Exit codes in Python
Question:
I got a message saying script xyz.py returned exit code 0
. What does this mean?
What do the exit codes in Python mean? How many are there? Which ones are important?
Answers:
Exit codes of 0 usually mean, “nothing wrong here.” However if the programmer of the script didn’t follow convention you may have to consult the source to see what it means. Usually a non-zero value is returned as an error code.
The exit codes only have meaning as assigned by the script author. The Unix tradition is that exit code 0 means ‘success’, anything else is failure. The only way to be sure what the exit codes for a given script mean is to examine the script itself.
Exit codes in many programming languages are up to programmers. So you have to look at your program source code (or manual). Zero usually means “everything went fine”.
You’re looking for calls to sys.exit()
in the script. The argument to that method is returned to the environment as the exit code.
It’s fairly likely that the script is never calling the exit method, and that 0 is the default exit code.
From the documentation for sys.exit
:
The optional argument arg can be an
integer giving the exit status
(defaulting to zero), or another type
of object. If it is an integer, zero
is considered “successful termination”
and any nonzero value is considered
“abnormal termination” by shells and
the like. Most systems require it to
be in the range 0-127, and produce
undefined results otherwise. Some
systems have a convention for
assigning specific meanings to
specific exit codes, but these are
generally underdeveloped; Unix
programs generally use 2 for command
line syntax errors and 1 for all other
kind of errors.
One example where exit codes are used are in shell scripts. In Bash you can check the special variable $?
for the last exit status:
me@mini:~$ python -c ""; echo $?
0
me@mini:~$ python -c "import sys; sys.exit(0)"; echo $?
0
me@mini:~$ python -c "import sys; sys.exit(43)"; echo $?
43
Personally I try to use the exit codes I find in /usr/include/asm-generic/errno.h
(on a Linux system), but I don’t know if this is the right thing to do.
Operating system commands have exit codes. Look for Linux exit codes to see some material on this. The shell uses the exit codes to decide if the program worked, had problems, or failed. There are some efforts to create standard (or at least commonly-used) exit codes. See this Advanced Shell Script posting.
There is an errno
module that defines standard exit codes:
For example, Permission denied is error code 13:
import errno, sys
if can_access_resource():
do_something()
else:
sys.exit(errno.EACCES)
For the record, you can use POSIX standard exit codes defined here.
Example:
import sys, os
try:
config()
except:
sys.exit(os.EX_CONFIG)
try:
do_stuff()
except:
sys.exit(os.EX_SOFTWARE)
sys.exit(os.EX_OK) # code 0, all ok
If you would like to portably use the standard POSIX exit codes, see the exitstatus
package on PyPI.
Install the package:
$ pip install exitstatus
Use in your code:
import sys
from exitstatus import ExitStatus
sys.exit(ExitStatus.success)
The answer is "Depends on what exit code zero means".
However, in most cases, this means "Everything is Ok".
I like POSIX:
So, in the shell, I would type:
python script.py && echo 'OK' || echo 'Not OK'
If my Python script calls sys.exit(0)
, the shell returns ‘OK’
If my Python script calls sys.exit(1)
(or any non-zero integer), the shell returns ‘Not OK’.
It’s your job to get clever with the shell, and read the documentation (or the source) for your script to see what the exit codes mean.
To allow exception handlers and other things execute I recommend a clean shutdown with the exit(0)
builtin rather than using sys.exit()
which terminates the process abruptly.
I recommend a clean shutdown with the exit(0) builtin rather than using sys.exit()
I’m not sure if this is good advice.
The very documentation mentioned reads:
The site module (which is imported automatically during startup, except if the -S command-line option is given) adds several constants to the built-in namespace. They are useful for the interactive interpreter shell and should not be used in programs.
Then:
exit(code=None)
Objects that when printed, print a message like “Use quit() or Ctrl-D (i.e. EOF) to exit”, and when called, raise SystemExit with the specified exit code.
If we compare it to sys.exit() documentation, it’s using the same mechanism which is raising a SystemExit
exception.
Following Unix exit codes 0 – for success / OK, 1 – non success / error. You could simply use exit(0)
or exit(1)
call without importing sys
module.
WORKAROUND:
To face the bug in gitlab-ci (https://gitlab.com/gitlab-org/gitlab-runner/-/issues/27668), I simply generated a file with the error code:
Python part:
with open("error_code.log", 'w') as file:
file.write('2') # this will trigger a warning
and then in the yml file:
.step_name:
variables:
# my vars
script:
- python my_code.py arg1 arg2
- exit `cat error_code.log`
allow_failure:
exit_codes:
- 2
I got a message saying script xyz.py returned exit code 0
. What does this mean?
What do the exit codes in Python mean? How many are there? Which ones are important?
Exit codes of 0 usually mean, “nothing wrong here.” However if the programmer of the script didn’t follow convention you may have to consult the source to see what it means. Usually a non-zero value is returned as an error code.
The exit codes only have meaning as assigned by the script author. The Unix tradition is that exit code 0 means ‘success’, anything else is failure. The only way to be sure what the exit codes for a given script mean is to examine the script itself.
Exit codes in many programming languages are up to programmers. So you have to look at your program source code (or manual). Zero usually means “everything went fine”.
You’re looking for calls to sys.exit()
in the script. The argument to that method is returned to the environment as the exit code.
It’s fairly likely that the script is never calling the exit method, and that 0 is the default exit code.
From the documentation for sys.exit
:
The optional argument arg can be an
integer giving the exit status
(defaulting to zero), or another type
of object. If it is an integer, zero
is considered “successful termination”
and any nonzero value is considered
“abnormal termination” by shells and
the like. Most systems require it to
be in the range 0-127, and produce
undefined results otherwise. Some
systems have a convention for
assigning specific meanings to
specific exit codes, but these are
generally underdeveloped; Unix
programs generally use 2 for command
line syntax errors and 1 for all other
kind of errors.
One example where exit codes are used are in shell scripts. In Bash you can check the special variable $?
for the last exit status:
me@mini:~$ python -c ""; echo $?
0
me@mini:~$ python -c "import sys; sys.exit(0)"; echo $?
0
me@mini:~$ python -c "import sys; sys.exit(43)"; echo $?
43
Personally I try to use the exit codes I find in /usr/include/asm-generic/errno.h
(on a Linux system), but I don’t know if this is the right thing to do.
Operating system commands have exit codes. Look for Linux exit codes to see some material on this. The shell uses the exit codes to decide if the program worked, had problems, or failed. There are some efforts to create standard (or at least commonly-used) exit codes. See this Advanced Shell Script posting.
There is an errno
module that defines standard exit codes:
For example, Permission denied is error code 13:
import errno, sys
if can_access_resource():
do_something()
else:
sys.exit(errno.EACCES)
For the record, you can use POSIX standard exit codes defined here.
Example:
import sys, os
try:
config()
except:
sys.exit(os.EX_CONFIG)
try:
do_stuff()
except:
sys.exit(os.EX_SOFTWARE)
sys.exit(os.EX_OK) # code 0, all ok
If you would like to portably use the standard POSIX exit codes, see the exitstatus
package on PyPI.
Install the package:
$ pip install exitstatus
Use in your code:
import sys
from exitstatus import ExitStatus
sys.exit(ExitStatus.success)
The answer is "Depends on what exit code zero means".
However, in most cases, this means "Everything is Ok".
I like POSIX:
So, in the shell, I would type:
python script.py && echo 'OK' || echo 'Not OK'
If my Python script calls sys.exit(0)
, the shell returns ‘OK’
If my Python script calls sys.exit(1)
(or any non-zero integer), the shell returns ‘Not OK’.
It’s your job to get clever with the shell, and read the documentation (or the source) for your script to see what the exit codes mean.
To allow exception handlers and other things execute I recommend a clean shutdown with the exit(0)
builtin rather than using sys.exit()
which terminates the process abruptly.
I recommend a clean shutdown with the exit(0) builtin rather than using sys.exit()
I’m not sure if this is good advice.
The very documentation mentioned reads:
The site module (which is imported automatically during startup, except if the -S command-line option is given) adds several constants to the built-in namespace. They are useful for the interactive interpreter shell and should not be used in programs.
Then:
exit(code=None)
Objects that when printed, print a message like “Use quit() or Ctrl-D (i.e. EOF) to exit”, and when called, raise SystemExit with the specified exit code.
If we compare it to sys.exit() documentation, it’s using the same mechanism which is raising a SystemExit
exception.
Following Unix exit codes 0 – for success / OK, 1 – non success / error. You could simply use exit(0)
or exit(1)
call without importing sys
module.
WORKAROUND:
To face the bug in gitlab-ci (https://gitlab.com/gitlab-org/gitlab-runner/-/issues/27668), I simply generated a file with the error code:
Python part:
with open("error_code.log", 'w') as file:
file.write('2') # this will trigger a warning
and then in the yml file:
.step_name:
variables:
# my vars
script:
- python my_code.py arg1 arg2
- exit `cat error_code.log`
allow_failure:
exit_codes:
- 2