После развертывания некоторых экземпляров Apache Kafka на удаленных узлах я обнаружил проблему с kafka-server-stop.sh
script, которая является частью архива Кафки.
По умолчанию он содержит:
#!/bin/sh
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
ps ax | grep -i 'kafka\.Kafka' | grep java | grep -v grep | awk '{print $1}' | xargs kill -SIGTERM
и этот script отлично работает, если я выполняю apache kafka как не фоновый процесс, например:
/var/lib/kafka/bin/kafka-server-start.sh /var/lib/kafka/config/server.properties
также работает, когда я выполняю его как фоновый процесс:
/var/lib/kafka/bin/kafka-server-start.sh /var/lib/kafka/config/server.properties &
но на моих удаленных узлах я выполняю его (с использованием Ansible) с этим python script:
#!/usr/bin/env python
import argparse
import os
import subprocess
KAFKA_PATH = "/var/lib/kafka/"
def execute_command_pipe_output(command_to_call):
return subprocess.Popen(command_to_call, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)
def execute_command_no_output(command_to_call):
with open(os.devnull, "w") as null_file:
return subprocess.Popen(command_to_call, stdout=null_file, stderr=subprocess.STDOUT)
def start_kafka(args):
command_to_call = ["nohup"]
command_to_call += [KAFKA_PATH + "bin/zookeeper-server-start.sh"]
command_to_call += [KAFKA_PATH + "config/zookeeper.properties"]
proc = execute_command_no_output(command_to_call)
command_to_call = ["nohup"]
command_to_call += [KAFKA_PATH + "bin/kafka-server-start.sh"]
command_to_call += [KAFKA_PATH + "config/server.properties"]
proc = execute_command_no_output(command_to_call)
def stop_kafka(args):
command_to_call = [KAFKA_PATH + "bin/kafka-server-stop.sh"]
proc = execute_command_pipe_output(command_to_call)
for line in iter(proc.stdout.readline, b''):
print line,
command_to_call = [KAFKA_PATH + "bin/zookeeper-server-stop.sh"]
proc = execute_command_pipe_output(command_to_call)
for line in iter(proc.stdout.readline, b''):
print line,
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Starting Zookeeper and Kafka instances")
parser.add_argument('action', choices=['start', 'stop'], help="action to take")
args = parser.parse_args()
if args.action == 'start':
start_kafka(args)
elif args.action == 'stop':
stop_kafka(args)
else:
parser.print_help()
после выполнения
manage-kafka.py start
manage-kafka.py stop
Zookeeper отключается (как и должно быть), но Kafka все еще работает.
Что более интересно, когда я вызываю (вручную)
nohup /var/lib/kafka/bin/kafka-server-stop.sh
или
nohup /var/lib/kafka/bin/kafka-server-stop.sh &
kafka-server-stop.sh
корректно завершает экземпляр Kafka. Я подозреваю, что эта проблема может быть вызвана некоторыми проблемами Linux/Python.