org.apache.hadoop.hive.ql.history

Interface HiveHistory

All Known Implementing Classes:

HiveHistoryImpl

public interface HiveHistory

HiveHistory. Logs information such as query, query plan, runtime statistics into a file. Each session uses a new object, which creates a new file.

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	HiveHistory.Info Info.
static class	HiveHistory.Keys Keys.
static interface	HiveHistory.Listener Listener interface.
static class	HiveHistory.QueryInfo QueryInfo.
static class	HiveHistory.RecordTypes RecordTypes.
static class	HiveHistory.SessionInfo SessionInfo.
static class	HiveHistory.TaskInfo TaskInfo.

Method Summary

Methods

Modifier and Type	Method and Description
void	closeStream() Close the log file stream
void	endQuery(String queryId) Called at the end of a query
void	endTask(String queryId, Task<? extends Serializable> task) Called at the end of a task.
String	getHistFileName()
void	logPlanProgress(QueryPlan plan) Logs the current plan state
void	printRowCount(String queryId)
void	progressTask(String queryId, Task<? extends Serializable> task) Logs progress of a task if ConfVars.HIVE_LOG_INCREMENTAL_PLAN_PROGRESS is set to true
void	setIdToTableMap(Map<String,String> map) Set the id to table name map
void	setQueryProperty(String queryId, HiveHistory.Keys propName, String propValue) Used to set query status and other attributes of a query
void	setTaskCounters(String queryId, String taskId, org.apache.hadoop.mapred.Counters ctrs) Serialize the task counters and set as a task property.
void	setTaskProperty(String queryId, String taskId, HiveHistory.Keys propName, String propValue) Used to set task properties.
void	startQuery(String cmd, String id) Called at the start of query execution in Driver.execute().
void	startTask(String queryId, Task<? extends Serializable> task, String taskName) Called at the start of a task.

Method Detail

getHistFileName

String getHistFileName()

Returns:

historyFileName

startQuery

void startQuery(String cmd,
              String id)

Called at the start of query execution in Driver.execute().

setQueryProperty

void setQueryProperty(String queryId,
                    HiveHistory.Keys propName,
                    String propValue)

Used to set query status and other attributes of a query

Parameters:

queryId -

propName -

propValue -

setTaskProperty

void setTaskProperty(String queryId,
                   String taskId,
                   HiveHistory.Keys propName,
                   String propValue)

Used to set task properties.

Parameters:

taskId -

propName -

propValue -

setTaskCounters

void setTaskCounters(String queryId,
                   String taskId,
                   org.apache.hadoop.mapred.Counters ctrs)

Serialize the task counters and set as a task property.

Parameters:

queryId -

taskId -

ctrs -

printRowCount

void printRowCount(String queryId)

endQuery

void endQuery(String queryId)

Called at the end of a query

Parameters:

queryId -

startTask

void startTask(String queryId,
             Task<? extends Serializable> task,
             String taskName)

Called at the start of a task. Called by Driver.run() A query can have multiple tasks. Tasks will have multiple operator.

Parameters:

task -

endTask

void endTask(String queryId,
           Task<? extends Serializable> task)

Called at the end of a task.

Parameters:

task -

progressTask

void progressTask(String queryId,
                Task<? extends Serializable> task)

Logs progress of a task if ConfVars.HIVE_LOG_INCREMENTAL_PLAN_PROGRESS is set to true

Parameters:

task -

logPlanProgress

void logPlanProgress(QueryPlan plan)
                     throws IOException

Logs the current plan state

Parameters:

plan -

Throws:

IOException

setIdToTableMap

void setIdToTableMap(Map<String,String> map)

Set the id to table name map

Parameters:

map -

closeStream

void closeStream()

Close the log file stream