kylin-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Na Zhai <na.z...@kyligence.io>
Subject 答复: Error (2.5.2) - fail to find the statistics file in base dir (Step 5: Save Cuboid Statistics)
Date Thu, 06 Dec 2018 04:38:44 GMT
Hi, Jon Shoberg.
Maybe you can try hive 1.x. Can you check the log of HDFS and hive? Maybe there will be the
error message.

发送自 Windows 10 版邮件<https://go.microsoft.com/fwlink/?LinkId=550986>应用

________________________________
发件人: Jon Shoberg <jon.shoberg@gmail.com>
发送时间: Wednesday, December 5, 2018 2:23:44 PM
收件人: user@kylin.apache.org
主题: Error (2.5.2) - fail to find the statistics file in base dir (Step 5: Save Cuboid
Statistics)

I'm getting the following error message below when reaching Step 5: Save Cuboid Statistics

Any ideas or suggestions? Below is the error and after are steps I've tried to fix.

java.io.IOException: fail to find the statistics file in base dir: hdfs://192.168.1.20:9000/kylin/kylin_metadata/kylin-46adf439-7f25-91fa-a3cf-a7c27732e77c/HoldingNodeCube/fact_distinct_columns/statistics<http://192.168.1.20:9000/kylin/kylin_metadata/kylin-46adf439-7f25-91fa-a3cf-a7c27732e77c/HoldingNodeCube/fact_distinct_columns/statistics>
at org.apache.kylin.engine.mr.steps.SaveStatisticsStep.doWork(SaveStatisticsStep.java:78)
at org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:164)
at org.apache.kylin.job.execution.DefaultChainedExecutable.doWork(DefaultChainedExecutable.java:70)
at org.apache.kylin.job.execution.AbstractExecutable.execute(AbstractExecutable.java:164)
at org.apache.kylin.job.impl.threadpool.DefaultScheduler$JobRunner.run(DefaultScheduler.java:114)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

The system is Centos 7, Hadoop 2.8.5, Hive 2.3.4, HBase 1.4.8, Kylin 2.5.2 - all installed
from tar

DFS/Yarn all seem to be working appropriately.  No DFS errors and is not in safe mode.

Hive is working correctly and source tables are populated with data and SQL tests seem OK.

HBase seems working OK. Its only installed/used for Kylin so even after re-installing Kylin
had installed its meta-data just fine.

Kylin's web UI comes up OK. The data model, cube, and processing are all done via the Web
UI and no errors other than the one above.

I looked at the GITHUB sources where the error message comes from and its not creating the
statistics directory and cannot get the statistics files (they're not created)

In the DFS/Hadoop logs I see some messages around statistics and the job ID in the HDFS URL
but nothing that shows me where to fix something.

Any thoughts, ideas, or experiences to share would be greatly appreciated!

J

Mime
View raw message