[
https://issues.apache.org/jira/browse/HIVE-15197?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
vishal.rajan updated HIVE-15197:
--------------------------------
Description:
When the below query is run in hive 1.2.0 it returns 'NULL NULL 0' on
empty table but when the same query is run on hive 2.1.0, nothing is returned
on empty table.
----------------hive 1.2.0(ORC)-------------------------
hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*) from
test_stage.geo_zone;
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1 Cumulative CPU: 4.79 sec HDFS Read: 7354 HDFS Write:
114 SUCCESS
Total MapReduce CPU Time Spent: 4 seconds 790 msec
OK
NULL NULL 0
Time taken: 38.168 seconds, Fetched: 1 row(s)
---------------------hive 2.1.0(ORC)-------------------------------------
hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*) from
test_stage.geo_zone
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the
future versions. Consider using a different execution engine (i.e. spark, tez)
or using Hive 1.X releases.
Total jobs = 1
Launching Job 1 out of 1
Number of reduce tasks is set to 0 since there's no reduce operator
2016-11-14 19:06:15,421 WARN [Thread-215] mapreduce.JobResourceUploader
(JobResourceUploader.java:uploadFiles(64)) - Hadoop command-line option parsing
not performed. Implement the Tool interface and execute your application with
ToolRunner to remedy this.
2016-11-14 19:06:19,222 INFO [Thread-215] input.FileInputFormat
(FileInputFormat.java:listStatus(283)) - Total input paths to process : 1
2016-11-14 19:06:20,000 INFO [Thread-215] mapreduce.JobSubmitter
(JobSubmitter.java:submitJobInternal(198)) - number of splits:0
Hadoop job information for Stage-1: number of mappers: 0; number of reducers: 0
2016-11-14 19:06:39,405 Stage-1 map = 0%, reduce = 0%
Stage-Stage-1: HDFS Read: 0 HDFS Write: 0 SUCCESS
Total MapReduce CPU Time Spent: 0 msec
OK
Time taken: 28.302 seconds
was:
When the below query is run in hive 1.2.0 it returns 'NULL NULL 0' on
empty table but when the same query is run on hive 2.1.0, nothing is returned
on empty table.
----------------hive 1.2.0 (avro table)-------------------------
hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*) from
test_stage.geo_zone;
MapReduce Jobs Launched:
Stage-Stage-1: Map: 1 Cumulative CPU: 4.79 sec HDFS Read: 7354 HDFS Write:
114 SUCCESS
Total MapReduce CPU Time Spent: 4 seconds 790 msec
OK
NULL NULL 0
Time taken: 38.168 seconds, Fetched: 1 row(s)
---------------------hive 2.1.0(ORC)-------------------------------------
hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*) from
test_stage.geo_zone
WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the
future versions. Consider using a different execution engine (i.e. spark, tez)
or using Hive 1.X releases.
Total jobs = 1
Launching Job 1 out of 1
Number of reduce tasks is set to 0 since there's no reduce operator
2016-11-14 19:06:15,421 WARN [Thread-215] mapreduce.JobResourceUploader
(JobResourceUploader.java:uploadFiles(64)) - Hadoop command-line option parsing
not performed. Implement the Tool interface and execute your application with
ToolRunner to remedy this.
2016-11-14 19:06:19,222 INFO [Thread-215] input.FileInputFormat
(FileInputFormat.java:listStatus(283)) - Total input paths to process : 1
2016-11-14 19:06:20,000 INFO [Thread-215] mapreduce.JobSubmitter
(JobSubmitter.java:submitJobInternal(198)) - number of splits:0
Hadoop job information for Stage-1: number of mappers: 0; number of reducers: 0
2016-11-14 19:06:39,405 Stage-1 map = 0%, reduce = 0%
Stage-Stage-1: HDFS Read: 0 HDFS Write: 0 SUCCESS
Total MapReduce CPU Time Spent: 0 msec
OK
Time taken: 28.302 seconds
> count and sum query on empty table, returning empty output
> -----------------------------------------------------------
>
> Key: HIVE-15197
> URL: https://issues.apache.org/jira/browse/HIVE-15197
> Project: Hive
> Issue Type: Bug
> Components: Hive
> Affects Versions: 2.0.0, 2.1.0, 2.0.1
> Reporter: vishal.rajan
>
> When the below query is run in hive 1.2.0 it returns 'NULL NULL 0' on
> empty table but when the same query is run on hive 2.1.0, nothing is returned
> on empty table.
> ----------------hive 1.2.0(ORC)-------------------------
> hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*)
> from test_stage.geo_zone;
> MapReduce Jobs Launched:
> Stage-Stage-1: Map: 1 Cumulative CPU: 4.79 sec HDFS Read: 7354 HDFS
> Write: 114 SUCCESS
> Total MapReduce CPU Time Spent: 4 seconds 790 msec
> OK
> NULL NULL 0
> Time taken: 38.168 seconds, Fetched: 1 row(s)
> ---------------------hive 2.1.0(ORC)-------------------------------------
> hive> SELECT sum(destination_pincode),sum(length(source_city)),count(*) from
> test_stage.geo_zone
> WARNING: Hive-on-MR is deprecated in Hive 2 and may not be available in the
> future versions. Consider using a different execution engine (i.e. spark,
> tez) or using Hive 1.X releases.
> Total jobs = 1
> Launching Job 1 out of 1
> Number of reduce tasks is set to 0 since there's no reduce operator
> 2016-11-14 19:06:15,421 WARN [Thread-215] mapreduce.JobResourceUploader
> (JobResourceUploader.java:uploadFiles(64)) - Hadoop command-line option
> parsing not performed. Implement the Tool interface and execute your
> application with ToolRunner to remedy this.
> 2016-11-14 19:06:19,222 INFO [Thread-215] input.FileInputFormat
> (FileInputFormat.java:listStatus(283)) - Total input paths to process : 1
> 2016-11-14 19:06:20,000 INFO [Thread-215] mapreduce.JobSubmitter
> (JobSubmitter.java:submitJobInternal(198)) - number of splits:0
> Hadoop job information for Stage-1: number of mappers: 0; number of reducers: > 0
> 2016-11-14 19:06:39,405 Stage-1 map = 0%, reduce = 0%
> Stage-Stage-1: HDFS Read: 0 HDFS Write: 0 SUCCESS
> Total MapReduce CPU Time Spent: 0 msec
> OK
> Time taken: 28.302 seconds
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
