Issues

Pipeline failure does not show cause of failure in logs

MapReduce jobs fail to launch on CDH 5.4

ProgramLifecycleHttpHandlerTest can hang the build

MapReduce can deadlock

Pipelines with many sinks may exceed the Variable substitution depth in the Hadoop conf

User should be able to call MapReduceContext#addOutput without a transaction

ETLMapReduce should add error datasets within a transaction

MapReduce programs emit too many metrics

PartitionConsumer onFinish should be called within the MR's transaction

Ability for MapReduce to append to an existing PFS partition

If a MapReduce job using PartitionConsumer fails, the partitions remain in-progress

NullPointerException when making a request to get /info when the MapReduce is Starting

Ability for DynamicPartitioner to append to an existing PFS partition

CombineFileInputFormat does not work with PFS as input

MapReduce jobs fail to be submitted on HDP 2.6.1

Add a configuration for the frequency of getting MapReduce task reports

MapReduce status is sometimes successful even though the job failed

There should be a way to retrieve the job conf of a MapReduce run

There should be a way to configure Spark/MapReduce job configuration through preferences

Mapreduce INFO logs are getting logged as WARN in CDAP

ReadlessIncrementTest sometimes fails due to memory limits

Compatibility Modules for different Hadoop distros/versions

Upgrade CDAP dependencies on Hadoop and HBase to a newer version

PartitionedFileSet does not clean up files under certain MapReduce failure conditions

Make file set permissions work with MultipleOutputs

Mapper Bytes Out is wrong

Readless Increments do not work from MapReduce

Remove all deprecated methods from Wise app

Better log message when a MapReduce fails

DynamicPartitioner should have a way to close a writer when it is known to be done

MultipleOutputs#close() should pass appropriate context rather than using same context

MapReduce classloader is closed prematurely in MRAppMaster

MapReduce's beforeSubmit is committed, even in case of some failures

DynamicPartitioner does not remove files upon failure

MapReduce should run the DatasetOutputCommitters in a separate transaction

Refactor MapReduceRuntimeService to configure inputs and outputs during initialize()

Remove deprecated setInput and setOutput methods in MapReduceConfigurer

If a program fails during startup, destroy() is never called

MRAppMaster java process running longer than expected.

If a mapreduce job is configured in a workflow, it should be discovered by the application during deploy

PartitionedFileSetArguments.setOutputPartitionMetadata does not work when Reducers are used

Docs for CDAP-5740

Concurrency protection for datasets and partitions

MapReduce programs fail on clusters with Phoenix enabled

Remove unsetting of 'mapreduce.jobhistory.address' from MapReduceRuntimeService

Multiple instances of a Table in the same transaction do not see each other's writes

CDAP should allow rules to react to programs that run longer than expected

Mapreduce jobs prints a lot of DEBUG messages for every single record that is processed

CDAP settings used in MapReduce framework should be settable in cdap-site.xml