Database Reference
In-Depth Information
Listing 13-10.
hive.log file showing some errors
2013-11-15 14:25:31,810 INFO ql.Driver (PerfLogger.java:PerfLogBegin(100))
- <PERFLOG method=Driver.run>
2013-11-15 14:25:31,811 INFO ql.Driver (PerfLogger.java:PerfLogBegin(100))
- <PERFLOG method=TimeToSubmit>
2013-11-15 14:25:31,811 INFO ql.Driver (PerfLogger.java:PerfLogBegin(100))
- <PERFLOG method=compile>
2013-11-15 14:25:31,812 INFO parse.ParseDriver (ParseDriver.java:parse(179))
- Parsing command: drop database hive
2013-11-15 14:25:31,813 INFO parse.ParseDriver (ParseDriver.java:parse(197))
- Parse Completed
2013-11-15 14:25:31,814 INFO ql.Driver (Driver.java:compile(442))
- Semantic Analysis Completed
2013-11-15 14:25:31,815 INFO ql.Driver (Driver.java:getSchema(259))
- Returning Hive schema: Schema(fieldSchemas:null, properties:null)
2013-11-15 14:25:31,816 INFO ql.Driver (PerfLogger.java:PerfLogEnd(127))
- </PERFLOG method=compile start=1384525531811 end=1384525531816 duration=5>
2013-11-15 14:25:31,816 INFO ql.Driver (PerfLogger.java:PerfLogBegin(100))
- <PERFLOG method=Driver.execute>
2013-11-15 14:25:31,816 INFO ql.Driver (Driver.java:execute(1066))
- Starting command: drop database hive
2013-11-15 14:25:31,816 INFO ql.Driver (PerfLogger.java:PerfLogEnd(127))
- </PERFLOG method=TimeToSubmit start=1384525531811 end=1384525531816 duration=5>
2013-11-15 14:25:31,846 ERROR exec.Task
(SessionState.java:printError(432))
- There is no database named hive
NoSuchObjectException(
message
:
There is no database named
hive
)Atorg.apache.hadoop.hive.metastore.api.ThriftHiveMetastore
$get_database_result$get_database_resultStandardScheme.read(ThriftHiveMetastore.java:9883)
There could be errors while executing DML commands like SELECT against your Hive tables. To understand and
troubleshoot such errors, you need to know the different phases that a HQL goes through. Table
13-1
summarizes the
phases of Hive query execution.
Table 13-1.
Hive query execution phases
Phases
Description
Parsing
Converts a Query into Parse Tree. If there are syntax errors in your query
(for example, a missing semi-colon at the end), it is likely to be failing at this stage.
Semantic Analysis
Builds a logical plan based on the information retrieved from the Hive metastore
database. Metadata failure errors, where the underlying schema has changed after
the query is submitted, are reported in this phase.
Physical Plan Generation
Converts the logical plan to a physical plan that generates a Direct Acyclic Graph of
MapReduce jobs that need to be executed. The errors reported in this stage or after
this are MapReduce job errors. Further insights can be gained about these failures
from the TaskTracker log files in the compute nodes.
