How can a Linux/Unix Bash script get its own PID?

Question

I have a script in Bash called Script.sh that needs to know its own PID. In other words, I need to get PID inside Script.sh.

Any idea how to do this?

score 310 · Accepted Answer · edited Apr 22 '21 at 09:31

310

The variable $$ contains the PID.

edited Apr 22 '21 at 09:31

vallentin

23,478
6
59
81

answered Mar 22 '10 at 15:53

Paul Tomblin

179,021
58
319
408

How can i get the pid from outside the script , i use `top` but it shows a lot of random processes – Murtaza Haji Apr 13 '22 at 21:38

score 98 · Answer 2 · edited Jan 15 '21 at 12:52

use $BASHPID or $$

See the [manual][1] for more information, including differences between the two.

TL;DRTFM

$$ Expands to the process ID of the shell.
- In a () subshell, it expands to the process ID of the invoking shell, not the subshell.
$BASHPID Expands to the process ID of the current Bash process (new to bash 4).
- In a () subshell, it expands to the process ID of the subshell [1]: http://www.gnu.org/software/bash/manual/bashref.html#Bash-Variables

score 46 · Answer 3 · answered Mar 22 '10 at 17:55

46

In addition to the example given in the Advanced Bash Scripting Guide referenced by Jefromi, these examples show how pipes create subshells:

$ echo $$ $BASHPID | cat -
11656 31528
$ echo $$ $BASHPID
11656 11656
$ echo $$ | while read line; do echo $line $$ $BASHPID; done
11656 11656 31497
$ while read line; do echo $line $$ $BASHPID; done <<< $$
11656 11656 11656

answered Mar 22 '10 at 17:55

Dennis Williamson

346,391
90
374
439

could you please explain what does "<<<" mean? Thanks. – Sopalajo de Arrierez Mar 29 '14 at 20:27
4

It redirects a string into the loop (or anything that reads `stdin`). The string is referred to as a "here string". – Dennis Williamson Mar 29 '14 at 21:29

score 10 · Answer 4 · edited Feb 15 '13 at 21:10

10

The PID is stored in $$.

Example: kill -9 $$ will kill the shell instance it is called from.

edited Feb 15 '13 at 21:10

n.st

946
12
27

answered Mar 22 '10 at 15:55

neo

1,260
11
22

`kill -9` (with `-9` flag) is considered to be harmful and only to be used if it is absolutely necessary). – Willem Van Onsem Sep 15 '15 at 15:01
4

It's considered "dangerous" because the process does not get a chance to respond to the signal (and possibly clean up after itself). Doing `kill -9 $$` does exactly 1 thing. It kills the **current shell process**. This is useful if you have done something in the shell session that you do not want written to `.bash_history` Like: `docker run -e PASSWORD=hunter2 ircbot` – Bruno Bronosky Oct 19 '17 at 04:05

score 8 · Answer 5 · answered Mar 22 '10 at 15:54

8

You can use the $$ variable.

answered Mar 22 '10 at 15:54

Klaus Byskov Pedersen

117,245
29
183
222

Joakim J. Rapp · Answer 6 · 2019-01-25T12:06:32.240

2

If the process is a child process and $BASHPID is not set, it is possible to query the ppid of a created child process of the running process. It might be a bit ugly, but it works. Example:

sleep 1 &
mypid=$(ps -o ppid= -p "$!")

edited Jan 25 '19 at 12:06

answered Jan 25 '19 at 11:58

Joakim J. Rapp

21
2

Pierre13fr · Answer 7 · 2021-11-25T08:34:53.840

Wherever you are (on an interactive command line or in a script), and in the case you do NOT have access to $BASHPID, you can get access to your current shell pid with this :

bash -c 'echo $PPID'

where simple quotes are essential to prevent premature string interpretation (so as to be sure the interpretation occurs in the child process, not in the current one). The principle is just to create a child process and ask for its parent pid, that is to say "myself". This code is simpler than ps-based Joakim solution.

How can a Linux/Unix Bash script get its own PID?

7 Answers7

Linked