[SPARK-19312] Spark gives wrong error message when failes to create file due to hdfs quota limit.

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Not A Problem
• Affects Version/s: 1.6.0
• Component/s: Spark Core
• Labels: hdfs quota spark,

Environment

CDH 5.8

• Assignee: Unassigned
• Reporter: Rivkin Andrey
• Created: 20/Jan/17 15:04
• Updated: 23/Jan/17 11:48
• Resolved: 23/Jan/17 11:48

Description

If we set quota on user space and then will try to create table through hive on spark, which will need more space then avaliable, spark will fail with:

Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: org.apache.hadoop.hive.ql.metadata.HiveException: org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.hdfs.protocol.AlreadyBeingCreatedException): failed to create file /user/xxxx/hive_db/.hive-staging_hive_..../task_tmp.-ext-10003/_tmp.000030_0 for DFSClient_NONMAPREDUCE-27052423_230 for client 192.168.x.x because current leaseholder is trying to recreate file.

If we will change hive execution engine to mr and execute the same command - create table, we will get:

Caused by: org.apache.hadoop.hdfs.protocol.DSQuotaExceededException: The DiskSpace quota of /user/xxxx is exceeded: quota = 10737418240 B = 10 GB but diskspace consumed = 11098812438 B = 10.34 GB

After increasing quota hive on spark is working.
The problem is with log message, which is inaccurate and not helpful.