Mailing-List: contact issues-help@drill.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@drill.apache.org
Date: Fri, 24 Feb 2017 05:57:44 +0000 (UTC)
From: "Usman Shahid (JIRA)" <jira@apache.org>
To: issues@drill.apache.org
Message-ID: <JIRA.13044786.1487671928000.52279.1487915864087@Atlassian.JIRA>
In-Reply-To: <JIRA.13044786.1487671928000@Atlassian.JIRA>
References: <JIRA.13044786.1487671928000@Atlassian.JIRA> <JIRA.13044786.1487671928607@jira-lw-us.apache.org>
Subject: [jira] [Commented] (DRILL-5280) Error during udf area creation due
 to NullPointerException during drillbit startup
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Fri, 24 Feb 2017 05:57:49 -0000


    [ https://issues.apache.org/jira/browse/DRILL-5280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15882030#comment-15882030 ] 

Usman Shahid commented on DRILL-5280:
-------------------------------------

[~arina]
It works perfectly with sqlline.bat, but I am running drill in distributed mode. 
The default way to run in the distributed mode is to use drillbit.sh which is essentially a shell script and I am a windows user. Using Cygwin to execute this shell script yields:
{code}
C:\apache-drill-1.9.0/bin/runbit: line 107: exec: c:\Program: not found
{code}

So currently I am directly running Drillbit from a debugger (Eclipse). I believe a windows batch job for running is required to set the HADOOP_HOME and run drill in distributed mode.

> Error during udf area creation due to NullPointerException during drillbit startup
> ----------------------------------------------------------------------------------
>
>                 Key: DRILL-5280
>                 URL: https://issues.apache.org/jira/browse/DRILL-5280
>             Project: Apache Drill
>          Issue Type: Bug
>         Environment: Windows 7 64-bit
> Java(TM) SE Runtime Environment (build 1.8.0_111-b14)
>            Reporter: Usman Shahid
>
> I tried to start drill in distributed mode but ran into this:
> {code}Exception in thread "main" org.apache.drill.exec.exception.DrillbitStartupException: Failure during initial startup of Drillbit.
> 	at org.apache.drill.exec.server.Drillbit.start(Drillbit.java:296)
> 	at org.apache.drill.exec.server.Drillbit.start(Drillbit.java:272)
> 	at org.apache.drill.exec.server.Drillbit.main(Drillbit.java:268)
> Caused by: org.apache.drill.common.exceptions.DrillRuntimeException: Error during udf area creation [/C:/Users/Usman.Shahid/drill/udf/registry] on file system [file:///]
> 	at org.apache.drill.common.exceptions.DrillRuntimeException.format(DrillRuntimeException.java:49)
> 	at org.apache.drill.exec.expr.fn.registry.RemoteFunctionRegistry.createArea(RemoteFunctionRegistry.java:251)
> 	at org.apache.drill.exec.expr.fn.registry.RemoteFunctionRegistry.prepareAreas(RemoteFunctionRegistry.java:208)
> 	at org.apache.drill.exec.expr.fn.registry.RemoteFunctionRegistry.init(RemoteFunctionRegistry.java:105)
> 	at org.apache.drill.exec.server.Drillbit.run(Drillbit.java:126)
> 	at org.apache.drill.exec.server.Drillbit.start(Drillbit.java:292)
> 	... 2 more
> Caused by: java.lang.NullPointerException
> 	at java.lang.ProcessBuilder.start(ProcessBuilder.java:1012)
> 	at org.apache.hadoop.util.Shell.runCommand(Shell.java:483)
> 	at org.apache.hadoop.util.Shell.run(Shell.java:456)
> 	at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:722)
> 	at org.apache.hadoop.util.Shell.execCommand(Shell.java:815)
> 	at org.apache.hadoop.util.Shell.execCommand(Shell.java:798)
> 	at org.apache.hadoop.fs.FileUtil.execCommand(FileUtil.java:1097)
> 	at org.apache.hadoop.fs.RawLocalFileSystem$DeprecatedRawLocalFileStatus.loadPermissionInfo(RawLocalFileSystem.java:654)
> 	at org.apache.hadoop.fs.RawLocalFileSystem$DeprecatedRawLocalFileStatus.getPermission(RawLocalFileSystem.java:629)
> 	at org.apache.drill.exec.expr.fn.registry.RemoteFunctionRegistry.createArea(RemoteFunctionRegistry.java:232)
> 	... 6 more
> {code}
> The complete stack trace for the `NullPointerException` is:
> {code}
> Thread [main] (Suspended (exception NullPointerException))	
> 	owns: Object  (id=57)	
> 	ProcessBuilder.start() line: 1012	
> 	Shell$ShellCommandExecutor(Shell).runCommand() line: 483	
> 	Shell$ShellCommandExecutor(Shell).run() line: 456	
> 	Shell$ShellCommandExecutor.execute() line: 722	
> 	Shell.execCommand(Map<String,String>, String[], long) line: 815	
> 	Shell.execCommand(String...) line: 798	
> 	FileUtil.execCommand(File, String...) line: 1097	
> 	RawLocalFileSystem$DeprecatedRawLocalFileStatus.loadPermissionInfo() line: 654	
> 	RawLocalFileSystem$DeprecatedRawLocalFileStatus.getPermission() line: 629	
> 	RemoteFunctionRegistry.createArea(FileSystem, String, String) line: 232	
> 	RemoteFunctionRegistry.prepareAreas(DrillConfig) line: 208	
> 	RemoteFunctionRegistry.init(DrillConfig, PersistentStoreProvider, ClusterCoordinator) line: 105	
> 	Drillbit.run() line: 126	
> 	Drillbit.start(DrillConfig, RemoteServiceSet) line: 292	
> 	Drillbit.start(StartupOptions) line: 272	
> 	Drillbit.main(String[]) line: 268	
> {code}
> To further pinpoint it, the call to _Shell.getGetPermissionCommand()_ at _RawLocalFileSystem$DeprecatedRawLocalFileStatus.loadPermissionInfo()_ returns a new String[] { WINUTILS, "ls", "-F" }. It's this WINUTILS that's null because my system doesn't have hadoop installed (nor do I intend to use it). 
> The shortest workaround to this is to have a _HADOOP_HOME_ environment variable pointing towards the _winutils.exe_ bin directory. 


--
This message was sent by Atlassian JIRA
(v6.3.15#6346)