[HIVE-12091] Merge file doesn't work for ORC table when running on Spark. [Spark Branch]

• Type: Sub-task
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.2.1
• Fix Version/s: spark-branch, 1.3.0, 2.0.0
• Component/s: Spark
• Assignee: Rui Li
• Reporter: Xin Hao
• Created: 12/Oct/15 08:10
• Updated: 16/Feb/16 23:51
• Resolved: 14/Oct/15 02:30

Description

This issue occurs when hive.merge.sparkfiles is set to true. And can be workaround by setting hive.merge.sparkfiles to false.
BTW, we did a local experiment to run the case with MR engine (set hive.merge.mapfiles=true; set hive.merge.mapredfiles=true, it can pass.

(1)Component Version：
– Hive Spark Branch 70eeadd2f019dcb2e301690290c8807731eab7a1 + Hive-11473 patch (HIVE-11473.3-spark.patch) ---> This is to support Spark 1.5 for Hive on Spark
– Spark 1.5.1

(2)Case used:
– Big-Bench Data Load (load data from HDFS to Hive warehouse, scored as ORC format). The related HiveQL:

DROP TABLE IF EXISTS customer_temporary;
CREATE EXTERNAL TABLE customer_temporary
  ( c_customer_sk             bigint              --not null
  , c_customer_id             string              --not null
  , c_current_cdemo_sk        bigint
  , c_current_hdemo_sk        bigint
  , c_current_addr_sk         bigint
  , c_first_shipto_date_sk    bigint
  , c_first_sales_date_sk     bigint
  , c_salutation              string
  , c_first_name              string
  , c_last_name               string
  , c_preferred_cust_flag     string
  , c_birth_day               int
  , c_birth_month             int
  , c_birth_year              int
  , c_birth_country           string
  , c_login                   string
  , c_email_address           string
  , c_last_review_date        string
  )
  ROW FORMAT DELIMITED FIELDS TERMINATED BY '|'
  STORED AS TEXTFILE LOCATION '/user/root/benchmarks/bigbench_n1t/data/customer'
;

DROP TABLE IF EXISTS customer;
CREATE TABLE customer
STORED AS ORC
AS
SELECT * FROM customer_temporary
;

(3)Error/Exception Message:

15/10/12 14:28:38 INFO exec.Utilities: PLAN PATH = hdfs://bhx2:8020/tmp/hive/root/4e145415-d4ea-4751-9e16-ff31edb0c258/hive_2015-10-12_14-28-12_485_2093357701513622173-1/-mr-10005/d891fdec-eacc-4f66-8827-e2b650c24810/map.xml
15/10/12 14:28:38 INFO OrcFileMergeOperator: ORC merge file input path: hdfs://bhx2:8020/user/hive/warehouse/bigbench_n100g.db/.hive-staging_hive_2015-10-12_14-28-12_485_2093357701513622173-1/-ext-10003/000001_0
15/10/12 14:28:38 INFO OrcFileMergeOperator: Merged stripe from file hdfs://bhx2:8020/user/hive/warehouse/bigbench_n100g.db/.hive-staging_hive_2015-10-12_14-28-12_485_2093357701513622173-1/-ext-10003/000001_0 [ offset : 3 length: 10525754 row: 247500 ]
15/10/12 14:28:38 INFO spark.SparkMergeFileRecordHandler: Closing Merge Operator OFM
15/10/12 14:28:38 ERROR executor.Executor: Exception in task 1.0 in stage 1.0 (TID 4)
java.lang.RuntimeException: org.apache.hadoop.hive.ql.metadata.HiveException: Failed to close AbstractFileMergeOperator
	at org.apache.hadoop.hive.ql.exec.spark.SparkMergeFileRecordHandler.close(SparkMergeFileRecordHandler.java:115)
	at org.apache.hadoop.hive.ql.exec.spark.HiveMapFunctionResultList.closeRecordProcessor(HiveMapFunctionResultList.java:58)
	at org.apache.hadoop.hive.ql.exec.spark.HiveBaseFunctionResultList$ResultIterator.hasNext(HiveBaseFunctionResultList.java:106)
	at scala.collection.convert.Wrappers$JIteratorWrapper.hasNext(Wrappers.scala:41)
	at scala.collection.Iterator$class.foreach(Iterator.scala:727)
	at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
	at org.apache.spark.rdd.AsyncRDDActions$$anonfun$foreachAsync$1$$anonfun$apply$15.apply(AsyncRDDActions.scala:118)
	at org.apache.spark.rdd.AsyncRDDActions$$anonfun$foreachAsync$1$$anonfun$apply$15.apply(AsyncRDDActions.scala:118)
	at org.apache.spark.SparkContext$$anonfun$37.apply(SparkContext.scala:1984)
	at org.apache.spark.SparkContext$$anonfun$37.apply(SparkContext.scala:1984)
	at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:66)
	at org.apache.spark.scheduler.Task.run(Task.scala:88)
	at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:214)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
	at java.lang.Thread.run(Thread.java:745)
Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Failed to close AbstractFileMergeOperator
	at org.apache.hadoop.hive.ql.exec.AbstractFileMergeOperator.closeOp(AbstractFileMergeOperator.java:235)
	at org.apache.hadoop.hive.ql.exec.OrcFileMergeOperator.closeOp(OrcFileMergeOperator.java:236)
	at org.apache.hadoop.hive.ql.exec.spark.SparkMergeFileRecordHandler.close(SparkMergeFileRecordHandler.java:113)
	... 15 more
Caused by: java.io.IOException: Unable to rename hdfs://bhx2:8020/user/hive/warehouse/bigbench_n100g.db/.hive-staging_hive_2015-10-12_14-28-12_485_2093357701513622173-1/_task_tmp.-ext-10001/_tmp.000001_0 to hdfs://bhx2:8020/user/hive/warehouse/bigbench_n100g.db/.hive-staging_hive_2015-10-12_14-28-12_485_2093357701513622173-1/_tmp.-ext-10001/000001_0
	at org.apache.hadoop.hive.ql.exec.AbstractFileMergeOperator.closeOp(AbstractFileMergeOperator.java:208)
	... 17 more

Issue Links

• is duplicated by: HIVE-12089 HiveException (Failed to close AbstractFileMergeOperator) occurs during loading data to ORC file, when hive.merge.sparkfiles is set to true. [Spark Branch]