1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
#!/bin/sh
# # # # # # # # # # # # # # # # # # # # # #
# @author : wangjf
# @date : 2019-08-28 18:06:59
# # # # # # # # # # # # # # # # # # # # # #
source ../dmp_env.sh
LOG_TIME=$(date -d "$ScheduleTime" +"%Y%m%d")
date_path=$(date -d "$ScheduleTime" +"%Y/%m/%d")
dt_four_days=$(date -d "$ScheduleTime 4 days ago" +"%Y/%m/%d")
OUTPUT_PATH="${IQiYi_LAHUO_DAILY_TMP_PATH}/$date_path"
spark-submit --class mobvista.dmp.datasource.iqiyi.IQiYiLaHuoDF \
--name "IQiYiLaHuoDF.${LOG_TIME}" \
--conf spark.sql.shuffle.partitions=1000 \
--conf spark.default.parallelism=1000 \
--conf spark.kryoserializer.buffer.max=256m \
--conf spark.driver.extraJavaOptions="-XX:+UseG1GC" \
--conf spark.executor.extraJavaOptions="-XX:+UseG1GC" \
--files /data/hadoop-config/command-home/hdp-spark-2.3.1-offline/conf/hive-site.xml \
--master yarn --deploy-mode cluster --executor-memory 4g --driver-memory 4g --executor-cores 3 --num-executors 10 \
../${JAR} -date ${LOG_TIME} -partNum 10 -output ${OUTPUT_PATH} -cluster 'cluster_1st'
if [[ $? -ne 0 ]];then
exit 255
fi
# 爱奇艺拉活是采用近四天的拉活结果入库，当天可拉活设备不到四天前的二分之一，就默认当天拉活出现问题，可能是接口访问为false，于是用四天前的数据覆盖
today_count=$(hadoop fs -text ${OUTPUT_PATH}/*|grep '"status":1'|wc -l)
FOUR_DAYS_OUTPUT_PATH="${IQiYi_LAHUO_DAILY_TMP_PATH}/$dt_four_days"
four_days_count=$(hadoop fs -text ${FOUR_DAYS_OUTPUT_PATH}/*|grep '"status":1'|wc -l)
four_days_count=`expr $four_days_count / 2`
if [[ $today_count -lt $four_days_count ]]
then
hadoop distcp -m 20 -overwrite ${FOUR_DAYS_OUTPUT_PATH} ${OUTPUT_PATH}
fi