Hive error (non-partition column appears in the Partition specification)

In the process of writing and testing, some of them may need further verification. Sometimes hive error location is not necessarily correct need more confirmation

1 failed:nullpointerexception NULL

You cannot use a view as the right table for a left outer join

2 failed:udfargumenttypeexception only numeric or string type arguments be accepted but decimal is passed.

In CDH hive0.10, the AVG column cannot be of type decimal, and Apache hive0.12 does not have this restriction

3 Failed:semanticexception [Error 10098]: non-partition column appears in the Partition specification

No partition was specified when the table was built, and the INSERT statement specified the

4 Failed:semanticexception [Error 10017]: line 5:5 Both left and right aliases encountered in JOIN ' ***_id '

Non-equivalent connections are not supported

5 failed:parseexception Line 1:799 missing Identifier @ ' group ' near ' <EOF> '

The alias of the child table is missing in front of group

6 field of the right table of the left semi join, cannot appear in the Select column

7 failed:semanticexception Column ***_id Found in more than one tables/subqueries

One column appears in more than one table, but no table name prefix is added

8 Diagnostic Messages for this Task:

Error:java.lang.RuntimeException:org.apache.hadoop.hive.ql.metadata.HiveException: Java.io.FileNotFoundException:/hadoop/tmp/nm-local-dir/usercache/hadoop/appcache/application_***/container_*** /stage-10.tar.gz/mapjoin-mapfile160--. Hashtable (No such file or directory)

Caused by:org.apache.hadoop.hive.ql.metadata.HiveException:java.io.FileNotFoundException:/hadoop/tmp/ nm-local-dir/usercache/hadoop/appcache/application_***/container_***/stage-10.tar.gz/ mapjoin-mapfile160--. Hashtable (No such file or directory)

Apache hive0.12 does not support the table after create as to join three tables, only two; cdhhive0.10 no this limit

9 cdhhive0.10 does not support field names defined as timestamp

10 add memory, should be specified with hiveconf when new task is created

The statement cannot be executed if the parameter is used directly and then the set is taken.

The parameter values of each SET statement do not support UDF settings, only constants, or variables that have been set

The use of the left outer join, the setting of the filter condition of the right table to pay particular attention to, if placed in the where, then it will become a inner join, the actual is to execute the first and then after the leave join, and then the filter, so that will be a part of the records in the side of the delete, does not conform to the left join the original intention to take the full amount of the table. The right table condition needs to be specified in the form of a child table, or on. The same is true in Oracle, where the conditional logic in on and where has a different meaning, which should not be a hive bug.

Hive error (non-partition column appears in the Partition specification)