Managing Linux processes
Table of contents
- Control processes
- See all processes
- GNU long parameters
- Real Time process monitoring
- Signals
- Killing processes
- Resources
Control processes
Managing processes in Linux using the pscommand can be a powerful way to view and control the processes running on your system. The ps command allows you to see a detailed list of all the processes currently running on your system, including their process IDs (PIDs), the user that started them, and their current status. You can also use various options and filters with the ps command to display information about specific processes, or to display information in a particular format.
Default behavior of ps command:
- The default behavior of the "ps" command is to display only the processes that are associated with the current user and are active on the current terminal.
- The standard output of the "ps" command includes the process ID (PID) of the programs, the terminal (TTY) on which they are running and the amount of CPU time the process has consumed.
Command line parameters
- Parameters in the Unix style, denoted by a preceding hyphen
- Parameters in the BSD style, denoted by no preceding hyphen
- Long parameters in the GNU style, denoted by a double hyphen.
The Unix-style parameters originated with the original ps command that run on the AT&T Unix systems invented by Bell Labs. Table show these parameters:
| Parameter | Description |
|---|---|
| -A | Shows all processes |
| -N | Show the opposite of the specified parameters |
| -a | Shows all processes except session headers and processes without a terminal |
| -d | Shows all processes except session headers |
| -e | Shows all processes |
| -C cmslist | Shows processes contained in the list cmdlist |
| -G grplist | Shows processes with a group ID listed in grplist |
| -U userlist | Shows processes owned by a userid listed in userlist |
| -g grplist | Shows processes by session or by groupid contained in grplist |
| -p pidlist | Shows processes with PIDs in the list pidList |
| -s sesslist | Shows processes with a session ID in the list sesslist |
| -t ttylist | Shows processes with terminal ID in the list ttylist |
| -u userlist | Shows processes by effective userid in the list userlist |
| -F | Uses extra full output |
| -O format | Displays specific columns in the list format along with default columns |
| -M | Displays security information about the process |
| -f | Displays a full format listing |
| -j | Shows job information |
| -l | Displays a long listing |
| -o format | Displays only specifc columns listed in format |
| -y | Prevents display of process flags |
| -Z | Displays the security context information |
| -H | Displays processes in a hierarchical format (showing parent processes) |
| -n namelist | Defines the values to display in the WCHAN column |
| -w | Uses wide output format, for unlimited width displays |
| -L | Shows process threads |
| -V | Displays the version of ps |
See all processes
We can see all processes running on the system by using -e and -f parameters. The output shows some useful information:
| Column | Description |
|---|---|
| UID | The user who launched the process |
| PID | The unique identification number of the process |
| PPID | The process ID of the parent process if the process is launched by another one |
| C | Processor utilization over the lifetime of the process |
| STIME | The time when the process started as per the system time |
| TTY | The terminal device where the process was launched |
| TIME | The total amount of CPU time that the process has consumed so far |
| CMD | The name of the program that was started to initiate the process |
When using the BSD-style parameters, the ps command automatically changes the output to simulate the BSD format. An example of this is using the "l" parameter:
ps l
The BSD-style parameters produce a more detailed state code for processes, shown in the "STAT" column. The two-character code provides more specific information about the current state of the process compared to the single-character Unix-style output.
Here is an example of the two-character state code:
| Code | Description |
|---|---|
< |
The process is running at high priority. |
N |
The process is running at low priority. |
L |
The process has pages locked in memory. |
s |
The process is a session header. |
l |
The process is multi-threaded. |
+ |
The process is running in the foreground. |
Additionally, the BSD-style parameters also produce the following additional columns:
| Column | Description |
|---|---|
| VSZ | The size in kilobytes of the process in memory |
| RSS | The physical memory used by the process that isn't swapped out |
GNU long parameters
The developers of GNU added some new options to the ps command, by introducing the GNU long parameters, some of which are similar to existing Unix- or BSD-style parameters, while others bring new capabilities. The table below provides a list of the available GNU long parameters.
| Parameter | Description |
|---|---|
| --deselect | Shows all processes except those listed in the command line |
| --Group grplist | Shows processes whose group ID is listed in grplist |
| --User userlist | Shows processes whose user ID is listed in userlist |
| --group grplist | Shows processes whose effective group ID is listed in grplist |
| --pid pidlist | Shows processes whose process ID is listed in pidlist |
| --ppid pidlist | Shows processes whose parent process ID is listed in pidlist |
| --sid sidlist | Shows processes whose session ID is listed in sidlist |
| --tty ttylist | Shows processes whose terminal device ID is listed in ttylist |
| --user userlist | Shows processes whose effective user ID is listed in userlist |
| --format format | Displays only columns specified in the format |
| --context | Displays additional security information |
| --cols n | Sets screen width to n columns |
| --columns n | Sets screen width to n columns |
| --cumulative | Includes stopped child process information |
| --forest | Displays processes in a hierarchical listing showing parent processes |
| --headers | Repeat column headers on each page of output |
| --no-headers | Prevents display of column headers |
| --lines n | Set the screen height to n lines |
| --rows n | Sets the screen height to n roews |
| --sort order | Defines the column(s) to use for sorting the output |
| --width n | Sets the screen width to n columns |
Real Time process monitoring
The ps command is a useful tool for viewing information about processes running on a system, but it has the limitation of only displaying information at a specific point in time. top command solves this problem by providing real-time process monitoring by updating the information continuously.
The top command output is divided into several sections. The first section provides general system information such as the current time, uptime, number of logged in users, and system load averages. The load averages (1-minute, 5-minute, and 15-minute) indicate the level of activity on the system. High load averages can indicate that the system is under heavy use, but the threshold for a "high" load average varies depending on the specific system and its configuration.
The following table presents the information that gives detail about the output of the top command:
| Parameter | Description |
|---|---|
| PID | The process ID of the process |
| User | The user name of the owner of the process |
| PR | The priority of the process |
| NI | The nice value of the process |
| VIRT | The total amount of virtual memory used by the process |
| RES | The amount of physical memory the process is using |
| SHR | The amount of memory the process is sharing with other processes. |
| S | The process status (D = interruptible sleep, R = running, S = sleeping, T = traced or stopped, or Z = zombies) |
| %CPU | The share of CPU time that the process is using |
| %MEM | The share of available physical memory the process is using |
| TIME+ | The total CPU time the process has used since starting |
| COMMAND | The command line name of the process (program started) |
Signals
In Linux, one way to control processes is through the use of signals. Signals are predefined messages that processes recognize and can choose to ignore or act on. Many well-written applications are able to receive and respond to standard Unix process signals. System administrators use signals to stop or interrupt a process that may be hung up or taking too much CPU.
The following table describes each signal.
| Signal | Name | Description |
|---|---|---|
| 1 | HUP | Hangs up |
| 2 | INT | Interrupts |
| 3 | QUIT | Stops running |
| 9 | KILL | Unconditionally terminates |
| 11 | SEGV | Produces segment violation |
| 15 | TERM | Terminates if possible |
| 17 | STOP | Stops unconditionally, but doesn't terminate |
| 18 | TSTP | Stops or pauses, but continues to run in background |
| 19 | CONT | Resumes execution after STOP or TSTP |
Killing processes
The kill and killall command in Linux allows you to send signals to processes based on their process ID(PID) or name respectively. The default signal sent is TERM, which requests the process to stop running. The kill command can also use the -s parameter to specify other signals such as HUP. The killall command is useful when you want to stop multiple processes by using their names with wildcard characters. To check if the command was effective, you need to run another ps or top command to check if the offending process stopped.
By understanding the commands and concepts presented in this article, you will be well equipped to manage and control processes in Linux.
Resources
Nemeth, E., Snyder, G., Hein, T., Whaley, B., & Mackin, D. (2018). Unix and Linux system administration handbook (5th edition).
Related Content
Author
