How To Use ps, kill, and nice to Manage Processes in Linux
Introduction
A Linux server, like any other computer you may be familiar with, runs applications. To the computer, these are considered "processes".
While Linux will handle the low-level, behind-the-scenes management in a process's life-cycle, you will need a way of interacting with the operating system to manage it from a higher-level.
In this guide, we will discuss some simple aspects of process management. Linux provides an abundant collection of tools for this purpose.
We will explore these ideas on an Ubuntu 12.04 VPS, but any modern Linux distribution will operate in a similar way.
How To View Running Processes in Linux
top
The easiest way to find out what processes are running on your server is to run the top command:
top
top - 15:14:40 up 46 min, 1 user, load average: 0.00, 0.01, 0.05
Tasks: 56 total, 1 running, 55 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 1019600k total, 316576k used, 703024k free, 7652k buffers
Swap: 0k total, 0k used, 0k free, 258976k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1 root 20 0 24188 2120 1300 S 0.0 0.2 0:00.56 init
2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kthreadd
3 root 20 0 0 0 0 S 0.0 0.0 0:00.07 ksoftirqd/0
6 root RT 0 0 0 0 S 0.0 0.0 0:00.00 migration/0
7 root RT 0 0 0 0 S 0.0 0.0 0:00.03 watchdog/0
8 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 cpuset
9 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 khelper
10 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kdevtmpfs
The top chunk of information give system statistics, such as system load and the total number of tasks.
You can easily see that there is 1 running process, and 55 processes are sleeping (aka idle/not using CPU resources).
The bottom portion has the running processes and their usage statistics.
htop
An improved version of top, called htop, is available in the repositories. Install it with this command:
sudo apt-get install htop
If we run the htop command, we will see that there is a more user-friendly display:
htop
Mem[||||||||||| 49/995MB] Load average: 0.00 0.03 0.05
CPU[ 0.0%] Tasks: 21, 3 thr; 1 running
Swp[ 0/0MB] Uptime: 00:58:11
PID USER PRI NI VIRT RES SHR S CPU% MEM% TIME+ Command
1259 root 20 0 25660 1880 1368 R 0.0 0.2 0:00.06 htop
1 root 20 0 24188 2120 1300 S 0.0 0.2 0:00.56 /sbin/init
311 root 20 0 17224 636 440 S 0.0 0.1 0:00.07 upstart-udev-brid
314 root 20 0 21592 1280 760 S 0.0 0.1 0:00.06 /sbin/udevd --dae
389 messagebu 20 0 23808 688 444 S 0.0 0.1 0:00.01 dbus-daemon --sys
407 syslog 20 0 243M 1404 1080 S 0.0 0.1 0:00.02 rsyslogd -c5
408 syslog 20 0 243M 1404 1080 S 0.0 0.1 0:00.00 rsyslogd -c5
409 syslog 20 0 243M 1404 1080 S 0.0 0.1 0:00.00 rsyslogd -c5
406 syslog 20 0 243M 1404 1080 S 0.0 0.1 0:00.04 rsyslogd -c5
553 root 20 0 15180 400 204 S 0.0 0.0 0:00.01 upstart-socket-br
You can learn more about how to use top and htop here.
How To Use ps to List Processes
Both top and htop provide a nice interface to view running processes similar to a graphical task manager.
However, these tools are not always flexible enough to adequately cover all scenarios. A powerful command called ps is often the answer to these problems.
When called without arguments, the output can be a bit lack-luster:
ps
PID TTY TIME CMD
1017 pts/0 00:00:00 bash
1262 pts/0 00:00:00 ps
This output shows all of the processes associated with the current user and terminal session. This makes sense because we are only running bash and ps with this terminal currently.
To get a more complete picture of the processes on this system, we can run the following:
ps aux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.2 24188 2120 ? Ss 14:28 0:00 /sbin/init
root 2 0.0 0.0 0 0 ? S 14:28 0:00 [kthreadd]
root 3 0.0 0.0 0 0 ? S 14:28 0:00 [ksoftirqd/0]
root 6 0.0 0.0 0 0 ? S 14:28 0:00 [migration/0]
root 7 0.0 0.0 0 0 ? S 14:28 0:00 [watchdog/0]
root 8 0.0 0.0 0 0 ? S< 14:28 0:00 [cpuset]
root 9 0.0 0.0 0 0 ? S< 14:28 0:00 [khelper]
. . .
These options tell ps to show processes owned by all users (regardless of their terminal association) in a user-friendly format.
To see a tree view, where hierarchal relationships are illustrated, we can run the command with these options:
ps axjf
PPID PID PGID SID TTY TPGID STAT UID TIME COMMAND
0 2 0 0 ? -1 S 0 0:00 [kthreadd]
2 3 0 0 ? -1 S 0 0:00 \_ [ksoftirqd/0]
2 6 0 0 ? -1 S 0 0:00 \_ [migration/0]
2 7 0 0 ? -1 S 0 0:00 \_ [watchdog/0]
2 8 0 0 ? -1 S< 0 0:00 \_ [cpuset]
2 9 0 0 ? -1 S< 0 0:00 \_ [khelper]
2 10 0 0 ? -1 S 0 0:00 \_ [kdevtmpfs]
2 11 0 0 ? -1 S< 0 0:00 \_ [netns]
. . .
As you can see, the process kthreadd is shown to be a parent of the ksoftirqd/0 process and the others.
A Note About Process IDs
In Linux and Unix-like systems, each process is assigned a process ID, or PID. This is how the operating system identifies and keeps track of processes.
A quick way of getting the PID of a process is with the pgrep command:
pgrep bash
1017
This will simply query the process ID and return it.
The first process spawned at boot, called init, is given the PID of "1".
pgrep init
1
This process is then responsible for spawning every other process on the system. The later processes are given larger PID numbers.
A process's parent is the process that was responsible for spawning it. Parent processes have a PPID, which you can see in the column headers in many process management applications, including top, htop and ps.
Any communication between the user and the operating system about processes involves translating between process names and PIDs at some point during the operation. This is why utilities tell you the PID.
Parent-Child Relationships
Creating a child process happens in two steps: fork(), which creates new address space and copies the resources owned by the parent via copy-on-write to be available to the child process; and exec(), which loads an executable into the address space and executes it.
In the event that a child process dies before its parent, the child becomes a zombie until the parent has collected information about it or indicated to the kernel that it does not need that information. The resources from the child process will then be freed. If the parent process dies before the child, however, the child will be adopted by init, though it can also be reassigned to another process.
How To Send Processes Signals in Linux
All processes in Linux respond to signals. Signals are an os-level way of telling programs to terminate or modify their behavior.
How To Send Processes Signals by PID
The most common way of passing signals to a program is with the kill command.
As you might expect, the default functionality of this utility is to attempt to kill a process:
kill PID_of_target_process
This sends the TERM signal to the process. The TERM signal tells the process to please terminate. This allows the program to perform clean-up operations and exit smoothly.
If the program is misbehaving and does not exit when given the TERM signal, we can escalate the signal by passing the KILL signal:
kill -KILL PID_of_target_process
This is a special signal that is not sent to the program.
Instead, it is given to the operating system kernel, which shuts down the process. This is used to bypass programs that ignore the signals sent to them.
Each signal has an associated number that can be passed instead of the name. For instance, You can pass "-15" instead of "-TERM", and "-9" instead of "-KILL".
How To Use Signals For Other Purposes
Signals are not only used to shut down programs. They can also be used to perform other actions.
For instance, many daemons will restart when they are given the HUP, or hang-up signal. Apache is one program that operates like this.
sudo kill -HUP pid_of_apache
The above command will cause Apache to reload its configuration file and resume serving content.
You can list all of the signals that are possible to send with kill by typing:
kill -l
1) SIGHUP 2) SIGINT 3) SIGQUIT 4) SIGILL 5) SIGTRAP
6) SIGABRT 7) SIGBUS 8) SIGFPE 9) SIGKILL 10) SIGUSR1
11) SIGSEGV 12) SIGUSR2 13) SIGPIPE 14) SIGALRM 15) SIGTERM
. . .
How To Send Processes Signals by Name
Although the conventional way of sending signals is through the use of PIDs, there are also methods of doing this with regular process names.
The pkill command works in almost exactly the same way as kill, but it operates on a process name instead:
pkill -9 ping
The above command is the equivalent of:
kill -9 `pgrep ping`
If you would like to send a signal to every instance of a certain process, you can use the killall command:
killall firefox
The above command will send the TERM signal to every instance of firefox running on the computer.
How To Adjust Process Priorities
Often, you will want to adjust which processes are given priority in a server environment.
Some processes might be considered mission critical for your situation, while others may be executed whenever there might be leftover resources.
Linux controls priority through a value called niceness.
High priority tasks are considered less nice, because they don't share resources as well. Low priority processes, on the other hand, are nice because they insist on only taking minimal resources.
When we ran top at the beginning of the article, there was a column marked "NI". This is the nice value of the process:
top
Tasks: 56 total, 1 running, 55 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.3%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 1019600k total, 324496k used, 695104k free, 8512k buffers
Swap: 0k total, 0k used, 0k free, 264812k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1635 root 20 0 17300 1200 920 R 0.3 0.1 0:00.01 top
1 root 20 0 24188 2120 1300 S 0.0 0.2 0:00.56 init
2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kthreadd
3 root 20 0 0 0 0 S 0.0 0.0 0:00.11 ksoftirqd/0
Nice values can range between "-19/-20" (highest priority) and "19/20" (lowest priority) depending on the system.
To run a program with a certain nice value, we can use the nice command:
nice -n 15 command_to_execute
This only works when beginning a new program.
To alter the nice value of a program that is already executing, we use a tool called renice:
renice 0 PID_to_prioritize
Note: While nice operates with a command name by necessity, renice operates by calling the process PID
Conclusion
Process management is a topic that is sometimes difficult for new users to grasp because the tools used are different from their graphical counterparts.
However, the ideas are familiar and intuitive, and with a little practice, will become natural. Because processes are involved in everything you do with a computer system, learning how to effectively control them is an essential skill.
13 Comments