Liu Zhao
Liu Zhao
## Proposed changes Describe the big picture of your changes here to communicate to the maintainers why we should accept this pull request. If it fixes a bug or resolves...
## Proposed changes Describe the big picture of your changes here to communicate to the maintainers why we should accept this pull request. If it fixes a bug or resolves...
## Proposed changes Describe the big picture of your changes here to communicate to the maintainers why we should accept this pull request. If it fixes a bug or resolves...
## Proposed changes log print format ## Branch to commit - [ ] Branch **kylin3** for v2.x to v3.x - [x] Branch **kylin4** for v4.x - [ ] Branch **kylin5**...
## Proposed changes 在 kylin4 的StorageCleanupJob 中增加对无引用的 cube_stataistics 数据清理,我认为是有意义的: 1. 可以降低无用数据占用的存储空间,同时避免过多无用小文件对nn的压力 2. 默认情况下清理无引用的 cube_stataistics 数据,但可以通过 -cleanupCubeStatistics false 禁用 ## Branch to commit - [ ] Branch **kylin3** for v2.x to...
## Proposed changes 1.初始发现:线上告警某节点存在大量的CLOSE_WAIT,通过 netstat -anp 发现来自于Kylin4 JobServer 进程,CLOSE_WAIT数达到9000多。并且 CLOSE_WAIT 来自的外部地址端口都是 50010,而该端口是 Hadoop DataNode 数据传输使用,故此怀疑是 JobServer在每次作业构建时 fileSystem.open() 一个流后没有进行close。 2.模拟复现:在研测环境提交cube构建任务,并观察 CLOSE_WAIT 数及增长情况,发现每次cube构建结束后,CLOSE_WAIT 数增加1,至此可以确定是JobServer代码中未关闭流导致。 3.定位代码:深入kylin4 构建代码进行debug,最终定位到 org.apache.kylin.engine.spark.utils.UpdateMetadataUtil#syncLocalMetadataToRemote 94行 (Apache Kylin main分支) String...
## Proposed changes 发现kylin4在build job后都会重新new Broadcaster,Broadcaster构造器中会创建一个线程池然后执行任务不停的take广播事件。这个会导致每次构建完job都会创建线程,同时线程也会阻塞在takeFirst处,时间推移服务会不可用。      ## Branch to commit - [ ] Branch **kylin3** for v2.x to v3.x - [x] Branch **kylin4** for v4.x...
## Proposed changes 创建model时在partition部分指定了天分区和时分区,格式分别为:yyyy-MM-dd、HH,在根据天分区值查询时结果非预期。原因是这种场景下segment裁剪有bug。 --不加天分区过滤得到的结果(18号3条数据,19号5条数据)  --加天分区在修复前查询结果 --18号实际有3条数据  --19号实际有5条数据  --18号19号共有8条数据  --加天分区在修复后查询结果    ## Branch to commit - [ ] Branch **kylin3** for v2.x to v3.x...
## Proposed changes 对已修复的bug,选用更合理的实现方式. ## Branch to commit - [ ] Branch **kylin3** for v2.x to v3.x - [x] Branch **kylin4** for v4.x - [ ] Branch **kylin5** for v5.x...
## Proposed changes statement "select name, replace(name, substring(name, 1, 1), '--') as new_name from LZ_TEST_YUFA " query failed, errorMsg:java.lang.ClassCastException: org.apache.spark.sql.Column cannot be cast to java.lang.String while executing SQL: "select *...