mirror of
https://github.com/LCTT/TranslateProject.git
synced 2024-12-26 21:30:55 +08:00
Merge pull request #2607 from Vic020/master
Translated:20150326 A Peep into Process Management Commands in Linux.md
This commit is contained in:
Xingyu.Wang
commit
2bdcb59cc6
@ -1,193 +0,0 @@
|
||||
Vic020
|
||||
|
||||
A Peep into Process Management Commands in Linux
|
||||
================================================================================
|
||||
A program in execution is called a process. While a program is an executable file present in storage and is passive, a process is a dynamic entity comprising of allocated system resources, memory, security attributes and has a state associated with it. There can be multiple processes associated with the same program and operating simultaneously without interfering with each other. The operating system efficiently manages and keeps track of all the processes running in the system.
|
||||
|
||||
In order to manage these processes, user should be able to
|
||||
|
||||
- See all the processes that are running
|
||||
- View the system resources consumed by the processes
|
||||
- Locate a particular process and take specific action on it
|
||||
- Change the priority levels associated with processes
|
||||
- Kill the required processes
|
||||
- Restrict the system resources available to processes etc.
|
||||
|
||||
Linux offers many commands to the user to effectively handle the above mentioned scenarios. Let's understand them one by one.
|
||||
|
||||
### 1. ps ###
|
||||
|
||||
'ps' is one of the basic commands in Linux to view the processes on the system. It lists the running processes in a system along with other details such as process id, command, cpu usage, memory usage etc. Some of the following options come handy to get more useful information
|
||||
|
||||
ps -a - List all the running / active processes
|
||||
|
||||
|
||||
|
||||
ps -ef |grep - List only the required process
|
||||
|
||||
ps -aux - Displays processes including those with no terminals(x) Output is user oriented (u) with fields like USER, PID, %CPU, %MEM etc
|
||||
|
||||
### 2. pstree ###
|
||||
|
||||
In Linux, every process gets spawned by its parent process. This command helps visualize the processes by displaying a tree diagram of the processes showing the relationship between them. If a pid is mentioned, the root of the tree will be the pid. Else it will be rooted at init.
|
||||
|
||||
|
||||
|
||||
### 3. top ###
|
||||
|
||||
'top' is a very useful command to monitor the system as it shows the system resources used by different processes. It gives a snapshot of the situation that the system is currently in. Its output includes data like process identification number(PID), user of the process, nice value, %CPU and %memory currently consumed by the process etc. One can use this output to figure out which process is hogging the CPU or memory.
|
||||
|
||||
|
||||
|
||||
### 4. htop ###
|
||||
|
||||
htop is similar to top, but is an interactive text mode process viewer. It displays the per CPU usage and memory, swap usage using a text graph. One can use the Up/Down arrow key to select processes, F7 and F8 to change the priority and F9 to kill a process. It is not present by default in the system and need to be installed explicitly.
|
||||
|
||||
|
||||
|
||||
### 5. nice ###
|
||||
|
||||
With the help of nice command, users can set or change the priorities of processes in Linux. Higher the priority of a process, more is the CPU time allocated by the kernel for it. By default, a process gets launched with priority 0. Process priority can be viewed using the top command output under the NI (nice value) column.
|
||||
|
||||
Values of process priority range from -20 to 19. Lower the nice value, higher the priority.
|
||||
|
||||
nice <priority value> <process name> - starts the process by setting its priority to the given value
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
In the above example, 'top' gets launched with a priority -3.
|
||||
|
||||
### 6. renice ###
|
||||
|
||||
It is similar to nice command. Use this command to change the priority of an already running process. Please note that users can change the priority of only the processes that they own.
|
||||
|
||||
renice -n -p - change the priority of the given process
|
||||
|
||||
|
||||
|
||||
Priority of process with id 3806 which had an initial priority of 0 is now changed to priority 4.
|
||||
|
||||
renice -u -g - change the priority of processes owned by the given user and group
|
||||
|
||||
|
||||
In the above example, priority of all processes owned by user 'mint' get changed to '-3'
|
||||
|
||||
### 7. kill ###
|
||||
|
||||
This is a command used to terminate processes by sending signals. If a process is not responding to kill command, then it can be forcefully killed using the kill -9 command. But this needs to be used carefully as it does not give a chance for the process to clean up and might end up in corrupted files. If we are not aware of the PID of the process to be killed or want to mention the process name to be killed, then killall comes to rescue.
|
||||
|
||||
kill <pid>
|
||||
|
||||
kill -9 <pid>
|
||||
|
||||
killall -9 - kill all instances having the same process name
|
||||
|
||||
If you use kill, you need to know the process id of the process to be killed. pkill is a similar command but can be used to kill processes using a pattern, i.e. process name, process owner etc.
|
||||
|
||||
pkill <process name>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
### 8. ulimit ###
|
||||
|
||||
Command useful in controlling the system-wide resources available to the shells and processes. Mostly useful for system administrators to manage systems that are heavily used and have performance problems. Limiting the resources ensures that important processes continue to run while other processes do not consume more resources.
|
||||
|
||||
ulimit -a - Displays the current limits associated with the current user.
|
||||
|
||||
|
||||
|
||||
-f - maximum file size
|
||||
|
||||
-v - maximum virtual memory size (in KB)
|
||||
|
||||
-n - maximum file descriptor plus 1
|
||||
|
||||
-H : To change and report the hard limit
|
||||
|
||||
-S : To change and report the soft limit
|
||||
|
||||
Check out the ulimit man page for more options.
|
||||
|
||||
### 9. w ###
|
||||
|
||||
w gives us information about the users who have currently logged in and the processes that they are running. The header details displayed contain information like current time, how long the system has been running, total number of users logged in, load average of the system for the last 1, 5 and 15 minutes
|
||||
|
||||
Based on the user information, one can take care before terminating any processes that do not belong to them.
|
||||
|
||||
|
||||
|
||||
**who** is a related command and gives a list of currently logged in users, time of last system boot, current run levels etc.
|
||||
|
||||
|
||||
|
||||
**whoami** command prints the username of the current user ID
|
||||
|
||||
|
||||
|
||||
### 10. pgrep ###
|
||||
|
||||
pgrep stands for "Process-ID Global Regular Expression Print". It scans the currently running processes and lists the process IDs that match the selection criteria mentioned on command line to stdout. Useful for retrieving the process id of a process by using its name.
|
||||
|
||||
pgrep -u mint sh
|
||||
|
||||
This command will display the process ID for the process named 'sh' and owned by user 'mint'
|
||||
|
||||
|
||||
|
||||
### 11. fg , bg ###
|
||||
|
||||
Sometimes, the commands that we execute take a long time to complete. In such situations, we can push the jobs to be executed in the background using 'bg' command and can be brought to the foreground with the 'fg' command.
|
||||
|
||||
We can start a program in background by using the '&' :
|
||||
|
||||
find . -name *iso > /tmp/res.txt &
|
||||
|
||||
A program that is already running can also be sent to the background using 'CTRL+Z' and 'bg' command:
|
||||
|
||||
find . -name *iso > /tmp/res.txt & - start the job in the background
|
||||
|
||||
ctrl+z - suspend the currently executing foreground job
|
||||
|
||||
bg - push the command execution to background
|
||||
|
||||
We can list all the background processes using 'jobs' command
|
||||
|
||||
jobs
|
||||
|
||||
We can bring back a background process to foreground using the 'fg' command.
|
||||
|
||||
fg %
|
||||
|
||||
|
||||
|
||||
### 12. ipcs ###
|
||||
|
||||
ipcs command is used for listing the inter-process communication facilities (shared memory, semaphores and message queues).
|
||||
|
||||
This command with -p option can be combined with options -m, -s or-q to display the process id which recently accessed the corresponding ipc facility.
|
||||
|
||||
ipcs -p -m
|
||||
|
||||
The screen shot below is listing the creator id and process id which accessed the shared memory recently.
|
||||
|
||||
|
||||
|
||||
### Conclusion ###
|
||||
|
||||
Generally, it is considered the job of administrators to fix problems and tweak the system to improve the performance. But users also need to deal with processes in Linux quite often. Hence, it is essential to familiarise ourselves with the various commands available to manage them effectively.
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: http://linoxide.com/linux-command/process-management-commands-linux/
|
||||
|
||||
作者：[B N Poornima][a]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创翻译，[Linux中国](http://linux.cn/) 荣誉推出
|
||||
|
||||
[a]:http://linoxide.com/author/bnpoornima/
|
||||
@ -0,0 +1,193 @@
|
||||
12个进程管理命令
|
||||
================================================================================
|
||||
每个程序在执行时叫进程。当程序是在存储中是可执行文件并且运行的时候，每个进程会被动态得分配系统资源，内存，安全属性和与之相关的状态。可以有多个进程关联同一个程序，并同时执行不会互相干扰。操作系统会有效地管理和追踪所有运行着的进程。
|
||||
|
||||
为了管理这些进程，用户应该能够：
|
||||
|
||||
- 查看所有运行中的进程
|
||||
- 查看进程消耗资源
|
||||
- 定位个别进程并且可以执行指定动作
|
||||
- 改变进程的优先级
|
||||
- 杀死指定进程
|
||||
- 限制进程的有效系统资源
|
||||
- 等
|
||||
|
||||
Linux提供了许多命令给用户来高效掌控上述的操作。接下来，一个一个的来讲解下。
|
||||
|
||||
### 1. ps ###
|
||||
|
||||
'ps'是最基础浏览系统中的进程的命令。能列出系统中运行的进程，包括进程号，命令，CPU使用量，内存使用量等。下述选项可以得到更多有用的消息。
|
||||
|
||||
ps -a - 列出所有运行中/激活进程
|
||||
|
||||
|
||||
|
||||
ps -ef |grep - 列出需要进程
|
||||
|
||||
ps -aux - 展示进程包括、终端（x）和用户（u）信息，如USER, PID, %CPU, %MEM等
|
||||
|
||||
### 2. pstree ###
|
||||
|
||||
linux中，每一个进程都是由父进程创建的。此命令帮助可视化进程，通过显示进程树状图表展示进程间关系。如果使用pid了，那么树的根是pid。不然将会是init。
|
||||
|
||||
|
||||
|
||||
### 3. top ###
|
||||
|
||||
‘top’是一个更加有用的命令，通过不同的进程所使用的资源可以监视系统。它提供实时的系统状态信息。显示进程的数据包括PID，进程用户，优先值，%CPU，%memory等。可以使用这些显示指示出资源使用量。
|
||||
|
||||
|
||||
|
||||
### 4. htop ###
|
||||
|
||||
htop与top很类似，但是htop是交互式的模式进程查看器。它通过文字图像显示每一个CPU和内存使用量、swap使用量。上下键选择进程，F7和F8改变优先级，F9杀死进程。Htop不是系统默认，所以需要额外安装。
|
||||
|
||||
|
||||
|
||||
### 5. nice ###
|
||||
|
||||
通过nice命令的帮助，用户可以设置和改变进程的优先级。提高一个进程的优先级，内核会分配更多CPU时间片给这个进程。默认情况下，进程以0的优先级启动。进程优先级可以通过top命令显示的NI（nice value）列查看。
|
||||
|
||||
进程优先级值的范围从-20到19。值越低，越优先。
|
||||
|
||||
nice <优先值> <进程名> - 通过给定的优先值启动一个程序
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
上述命令例子，可以看到‘top’命令获得了-3的优先值。
|
||||
|
||||
### 6. renice ###
|
||||
|
||||
renice命令类似nice命令。使用这个命令可以改变正在运行的进程优先值。注意，用户只能改变属于他们的进程的优先值。
|
||||
|
||||
renice -n -p - 改变指定进程的优先值
|
||||
|
||||
|
||||
|
||||
优先值初始化为0的3806号进程优先值已经变成了4.
|
||||
|
||||
renice -u -g - 通过指定用户和组来改变进程优先值
|
||||
|
||||
|
||||
|
||||
上述例子中，用户为‘mint’的所有进程优先值变为‘-3’。
|
||||
|
||||
### 7. kill ###
|
||||
|
||||
这个命令通过发送信号结束进程。如果一个进程没有响应杀死命令，这也许就需要强制杀死，使用-9参数来执行。注意，使用强制杀死的时候一定要小心，因为没有机会确定是否写入完成、是否结束等。如果我们不知道进程PID或者打算用名字杀死进程时候，killall就能派上用场。
|
||||
|
||||
kill <pid>
|
||||
|
||||
kill -9 <pid>
|
||||
|
||||
killall -9 - 杀死所有拥有同样名字的进程
|
||||
|
||||
如果你使用kill，你需要知道进程ID号。pkill是类似的命令但只需要一个模式，如果进程名，进程拥有者等
|
||||
|
||||
pkill <进程名>
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
### 8. ulimit ###
|
||||
|
||||
命令在控制系统资源在shell和进程上的分配量。对于系统管理员是最有用的，可以管理系统倾向和性能问题。限制资源大小可以确保重要进程持续运行，其他进程不会占用过多资源。
|
||||
|
||||
ulimit -a - 显示当前用户关联的资源限制
|
||||
|
||||
|
||||
|
||||
-f - 最大文件大小
|
||||
|
||||
-v - 最大虚拟内存大小（KB）
|
||||
|
||||
-n - 最大文件描述符加1
|
||||
|
||||
-H : 改变和报告硬限制
|
||||
|
||||
-S : 改变和报告软限制
|
||||
|
||||
浏览ulimit man页面获取更多选项。
|
||||
|
||||
### 9. w ###
|
||||
|
||||
w提供正确登录的用户和其执行的进程的信息。显示信息头包含信息像现在时间，系统运行时长，登录用户总数，过去1，5，15分钟负载均衡数。
|
||||
|
||||
基于这些用户信息，用户可以在终止进程前查看不属于他们的进程。
|
||||
|
||||
|
||||
|
||||
**who**是类似命令，提供列表，包含当前登录用户，系统启动时间，运行级别等。
|
||||
|
||||
|
||||
|
||||
**whoami** 命令输出当前用户ID
|
||||
|
||||
|
||||
|
||||
### 10. pgrep ###
|
||||
|
||||
pgrep为"进程号全局正则匹配打印"。命令扫描当前运行进程，然后列出选择标准匹配出的进程ID到标准输出。对于通过名字检索进程号是很有用。
|
||||
|
||||
pgrep -u mint sh
|
||||
|
||||
这个命令将会显示用户为‘mint’和进程名为‘sh’的进程ID。
|
||||
|
||||
|
||||
|
||||
### 11. fg , bg ###
|
||||
|
||||
有时，命令需要很长的时间才能执行完成。对于这种情况，我们使用‘bg’命令可以将任务放在后台执行，而是用‘fg’可以调到前台来使用。
|
||||
|
||||
通过‘&’,我们后台启动一个程序:
|
||||
|
||||
find . -name *iso > /tmp/res.txt &
|
||||
|
||||
一个正在运行的程序也可以通过“CTRL+Z”和“bg”命令组合放到后台运行。
|
||||
|
||||
find . -name *iso > /tmp/res.txt & - 启动一个程序
|
||||
|
||||
ctrl+z - 挂起当前执行程序
|
||||
|
||||
bg - 将程序发送到后台运行
|
||||
|
||||
我们可以使用‘jobs’命令列出所有后台进程。
|
||||
|
||||
jobs
|
||||
|
||||
使用‘fg’命令可以将后台程序调到前台执行。
|
||||
|
||||
fg %
|
||||
|
||||
|
||||
|
||||
### 12. ipcs ###
|
||||
|
||||
ipcs命令报告进程间通信设施状态。（共享内存，信号量和消息队列）
|
||||
|
||||
用-p参数联合-m、-s或-q使用，可以获得相关的进程间通信的进程ID。
|
||||
|
||||
ipcs -p -m
|
||||
|
||||
下面屏幕截图列出了共享内存段的创建者ID和进程ID。
|
||||
|
||||
|
||||
|
||||
### 总结 ###
|
||||
|
||||
总之 ，这些命令可以帮助管理员修复问题和改善性能。同样作为一名普通用户也需要解决进程间出现的问题。所以，熟悉如此繁多的命令，从能有效管理进程是行之有效。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: http://linoxide.com/linux-command/process-management-commands-linux/
|
||||
|
||||
作者：[B N Poornima][a]
|
||||
译者：[VicYu/Vic020](http://vicyu.net)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创翻译，[Linux中国](http://linux.cn/) 荣誉推出
|
||||
|
||||
[a]:http://linoxide.com/author/bnpoornima/
|
||||
Loading…
Reference in New Issue
Block a user