#!/usr/bin/env bash source ../../dmp_env.sh source ././../../ga_rawdata_analysis/common/tools.sh dt_today=$(date -d "$ScheduleTime tomorrow" +"%Y%m%d") dt_oneday_ago=$(date -d "$ScheduleTime 1 days ago" +"%Y%m%d") dt_slash_today=$(date -d "$ScheduleTime tomorrow" +"%Y/%m/%d") hour="01" dt_slash_oneday=$(date -d "$ScheduleTime 1 days ago" +"%Y/%m/%d") expire_date=$(date -d "$ScheduleTime 61 days ago" +"%Y%m%d") expire_date_path=$(date -d "$ScheduleTime 61 days ago" +"%Y/%m/%d") check_await "${DM_INSTALL_LIST}_v2/${dt_slash_oneday}/dsp_req/_SUCCESS" OUTPUT_PATH01="${ALIPAY_ACTIVATION_DAILY_PATH}/${dt_slash_today}/${hour}/other_data/foractivation_l00016_alipayoppo" OUTPUT_PATH02="${ALIPAY_ACTIVATION_DAILY_PATH}/${dt_slash_today}/${hour}/other_data/foractivation_l00016_alipaybes" OUTPUT_PATH="${ALIPAY_ACTIVATION_DAILY_PATH}/${dt_slash_today}/${hour}/other_data" hadoop fs -rm -r "${OUTPUT_PATH}" spark-submit --class mobvista.dmp.datasource.taobao.AlipayOtherDataToDmp \ --conf spark.network.timeout=720s \ --conf spark.default.parallelism=3000 \ --conf spark.sql.shuffle.partitions=3000 \ --conf spark.sql.broadcastTimeout=1200 \ --conf spark.sql.autoBroadcastJoinThreshold=31457280 \ --master yarn --deploy-mode cluster --executor-memory 8g --driver-memory 4g --executor-cores 5 --num-executors 140 \ ../../${JAR} \ -output01 ${OUTPUT_PATH01} -output02 ${OUTPUT_PATH02} \ -dt_today ${dt_today} -dt_oneday_ago ${dt_oneday_ago} -hour ${hour} if [ $? -ne 0 ];then exit 255 fi mount_partition "etl_alipay_activation_daily" "dt='${dt_today}', hh='${hour}', business='other_data'" "${OUTPUT_PATH}" #HIVE_CMD=$(hive_func) #$HIVE_CMD -v -hivevar dt_today ${dt_today} hour ${hour} -f alipay_other_data_to_dmp.sql if [ $? -ne 0 ];then exit 255 fi hadoop fs -touchz "${ALIPAY_ACTIVATION_DAILY_PATH}/${dt_slash_today}/${hour}/_SUCCESS"