org.apache.hadoop.mapreduce.v2.hs

インタフェース HistoryStorage

@InterfaceAudience.Public
 @InterfaceStability.Unstable
public interface HistoryStorage

Provides an API to query jobs that have finished. For those implementing this API be aware that there is no feedback when files are removed from HDFS. You may rely on HistoryFileManager to help you know when that has happened if you have not made a complete backup of the data stored on HDFS.

メソッドのサマリー

修飾子とタイプ	メソッドと説明
Map<org.apache.hadoop.mapreduce.v2.api.records.JobId,org.apache.hadoop.mapreduce.v2.app.job.Job>	getAllPartialJobs() Get all of the cached jobs.
org.apache.hadoop.mapreduce.v2.app.job.Job	getFullJob(org.apache.hadoop.mapreduce.v2.api.records.JobId jobId) Get a fully parsed job.
org.apache.hadoop.mapreduce.v2.hs.webapp.dao.JobsInfo	getPartialJobs(Long offset, Long count, String user, String queue, Long sBegin, Long sEnd, Long fBegin, Long fEnd, org.apache.hadoop.mapreduce.v2.api.records.JobState jobState) Look for a set of partial jobs.
void	setHistoryFileManager(HistoryFileManager hsManager) Give the Storage a reference to a class that can be used to interact with history files.

メソッドの詳細

setHistoryFileManager

void setHistoryFileManager(HistoryFileManager hsManager)

Give the Storage a reference to a class that can be used to interact with history files.

パラメータ:

hsManager - the class that is used to interact with history files.

getPartialJobs

org.apache.hadoop.mapreduce.v2.hs.webapp.dao.JobsInfo getPartialJobs(Long offset,
                                                                     Long count,
                                                                     String user,
                                                                     String queue,
                                                                     Long sBegin,
                                                                     Long sEnd,
                                                                     Long fBegin,
                                                                     Long fEnd,
                                                                     org.apache.hadoop.mapreduce.v2.api.records.JobState jobState)

Look for a set of partial jobs.

パラメータ:

offset - the offset into the list of jobs.

count - the maximum number of jobs to return.

user - only return jobs for the given user.

queue - only return jobs for in the given queue.

sBegin - only return Jobs that started on or after the given time.

sEnd - only return Jobs that started on or before the given time.

fBegin - only return Jobs that ended on or after the given time.

fEnd - only return Jobs that ended on or before the given time.

jobState - only return Jobs that are in the given job state.

戻り値:

The list of filtered jobs.

getAllPartialJobs

Map<org.apache.hadoop.mapreduce.v2.api.records.JobId,org.apache.hadoop.mapreduce.v2.app.job.Job> getAllPartialJobs()

Get all of the cached jobs. This only returns partial jobs and is here for legacy reasons.

戻り値:

all of the cached jobs

getFullJob

org.apache.hadoop.mapreduce.v2.app.job.Job getFullJob(org.apache.hadoop.mapreduce.v2.api.records.JobId jobId)

Get a fully parsed job.

パラメータ:

jobId - the id of the job

戻り値:

the job, or null if it is not found.