请问执行一键脚本后，如果不出错、vps也没有重启，它会一直自动运行下去吗？按照一键脚本固定的规则消耗甲骨文的资源
谢谢

I'm using a 4 core VM.Standard.A1.Flex, and also your NeverIdle-linux-arm64 script, and noticed that the CPU activity does not reach the 95th Percentile target set by Oracle.
Even if the CPU feature was set to run -c 1m, it would still not reach the 95th Percentile target, as it only runs for less than a minute each time, and then only reached 8% at it's peak, and most times it's much lower.

FROM alpine:latest

Install dependencies

RUN apk update && apk add --no-cache wget

Download script

RUN wget https://github.com/layou233/NeverIdle/releases/download/0.1/NeverIdle-linux-arm64 -O /usr/local/bin/NeverIdle

Change file permission

RUN chmod 777 /usr/local/bin/NeverIdle

Run the script with default arguments

CMD ["/bin/sh", "-c", "/usr/local/bin/NeverIdle -c 2h -m 2 -n 4h"]

i use this with caprover so i can run this with docker
NOTE :- change wget url as reuired for amd or arm server

FROM alpine:latest

# Install dependencies
RUN apk update && apk add --no-cache wget

# Download script
RUN wget https://github.com/layou233/NeverIdle/releases/download/0.1/NeverIdle-linux-arm64 -O /usr/local/bin/NeverIdle

# Change file permission
RUN chmod 777 /usr/local/bin/NeverIdle

# Run the script with default arguments
CMD ["/bin/sh", "-c", "/usr/local/bin/NeverIdle -c 2h -m 2 -n 4h"]

但心vps哪天自动重启后不跑了，自己又不知道，谢谢

大佬，小白不懂弱弱的问一句CPU占用模块是否会生产垃圾文件

ARM DD过的Debian11，内存占用不生效，依然识别为AMD

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

This repository currently has no open or pending branches.

Detected dependencies

• Check this box to trigger a request for Renovate to run again on this repository

比如请求的url之类的，然后被团灭.....
今天决定试试用这个，然后不知道有没有设置程序的nice值，让cpu在关键时候能让位一下。
昨天常用来翻墙的被停机了，让手动去打开，另外两个不用来翻墙的，却没事...但是那两台偶尔连一下ssh编译个程序啥的，但是肯定三项都不达标。

已经用上了，感谢！
有个问题是比如默认是设定2个小时占用一次cpu，请问下这个会持续多久时间

RT

Hi there,

i'm getting the following error when starting version 0.2

Start:

./NeverIdle -c 1h -n 3h > NeverIdle.log 2>&1 &

Logfile shows:

====================
Starting CPU wasting with interval 1h0m0s
====================
====================
Starting network speed testing with interval 3h0m0s
====================
[CPU] Successfully wasted on 2023-03-04 11:17:16.885686988 +0000 GMT m=+0.009873685
panic: invalid argument to Int31n

goroutine 17 [running]:
math/rand.(*Rand).Int31n(0x0?, 0x0?)
        math/rand/rand.go:138 +0xdb
math/rand.Int31n(...)
        math/rand/rand.go:353
github.com/layou233/neveridle/waste.Network(0x0?)
        github.com/layou233/neveridle/waste/network.go:35 +0x211
created by main.main
        ./main.go:52 +0x69d

Version 0.1 works.

uname -a
Linux 4.14.35-2047.522.3.el7uek.x86_64 #2 SMP Fri Jan 20 16:05:02 PST 2023 x86_64 x86_64 x86_64 GNU/Linux

Image is "Oracle-Linux-7.7-2019.08.28-0"

Thanks

[CPU] Successfully wasted on 2023-02-21 22:57:18.687024367 +0000 UTC m=+1850408.314042512
panic: chacha20: counter overflow

goroutine 52850 [running]:
golang.org/x/crypto/chacha20.(*Cipher).XORKeyStream(0xc000108160?, {0xc000180000?, 0xc000147b80?, 0xc000880b40?}, {0xc000180000?, 0x6ed310?, 0xc0008f8600?})
golang.org/x/[email protected]/chacha20/chacha_generic.go:218 +0x408
main.WasteCPU.func1()
NeverIdle/cpu.go:30 +0x48
created by main.WasteCPU
NeverIdle/cpu.go:28 +0x1e5
Ubuntu 18.04.4 LTS (GNU/Linux 5.4.0-1066-oracle x86_64)

看了两台arm机器，好像都没有显示，一些数据如下：

[NETWORK] SpeedTest Ping: 152.278863ms , 2.1863757762040916 , Upload: 0.6479155546829458 via [37235] 8331.62km
Shenyang (China) by Unicom

[NETWORK] SpeedTest Ping: 84.991532ms , 34.01522008046169 , Upload: 37.481146947397846 via [37235] 8331.62km
Shenyang (China) by Unicom

[NETWORK] SpeedTest Ping: 124.063885ms , -1 , Upload: -1 via [37235] 8331.62km
Shenyang (China) by Unicom
[NETWORK] SpeedTest Ping: -1ns , -1 , Upload: -1 via [5249] 2.21km
Seoul (South Korea) by Ookla

[NETWORK] SpeedTest Ping: -1ns , -1 , Upload: -1 via [5249] 2.21km
Seoul (South Korea) by Ookla

[NETWORK] SpeedTest Ping: -1ns , -1 , Upload: -1 via [5249] 2.21km
Seoul (South Korea) by Ookla

下载好程序后如何使用，小白请教，谢谢。

请问这个CPU消耗能指定线程数吗
每次消耗会跑多少时间呢？
如果只是突发的高占用可能被95百分位数忽略？
能考虑弄个限制CPU使用率吗。比如设置常年占用10%就不用指定间隔，一直跑就好了。

刚看了下nice值和普通程序是一样的都是0，还得手动renice修改，go语言里应该也有nice函数可以实现吧。
cpu好像也是双核跑了20秒。我觉得占多少cpu最好可以由用户设置。至少可以控制跑几个线程。

请求支持。

RT，本人ARM A1实例(2ocpu 24G)，目前使用宝塔面板的计划任务执行该脚本（因为不懂代码），已将脚本参数设置为 -cp 0.3 -m 2 -n 2h，如下图所示

运行后cpu占用、内存占用都在30左右、网络测速也正常进行。感谢开发者的无私分享。但有些疑问：

1. 计划任务里已经定义了运行脚本的时间间隔，脚本本身也有个-n两小时的时间间隔，叠加在一起的话会怎么计算的。我看Oracle最新的要求已经变成了内存、CPU、网络使用率最低需要95%时间>20%了，我这里的设置是否能满足需求呢？
2. 我发现脚本每执行一次，系统就会多出一个相同的NerverIdle-linux进程，如此CPU和内存越越来越大，直到超过80%收到告警，然后为了避免100%我每次都要手动kill一下多出来的重复进程。所以按我的理解这个脚本只要跑起来就已经起到了CPU和内存占用的目的，似乎没必要一直重复去运行它？就是测速是一次性的，可能需要隔段时间去运行一次。如果是这样的话，那我应该怎么去设定这个计划呢（不知道我的理解是否有问题）
   以上，希望有大佬帮忙解答一下，谢谢！！

RT

如图

内存能否是小数？整个G的消耗太浪费了呀