记一次logstash-input-jdbc配置文件type冲突血泪史 作者: admin 时间: 2018-06-26 分类: elasticsearch 评论 # 记一次logstash-input-jdbc配置文件type冲突血泪史 ### 起因 统计需求，需要将某几份数据按时间分组进行统计分析，使用MySQL进行分组查询，不只SQL语句繁琐，且效率较慢，按时间分组查询SQL如下： ```sql -- month select CONCAT(YEAR(stat_date),'_',DATE_FORMAT(stat_date,'%m')) months ,sum(total_amount) as count_amount, sum(total_new_user) as count_new_user, sum(da_active_user) as count_active_user from xxx WHERE `stat_date` BETWEEN '2016-01-02' AND '2017-05-30' group by months; -- 季度 select CONCAT(YEAR(stat_date),'_',quarter(stat_date)) qu,sum(total_amount) as count_amount, sum(total_new_user) as count_new_user, sum(da_active_user) as count_active_user from xxx WHERE `stat_date` BETWEEN '2016-01-02' AND '2017-05-30' group by qu; -- 周 select CONCAT(YEAR(stat_date),'_',DATE_FORMAT(stat_date,'%U')) weeks,sum(total_amount) as count_amount, sum(total_new_user) as count_new_user, sum(da_active_user) as count_active_user from xxx WHERE `stat_date` BETWEEN '2016-01-02' AND '2017-05-30' group by weeks; -- 天 select CONCAT(YEAR(stat_date),'_',DATE_FORMAT(stat_date,'%m'),'_',DATE_FORMAT(stat_date,'%d')) days, sum(total_amount) as count_amount, sum(total_new_user) as count_new_user, sum(da_active_user) as count_active_user from xxx WHERE `stat_date` BETWEEN '2016-01-02' AND '2017-05-30' group by days; ``` - 阅读剩余部分 -