[
https://issues.apache.org/jira/browse/HIVE-18442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16359802#comment-16359802
]
Rui Li commented on HIVE-18442:
-------------------------------
[~xuefuz], my concern about {{spark.yarn.user.classpath.first}} is it's not a
public config. On the other hand, setting {{fs.SCHEME.impl}} in the
configuration is somehow a more "official" solution -- at least the java doc
implies the filesystem binding will be checked in the conf when getting a
FileSystem instance:
https://github.com/apache/hadoop/blob/release-3.0.0-beta1-RC0/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FileSystem.java#L3237
> HoS: No FileSystem for scheme: nullscan
> ---------------------------------------
>
> Key: HIVE-18442
> URL: https://issues.apache.org/jira/browse/HIVE-18442
> Project: Hive
> Issue Type: Bug
> Components: Spark
> Reporter: Rui Li
> Assignee: Rui Li
> Priority: Major
> Attachments: HIVE-18442.1.patch, HIVE-18442.2.patch
>
>
> Hit the issue when I run following query in yarn-cluster mode:
> {code}
> select * from (select key from src where false) a left outer join (select key
> from srcpart limit 0) b on a.key=b.key;
> {code}
> Stack trace:
> {noformat}
> Job failed with java.io.IOException: No FileSystem for scheme: nullscan
> at
> org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2799)
> at
> org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2810)
> at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:100)
> at
> org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:2849)
> at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:2831)
> at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:389)
> at org.apache.hadoop.fs.Path.getFileSystem(Path.java:356)
> at
> org.apache.hadoop.hive.ql.exec.Utilities.isEmptyPath(Utilities.java:2605)
> at
> org.apache.hadoop.hive.ql.exec.Utilities.isEmptyPath(Utilities.java:2601)
> at
> org.apache.hadoop.hive.ql.exec.Utilities$GetInputPathsCallable.call(Utilities.java:3409)
> at
> org.apache.hadoop.hive.ql.exec.Utilities.getInputPaths(Utilities.java:3347)
> at
> org.apache.hadoop.hive.ql.exec.spark.SparkPlanGenerator.cloneJobConf(SparkPlanGenerator.java:299)
> at
> org.apache.hadoop.hive.ql.exec.spark.SparkPlanGenerator.generate(SparkPlanGenerator.java:222)
> at
> org.apache.hadoop.hive.ql.exec.spark.SparkPlanGenerator.generate(SparkPlanGenerator.java:109)
> at
> org.apache.hadoop.hive.ql.exec.spark.RemoteHiveSparkClient$JobStatusJob.call(RemoteHiveSparkClient.java:354)
> at
> org.apache.hive.spark.client.RemoteDriver$JobWrapper.call(RemoteDriver.java:358)
> at
> org.apache.hive.spark.client.RemoteDriver$JobWrapper.call(RemoteDriver.java:323)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748)
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
