使用情况等信息,并将测试结果记录到日志文件中。测试完成后,脚本会提供详细的反馈,并允许用户选择是否重新执行测试。通过这种方式,用户可以轻松地评估设备的性能和健康状况。

#!/bin/bash

# 请设置老化时长(小时)
set_aging_time=1


# 获取脚本绝对路径
SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
# echo "脚本所在的绝对路径是: ${SCRIPT_DIR}"

# 日志存放路径
log_file=${SCRIPT_DIR}/log_file.log
stress_ng=${SCRIPT_DIR}/stress_ng.log

# 安装stress-ng、figlet和glmark2-es2工具,当前环境为Ubuntu系统,根据系统替换相关安装命令
install_packages() {
	if command -v stress-ng > /dev/null 2>&1 && command -v figlet > /dev/null 2>&1 && command -v glmark2-es2 > /dev/null 2>&1; then
		echo "All software has been installed."
	else
		sudo apt update
		sudo apt install -y stress-ng
		sudo apt install -y glmark2-es2
		sudo apt install -y figlet

		# 检查 stress-ng 是否缺失
		if ! command -v stress-ng > /dev/null 2>&1; then
			echo "stress-ng is not installed"
			exit 1
		fi

		# 检查 figlet 是否缺失
		if ! command -v figlet > /dev/null 2>&1; then
			echo "figlet is not installed"
			exit 1
		fi

		# 检查 glmark2-es2 是否缺失
		if ! command -v glmark2-es2 > /dev/null 2>&1; then
			echo "glmark2-es2 is not installed"
			exit 1
		fi

	fi
}

# CPU stress test
run_cpu_test() {
    # 查看cpu信息 lscpu
    stress-ng --cpu $(nproc) --metrics-brief --timeout ${aging_time}s >> ${stress_ng} 2>&1
}

# Memory stress test
run_memory_test() {
	available_memory=$(free -m | grep -E 'Mem|内存' | awk '{print $7}')
	half_memory=$(echo "$available_memory / $(nproc)" | bc)
	# 运行内存分配释放模式
	stress-ng --vm $(nproc) --vm-bytes ${half_memory}M --metrics-brief --timeout ${aging_time}s >> ${stress_ng} 2>&1
	# 运行内存持续占用模式
	# stress-ng --vm 1 --vm-bytes ${available_memory}M --metrics-brief --vm-keep --timeout ${aging_time}s >> ${stress_ng} 2>&1
}

#HDD stress test
run_disk_stress_test() {
	stress-ng --hdd $(nproc) -i $(nproc) --metrics-brief --timeout ${aging_time}s >> ${stress_ng} 2>&1
}

# GPU stress test
run_gpu_test() {
	timeout ${aging_time} glmark2-es2  --run-forever --annotate > /dev/null 2>&1
}

get_cpu_info() {
	# 当前 CPU 温度路径
	cpu_temp_path="/sys/class/thermal/thermal_zone1/temp"
	cat ${cpu_temp_path} > /dev/null 2>&1
	if [ $? -eq 0 ]; then
		cpu_temp=$(echo "scale=1; $(cat ${cpu_temp_path}) / 1000" | bc)
	else
		cpu_temp=" ---"
	fi

	# 当前 CPU 频率路径
	cpu_cur_freq_path="/sys/devices/system/cpu/cpufreq/policy0/cpuinfo_cur_freq"
	cat ${cpu_cur_freq_path} > /dev/null 2>&1
	if [ $? -eq 0 ]; then
		cpu_cur_freq=$(echo "scale=2; $(cat ${cpu_cur_freq_path}) / 1000" | bc)
	else
		cpu_cur_freq=" --- "
	fi

	# 获取当前CPU使用率
	cpu_usage=$(top -bn1 | grep "Cpu(s)" | sed "s/.*, *\([0-9.]*\)%* id.*/\1/" | awk '{print 100 - $1}')
}

get_memory_info() {
	# 获取总内存、已用内存、缓存
	memory_info=$(free -m)
	total_mem=$(echo "$memory_info" | grep -E 'Mem|内存' | awk '{print $2}')
	used_mem=$(echo "$memory_info" | grep -E 'Mem|内存' | awk '{print $3}')
	# 计算内存使用百分比(保留一位小数)
	used_percent=$(echo "scale=1; $used_mem * 100 / $total_mem" | bc)
}

run_test() {
	touch ${SCRIPT_DIR}/start_state.zz
	echo "Device MAC: $(ip -o link show up | awk '$2 == "eth0:" {print $17}')" >> ${log_file}

	aging_time=$((set_aging_time * 60 * 60))
	echo "Aging duration: ${aging_time}S" >> ${log_file}

    run_cpu_test &
	run_memory_test &
	run_disk_stress_test &
	run_gpu_test &

	echo "start time: $(date)" >> ${log_file}
	start_time=$(date +%s)

	# 清屏
	clear
	# 隐藏光标:使用 ANSI 转义序列
	echo -e "\e[?25l"
	
	# 循环直到老化时间结束
	while true; do
		current_time=$(date +%s)  # 获取当前时间
		elapsed_time=$((current_time - start_time))  # 计算已老化时间

		# 将已老化时间转换为小时、分钟和秒
		hours=$((elapsed_time / 3600))
		minutes=$(( (elapsed_time % 3600) / 60 ))
		seconds=$((elapsed_time % 60))

		tput cup 0 0
		echo -n "Aging time: $(printf "%02d:%02d:%02d" $hours $minutes $seconds)    "

		get_cpu_info
		tput cup 2 0
		echo -n "CPU Usage: ${cpu_usage}%   |  CPU Temp: ${cpu_temp} °C   |  CPU Cur Freq: ${cpu_cur_freq} MHz    "

		get_memory_info
		tput cup 4 0
		echo -n "Total Mem: ${total_mem}M  |  Used Mem: ${used_mem}M     |  Mem Usage: ${used_percent}%    "

		# 检查是否已经达到老化时间
		if [ "$elapsed_time" -ge "$aging_time" ]; then
			echo "stop time: $(date)" >> ${log_file}
			break
		fi

		# 每隔1秒更新一次显示
		sleep 1
	done

	# 等待所有测试完成
    wait
	
	# 显示光标:使用 ANSI 转义序列
	echo -e "\e[?25h"	

	echo ""
	printf "Aging test passed, aging duration: %02d:%02d:%02d" $hours $minutes $seconds 2>&1 | tee -a ${log_file}
	echo ""
	echo -e "\033[32m$(figlet "PASS")\033[0m"
	touch ${SCRIPT_DIR}/end_state.zz
}

install_packages

start_state="${SCRIPT_DIR}/start_state.zz"
end_state="${SCRIPT_DIR}/end_state.zz"
if [[ -e "$start_state" ]] && [[ ! -e "$end_state" ]]; then
	echo -e "\033[31m$(figlet "FAIL")\033[0m"
	read -p "Aging test failed, please choose whether to re-execute aging test? (y/n):" answer
	if [ "$answer" = "Y" ] || [ "$answer" == "y" ]; then
		rm -rf ${SCRIPT_DIR}/log_file.log
		rm -rf ${SCRIPT_DIR}/stress_ng.log
		rm -rf ${SCRIPT_DIR}/start_state.zz
		run_test
	else
		exit 0
	fi

elif [[ -e "$start_state" ]] && [[ -e "$end_state" ]]; then
	echo -e "\033[32m$(figlet "PASS")\033[0m"
	read -p "The equipment has completed the aging test and passed. Would you like to re-execute the aging test? (y/n):" answer
	if [ "$answer" = "Y" ] || [ "$answer" == "y" ]; then
		rm -rf ${SCRIPT_DIR}/log_file.log
		rm -rf ${SCRIPT_DIR}/stress_ng.log
		rm -rf ${SCRIPT_DIR}/start_state.zz
		rm -rf ${SCRIPT_DIR}/end_state.zz
		run_test
	else
		exit 0
	fi

else
	run_test
fi