hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bharath R (JIRA)" <>
Subject [jira] Commented: (HIVE-1872) Hive process is exiting on executing ALTER query
Date Thu, 27 Jan 2011 06:09:44 GMT


Bharath R  commented on HIVE-1872:

Based on the thread  HIVE-549 ,  there was a comment to cleanup the running jobs if one of
them have failed. 

	* Shall we immediately stop all other running jobs if one of them have failed?
		+ console.printError(errorMessage);
		+ taskCleanup(runnable); 

This comment was implemented by calling System.exit(9). This makes the JVM to exit. 

Rather than relying on the ShutdownHook , we could directly invoke that functionality in the

As per my proposed solution , in taskCleanup. 

                 1)  Sending an interrupt to runningJobs "TaskRunner" so that the task associated
will be interrupted. 
                 2)  And  if the task is "MapRed" task, then the task cleanup should post
the "kill url" from "runningJobKillURIs" hashmap in ExecDriver



> Hive process is exiting on executing ALTER query
> ------------------------------------------------
>                 Key: HIVE-1872
>                 URL:
>             Project: Hive
>          Issue Type: Bug
>          Components: CLI, Server Infrastructure
>    Affects Versions: 0.6.0
>         Environment: SUSE Linux Enterprise Server 10 SP2 (i586) - Kernel
> Hadoop 0.20.1
> Hive 0.6.0
>            Reporter: Bharath R 
>            Assignee: Bharath R 
>         Attachments: HIVE-1872.1.patch
> Hive process is exiting on executing the below queries in the same order as mentioned
> 1) CREATE TABLE SAMPLETABLE(IP STRING , showtime BIGINT ) partitioned by (ds string,ipz
> 2) ALTER TABLE SAMPLETABLE add Partition(ds='sf') location '/user/hive/warehouse' Partition(ipz=100)
location '/user/hive/warehouse'
> After the second query execution , the hive throws the below exception and exiting the
> 10:09:03 ERROR exec.DDLTask: FAILED: Error in metadata: table is partitioned but partition
spec is not specified or tab: {ipz=100}
> org.apache.hadoop.hive.ql.metadata.HiveException: table is partitioned but partition
spec is not specified or tab: {ipz=100}
>         at org.apache.hadoop.hive.ql.metadata.Table.isValidSpec(
>         at org.apache.hadoop.hive.ql.metadata.Hive.getPartition(
>         at org.apache.hadoop.hive.ql.exec.DDLTask.addPartition(
>         at org.apache.hadoop.hive.ql.exec.DDLTask.execute(
>         at org.apache.hadoop.hive.ql.exec.Task.executeTask(
>         at org.apache.hadoop.hive.ql.exec.TaskRunner.runSequential(
>         at org.apache.hadoop.hive.ql.Driver.launchTask(
>         at org.apache.hadoop.hive.ql.Driver.execute(
>         at
>         at org.apache.hadoop.hive.service.HiveServer$HiveServerHandler.execute(
>         at org.apache.hadoop.hive.service.ThriftHive$Processor$execute.process(
>         at org.apache.hadoop.hive.service.ThriftHive$Processor.process(
>         at org.apache.thrift.server.TThreadPoolServer$
>         at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(
>         at java.util.concurrent.ThreadPoolExecutor$
>         at
> As the "alter" query is incorrect the exception was thrown, ideally it should be "ALTER
TABLE SAMPLETABLE add Partition(ds='sf',ipz=100) location '/user/hive/warehouse'". 
> It is not good to exit the HIVE process when the query is incorrect.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message