Hi Shylin,
I think the description part in https://issues.apache.org/jira/browse/KYLIN-4153 is clear enough, you can find the root cause and workaround there, it happened due to the lack of atomicity during the rollback for the "big resource". As a workaround, you may delete the remaining marker in your resource store, but it’s more advised to do an upgrade for your Kylin. Best regards, Ni Chunen / George On 02/16/2020 16:22，shylinzhang<[email protected]> wrote： Dear all, The cube is ok when 2:00 am,but after 4:30 am it always shows error ,it shows ``` #4 Step Name: Build Dimension Dictionary org.apache.kylin.engine.mr.exception.HadoopShellException: java.lang.RuntimeException: java.io.IOException: Failed to read big resource /dict/KYLIN.HIVE_METRICS_QUERY_CUBE_QA/CUBE_NAME/b9bb85ec-b72e-4df1-9ee7-e8c971ec5c72.dict at org.apache.kylin.dict.DictionaryManager.getDictionaryInfo(DictionaryManager.java:108) at org.apache.kylin.dict.DictionaryManager.checkDupByContent(DictionaryManager.java:173) at org.apache.kylin.dict.DictionaryManager.trySaveNewDict(DictionaryManager.java:151) at org.apache.kylin.dict.DictionaryManager.saveDictionary(DictionaryManager.java:320) at org.apache.kylin.cube.CubeManager$DictionaryAssist.saveDictionary(CubeManager.java:1117) at org.apache.kylin.cube.CubeManager.saveDictionary(CubeManager.java:1079) at org.apache.kylin.cube.cli.DictionaryGeneratorCLI.processSegment(DictionaryGeneratorCLI.java:74) at org.apache.kylin.cube.cli.DictionaryGeneratorCLI.processSegment(DictionaryGeneratorCLI.java:55) at org.apache.kylin.engine.mr.steps.CreateDictionaryJob.run(CreateDictionaryJob.java:73) at org.apache.kylin.engine.mr.MRUtil.runMRJob(MRUtil.java:93) at org.apache.kylin.engine.mr.common.HadoopShellExecutable.doWork(HadoopShellExecutable.java:63) at org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:166) at org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:71) at org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:166) at org.apache.kylin.job.impl.threadpool.DistributedScheduler$JobRunner.run(DistributedScheduler.java:110) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748) ``` What I do: 1. I disable the cube and purge the cube and rebulid, it also show the same error; 2. Because the cube at 2:00 am is ok, so I use the meta_backs in 2:00 am and restore and cube; I try short period to build the cube, 2:00 am~3:00am~4:00am~4:00am~6:00am are ok, but in 6:am~7:00 am it also shows above error; I found the solution in kylin jira, it tells me to update to release 2.6.4; our kylin clusters are used in 2019-06; just today shows the error. Here’s the jira url: https://issues.apache.org/jira/browse/KYLIN-4153?jql=summary%20~%20%22Failed%20to%20read%20big%20resource*%22%20OR%20description%20~%20%22Failed%20to%20read%20big%20resource*%22%20ORDER%20BY%20lastViewed%20DESC Therefore I want to know why it occured and if there is a way to fix it except updating. I am looking forward for your replay, thank you. Kylin version: 2.6.1 Hadoop version: 3.0.0+cdh6.0.0 Best Wishes! Shylin Zhang | | shylinzhang | | [email protected] | 签名由网易邮箱大师定制
