infrastructure-provisioning/src/general/scripts/os/common_configure_spark.py

#!/usr/bin/python3

# *****************************************************************************
#
# 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.
#
# ******************************************************************************

import argparse
import ast
import os
import sys
import time
from datalab.fab import *
from datalab.notebook_lib import *
from fabric import *

parser = argparse.ArgumentParser()
parser.add_argument('--hostname', type=str, default='')
parser.add_argument('--keyfile', type=str, default='')
parser.add_argument('--os_user', type=str, default='')
parser.add_argument('--spark_conf', type=str, default='')
parser.add_argument('--cluster_name', type=str, default='')
args = parser.parse_args()


def update_spark_defaults_conf(spark_conf):
    try:
        timestamp = time.strftime("%a, %d %b %Y %H:%M:%S %Z", time.gmtime())
        configs = conn.sudo('find /opt/ /etc/ /usr/lib/ -name spark-defaults.conf -type f').stdout.split('\n')
        for conf in filter(None, configs):
            conn.sudo('''sed -i '/^# Updated/d' {0}'''.format(conf))
            conn.sudo('''echo "# Updated by DATALAB at {0} >> {1}'''.format(timestamp, conf))
    except Exception as err:
        print('Error: {0}'.format(err))
        sys.exit(1)


def add_custom_spark_properties(cluster_name):
    try:
        if os.path.exists('/opt/{0}'.format(cluster_name)):
            datalab_header = conn.sudo('cat /tmp/{0}/notebook_spark-defaults_local.conf | grep "^#"'.format(cluster_name)).stdout
            spark_configurations = ast.literal_eval(os.environ['spark_configurations'])
            new_spark_defaults = list()
            spark_defaults = conn.sudo('cat /opt/{0}/spark/conf/spark-defaults.conf'.format(cluster_name)).stdout
            current_spark_properties = spark_defaults.split('\n')
            for param in current_spark_properties:
                if param.split(' ')[0] != '#':
                    for config in spark_configurations:
                        if config['Classification'] == 'spark-defaults':
                            for property in config['Properties']:
                                if property == param.split(' ')[0]:
                                    param = property + ' ' + config['Properties'][property]
                                else:
                                    new_spark_defaults.append(property + ' ' + config['Properties'][property])
                    new_spark_defaults.append(param)
            new_spark_defaults = set(new_spark_defaults)
            conn.sudo("echo '{0}' > /opt/{1}/spark/conf/spark-defaults.conf".format(datalab_header, cluster_name))
            for prop in new_spark_defaults:
                prop = prop.rstrip()
                conn.sudo('echo "{0}" >> /opt/{1}/spark/conf/spark-defaults.conf'.format(prop, cluster_name))
            conn.sudo('sed -i "/^\s*$/d" /opt/{0}/spark/conf/spark-defaults.conf'.format(cluster_name))
    except Exception as err:
        print('Error: {0}'.format(err))
        sys.exit(1)


if __name__ == "__main__":
    print('Configure connections')
    global conn
    conn = datalab.fab.init_datalab_connection(args.hostname, args.os_user, args.keyfile)

    if (args.spark_conf != ''):
        update_spark_defaults_conf(args.spark_conf)

    update_spark_jars()

    if 'spark_configurations' in os.environ:
        add_custom_spark_properties(args.cluster_name)

    conn.close()