Python Forum

I have pyspark script that is working fine. This script will fetch data from mysql and create hive tables in HDFS.

The pyspark script is below.

    #!/usr/bin/env python
    import sys
    from pyspark import SparkContext, SparkConf
    from pyspark.sql import HiveContext
    conf = SparkConf()
    sc = SparkContext(conf=conf)
    sqlContext = HiveContext(sc)

    #Condition to specify exact number of arguments in the spark-submit command line
    if len(sys.argv) != 8:
        print "Invalid number of args......"
        print "Usage: spark-submit import.py Arguments"
        exit()
    table = sys.argv[1]
    hivedb = sys.argv[2]
    domain = sys.argv[3]
    port=sys.argv[4]
    mysqldb=sys.argv[5]
    username=sys.argv[6]
    password=sys.argv[7]

    df = sqlContext.read.format("jdbc").option("url", "{}:{}/{}".format(domain,port,mysqldb)).option("driver", "com.mysql.jdbc.Driver").option("dbtable","{}".format(table)).option("user", "{}".format(username)).option("password", "{}".format(password)).load()

    #Register dataframe as table
    df.registerTempTable("mytempTable")

    # create hive table from temp table:
    sqlContext.sql("create table {}.{} as select * from mytempTable".format(hivedb,table))

    sc.stop()

Now this pyspark script will be invoked by using a shell script. For this shell script I am passing table names as arguments from a file.

The shell script is below.


#!/bin/bash

source /home/$USER/spark/source.sh
[ $# -ne 1 ] && { echo "Usage : $0 table ";exit 1; }

args_file=$1

TIMESTAMP=date "+%Y-%m-%d"
touch /home/$USER/logs/${TIMESTAMP}.success_log
touch /home/$USER/logs/${TIMESTAMP}.fail_log
success_logs=/home/$USER/logs/${TIMESTAMP}.success_log
failed_logs=/home/$USER/logs/${TIMESTAMP}.fail_log

#Function to get the status of the job creation
function log_status
{
status=$1
message=$2
if [ "$status" -ne 0 ]; then
echo "date +\"%Y-%m-%d %H:%M:%S\" [ERROR] $message [Status] $status : failed" | tee -a "${failed_logs}"
#echo "Please find the attached log file for more details"
exit 1
else
echo "date +\"%Y-%m-%d %H:%M:%S\" [INFO] $message [Status] $status : success" | tee -a "${success_logs}"
fi
}
while read -r table ;do
spark-submit --name "${table}" --master "yarn-client" --num-executors 2 --executor-memory 6g --executor-cores 1 --conf "spark.yarn.executor.memoryOverhead=609" /home/$USER/spark/sql_spark.py ${table} ${hivedb} ${domain} ${port} ${mysqldb} ${username} ${password} > /tmp/logging/${table}.log 2>&1
g_STATUS=$?
log_status $g_STATUS "Spark job ${table} Execution"
done < "${args_file}"

echo "************************************************************************************************************************************************************************"

I am able to collect logs for each individual table in the args_file using the above shell script.

Now I have more than 200 tables in mysql. I have modified the pyspark script like below. I have create a function to itreate over the args_file and execute the code.

New spark script

    #!/usr/bin/env python
    import sys
    from pyspark import SparkContext, SparkConf
    from pyspark.sql import HiveContext
    conf = SparkConf()
    sc = SparkContext(conf=conf)
    sqlContext = HiveContext(sc)

    #Condition to specify exact number of arguments in the spark-submit command line
    if len(sys.argv) != 8:
        print "Invalid number of args......"
        print "Usage: spark-submit import.py Arguments"
        exit()
    args_file = sys.argv[1]
    hivedb = sys.argv[2]
    domain = sys.argv[3]
    port=sys.argv[4]
    mysqldb=sys.argv[5]
    username=sys.argv[6]
    password=sys.argv[7]

    def testing(table, hivedb, domain, port, mysqldb, username, password):

        print "*********************************************************table = {} ***************************".format(table)
        df = sqlContext.read.format("jdbc").option("url", "{}:{}/{}".format(domain,port,mysqldb)).option("driver", "com.mysql.jdbc.Driver").option("dbtable","{}".format(table)).option("user", "{}".format(username)).option("password", "{}".format(password)).load()

        #Register dataframe as table
        df.registerTempTable("mytempTable")

        # create hive table from temp table:
        sqlContext.sql("create table {}.{} stored as parquet as select * from mytempTable".format(hivedb,table))

    input = sc.textFile('/user/XXXXXXX/spark_args/%s' %args_file).collect()

    for table in input:
     testing(table, hivedb, domain, port, mysqldb, username, password)

    sc.stop()

Now I want to collect the logs for individual table in args_file. But I am getting only one log file that has the log for all the tables.

How can I achieve my requirement? Or is the method I am doing is completely wrong

> New shell script:

spark-submit --name "${args_file}" --master "yarn-client" --num-executors 2 --executor-memory 6g --executor-cores 1 --conf "spark.yarn.executor.memoryOverhead=609" /home/$USER/spark/sql_spark.py ${table} ${hivedb} ${domain} ${port} ${mysqldb} ${username} ${password} > /tmp/logging/${args_file}.log 2>&

(Aug-28-2017, 06:20 PM)viru Wrote: [ -> ]> /tmp/logging/${args_file}.log 2>&

If you want a different log for each table, why not have the table name in the log file name?