[Python] How to kill zombie process with ppid = 1?

root@9948254b178c:/usr/src/app# ps -ef
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 01:42 ? 00:00:00 node app.js
root 35 1 0 01:43 ? 00:00:02 [javac] <defunct>
root 51 1 0 01:43 ? 00:00:00 [g++] <defunct>
root 66 0 0 01:43 pts/0 00:00:00 bash
root 73 66 0 01:43 pts/0 00:00:00 top
root 77 66 0 01:45 pts/0 00:00:00 top
root 79 66 0 01:51 pts/0 00:00:00 top
root 83 66 0 01:54 pts/0 00:00:00 ps -ef


In the table we can see that there are two zombie processes with PID 35 and 51. How could I remove it without stopping the parent (PID=1).

Because the parent is the whole system.

Here is a python script I am running for compiling and running different languages. (C/C++, Python3, Java)

Note: The script for python does not create any zombie process. But the other does. I am running nodejs ecec to run this python script.

import sys
import subprocess
import re

# "python3 run.py "+ json_msg.filename +" "+extensions[json_msg.lang]+" "+json_msg.timeout
filename = str(sys.argv[1])
extension = str(sys.argv[2])
timeout = str(sys.argv[3])


# for java
def changing_class_name():

file = open(f"temp/{filename}.java" , 'r')

newfile = filename
new_file_content = str()

for line in file:
if re.search('^class ',line):
line = line.strip()
ls = line.split(" ")

# print(ls)
string = ls[1]
if(string[-1] == '{'):
new_line = line.replace(string , newfile+'{')
new_file_content += new_line + "\n"

elif(string[-1]!='{'):

new_line = line.replace(string , newfile)
new_file_content += new_line + "\n"

else:
new_file_content += line


writing_file = open(f"temp/{filename}.java", "w")
writing_file.write(new_file_content)
writing_file.close()



status = True

# we have to get the input file first
try:
inputfile = subprocess.run(f"cd temp/ && cat input.txt" ,shell=True , stdout=subprocess.PIPE,stderr=subprocess.STDOUT , timeout=int("5"))

except :
result = 'Something went wrong while reading input file'
status = False
# print('Something went wrong while reading input file')


# for compiling the file
if(status):
try:
if(extension == "cpp" or extension == "c"):
comp = subprocess.run(f"cd temp/ && g++ {filename}.{extension} -o {filename}" ,shell=True , stdout=subprocess.PIPE,stderr=subprocess.STDOUT , timeout=int("5"))
if(comp.stdout.decode()):
result = comp.stdout.decode()
status = False

if(extension == "java"):
changing_class_name()
comp = subprocess.run(f"cd temp/ && javac {filename}.java" ,shell=True , stdout=subprocess.PIPE,stderr=subprocess.STDOUT ,timeout=5 )

if(comp.stdout.decode()):
result = comp.stdout.decode()
status = False

except:
result = 'Something went wrong while compiling the file'
status = False


# running the file
if(status):
try:
if(extension == "py"):
output = subprocess.run(f"cd temp/ && python3 {filename}.{extension}" ,shell=True , stdout=subprocess.PIPE, stderr=subprocess.PIPE , input=(inputfile.stdout.decode()).encode() , timeout=int(timeout))
result = output.stdout.decode()

elif(extension=="cpp" or extension == "c"):
output = subprocess.run(f"cd temp/ && ./{filename}" ,shell=True , stdout=subprocess.PIPE,stderr=subprocess.PIPE , input=(inputfile.stdout.decode()).encode() , timeout=int(timeout))
result = output.stdout.decode()

elif(extension == "java"):
output = subprocess.run(f"cd temp/ && java {filename}" ,shell=True , stdout=subprocess.PIPE,stderr=subprocess.PIPE , input=(inputfile.stdout.decode()).encode() , timeout=int(timeout))
result = output.stdout.decode()

# if there is any error we will also add the error
if(output.stderr.decode() != ""):
result += output.stderr.decode()
status = False


except Exception as e:
result = "Time limit exceeded"
status = False
# print(e)



# getting the result and writting it on output.txt
file = open("./temp/output.txt" , "w")
file.write(result)
file.close()


if(status == True):
print('Successful' ,end="")
else:
print("Failed",end="")

Continue reading...