What does ampersand mean at the end of a shell script line?
sh sys-snap.sh &
What is sh?
What is sys-snap.sh?
Why I should put & at the end of the line?
Can anyone explain the syntax?
Without the & the script won’t go back to the prompt till I press Ctrl+C.
With & I can press enter and it works.
sh is a shell. Which shell exactly depends on the system. Example for a system that uses bash as its standard shell:
$ ls -l /bin/sh
lrwxrwxrwx 1 root root 4 Mar 13 16:12 /bin/sh -> bash
$ sh --version
GNU bash, version 4.2.45(1)-release (x86_64-pc-linux-gnu)
sys-snap.sh will be a shell script of some kind, so sh sys-snap.sh will execute the shell script.
& will cause the shell process to run in the background. Without & it will stay in the foreground until the script ends. The script will work either way, it’s just a question of waiting for the script to end or not before executing further commands.
sh is the default Bourne-compatible shell (usually bash or dash)
sys-snap.sh is a shell script, which contains commands that sh executes.
As you do not post its content, I can only guess from its name, what it does.
I can find a script related to CPanel with the same file name, that make a log file with all current processes, current memory usage, database status etc.
If the script starts with a shebang line (#!/bin/sh or similar), you can make it executable with chmod +x sys-snap.sh and start it directly by using ./sys-snap.sh if it is in the current directory.
With & the process starts in the background, so you can continue to use the shell and do not have to wait until the script is finished.
If you forget it, you can stop the current running process with Ctrl-Z and continue it in the background with bg (or in the foreground with fg).
For more information, see job control
This is known as job control under unix. The & informs the shell to put the command in the background. This means it continues to run the sys-snap.sh command but returns you to your shell to allows you to continue doing parallel commands.
You can see the list of jobs presently running with the jobs command. You can return to the command (bring to the ‘foreground) with use of the fg command. Which pulls the command back to the state where you see no prompt and you have to issue Ctrl–C to kill the process. You can however suspend (pause) that process, issuing Ctrl–Z. This will pause sys-snap.sh and return you to your prompt. You can then background it (as though you had issued it with the &) with the bg command, and it will resume running from it’s paused state the Ctrl–Z had put it in.
Note that you can have more than one job at a time (as shown by jobs):
[1]- Running sys-snap.sh &
[2]+ Running another-script.sh &
You can background and foreground them using their job number, %1 will be sys-snap.sh and %2 will be another-script.sh. If you use fg or bg without arguments it will action the command on the job marked by + in the jobs output above.
fg %1
will put sys-snap.sh back into the foreground, whilst leaving another-script.sh in the background.
You can issue the Ctrl–C sequence to running jobs without having to foreground them with the kill command, kill %1 will send the equivalent of Ctrl–C to sys-snap.sh.
If you are using bash shell, the man bash command has a detailed section under the section headed ‘JOB CONTROL’ going into more detail.
As to the name of sys-snap.sh, under unix file names are arbitrary (with a couple of exceptions like dynamic loader files). There is no requirement for them to have specific file extentions to make them run as a shell script, invoke other commands such as perl or php etc. It is usually used to provide clarity, that in this instance, it is .sh a Shell Script using the Bourne Shell /bin/sh.
The functional part of sys-snap.sh (when you look at it’s contents with something like the less command) is the Shebang. On the first line you will probably find one of:
#! /bin/sh
#! /bin/bash
#! /usr/local/bin/bash
or similar. In basic terms, a the command after the #! (such as /bin/sh) is ran and the contents of the rest of the script file is fed to it a line at a time. Note that the file must also be set executable with chmod so that you can run it as a command. If the permissions were not set, then the shebang has no effect, because you would either get the error:
bash: sys-snap.sh: command not found
or if you ran it by explicit path ./sys-snap.sh (. meaning the current working directory) you would get:
bash: ./sys-snap.sh: Permission denied
The other alternative is to leave it without execute permissions and explicitly ask /bin/sh to run it:
/bin/sh sys-snap.sh &
Is your server in hostgator?
The script sys-snap.sh is a special script written by hostgator team to collect all the log files and data for the system, so they use its output in monitoring and collecting the system information.