Apache Hadoop Changelog

Release 0.18.0 - 2008-08-22

INCOMPATIBLE CHANGES:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3837hadop streaming does not use progress reporting to detect hung tasksMajor.dhruba borthakurdhruba borthakur
HADOOP-3808[HOD] Include job tracker RPC in notes attribute after job submissionBlockercontrib/hodHemanth YamijalaHemanth Yamijala
HADOOP-3703[HOD] logcondense needs to use the new pattern of output in hadoop dfs -lsrBlockercontrib/hodHemanth YamijalaVinod Kumar Vavilapalli
HADOOP-3683Hadoop dfs metric FilesListed shows number of files listed instead of operationsMajormetricsLohit VijayarenuLohit Vijayarenu
HADOOP-3665WritableComparator newKey() fails for NullWritableMinorioLukas VlcekChris Douglas
HADOOP-3610[HOD] HOD does not automatically create a cluster directory for the script optionBlockercontrib/hodHemanth YamijalaVinod Kumar Vavilapalli
HADOOP-3598Map-Reduce framework needlessly creates temporary _${taskid} directories for MapsBlocker.Arun C MurthyArun C Murthy
HADOOP-3569KFS input stream read() returns 4 bytes instead of 1Minor.Sriram RaoSriram Rao
HADOOP-3512Split map/reduce tools into separate jarsMajor.Owen O'MalleyOwen O'Malley
HADOOP-3486Change default for initial block report to 0 sec and document it in hadoop-defaults.xmlMajor.Sanjay RadiaSanjay Radia
HADOOP-3483[HOD] Improvements with cluster directory handlingMajorcontrib/hodHemanth YamijalaHemanth Yamijala
HADOOP-3459Change dfs -ls listing to closely match format on LinuxMajor.Mukund MadhugiriMukund Madhugiri
HADOOP-3452fsck exit code would be better if non-zero when FS corruptMinor.Pete WyckoffLohit Vijayarenu
HADOOP-3417JobClient should not have a static configuration for cli parsingMajor.Owen O'MalleyAmareshwari Sriramadasu
HADOOP-3405Make mapred internal classes package-localMajor.Enis SoztutarEnis Soztutar
HADOOP-3390Remove deprecated ClientProtocol.abandonFileInProgress()Major.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3379Document the “stream.non.zero.exit.status.is.failure” knob for streamingBlockerdocumentationArun C MurthyAmareshwari Sriramadasu
HADOOP-3329DatanodeDescriptor objects stored in FSImage may be out dated.Major.Tsz Wo Nicholas Szedhruba borthakur
HADOOP-3317add default port for hdfs namenodeMinor.Doug CuttingDoug Cutting
HADOOP-3310Lease recovery for appendMajor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3283Need a mechanism for data nodes to update generation stamps.Major.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3265Remove deprecated API getFileCacheHintsMajorfsLohit VijayarenuLohit Vijayarenu
HADOOP-3226Run combiner when merging spills from map outputMajor.Chris DouglasChris Douglas
HADOOP-3193Discovery of corrupt block reported in name node logMinor.Robert ChanslerChris Douglas
HADOOP-3184HOD gracefully exclude “bad” nodes during ring formationMajorcontrib/hodMarco NicosiaHemanth Yamijala
HADOOP-3113DFSOututStream.flush() should flush data to real block file on DataNode.Major.dhruba borthakurdhruba borthakur
HADOOP-3095Validating input paths and creating splits is slow on S3Majorfs, fs/s3Tom WhiteTom White
HADOOP-3035Data nodes should inform the name-node about block crc errors.Major.Konstantin ShvachkoLohit Vijayarenu
HADOOP-2909Improve IPC idle connection managementMajoripcHairong KuangHairong Kuang
HADOOP-2865FsShell.ls() should print file attributes first then the path name.Major.Konstantin ShvachkoEdward J. Yoon
HADOOP-2797Withdraw CRC upgrade from HDFSCritical.Robert ChanslerRaghu Angadi
HADOOP-2703New files under lease (before close) still shows up as MISSING files/blocks in fsckMinor.Koji NoguchiLohit Vijayarenu
HADOOP-2656Support for upgrading existing cluster to facilitate appends to HDFS filesMajor.dhruba borthakurdhruba borthakur
HADOOP-2585Automatic namespace recovery from the secondary image.Major.Konstantin ShvachkoKonstantin Shvachko
HADOOP-2188RPC should send a ping rather than use client timeoutsMajoripcOwen O'MalleyHairong Kuang
HADOOP-2181Input Split details for maps should be loggedMinor.Lohit VijayarenuAmareshwari Sriramadasu
HADOOP-2065Replication policy for corrupted blockMajor.Koji NoguchiLohit Vijayarenu
HADOOP-1702Reduce buffer copies when data is written to DFSMajor.Raghu AngadiRaghu Angadi
HADOOP-544Replace the job, tip and task ids with objects.Major.Owen O'MalleyEnis Soztutar

IMPORTANT ISSUES:

JIRASummaryPriorityComponentReporterContributor

NEW FEATURES:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3502Quota API needs documentation in ForrestMajor.Hairong KuangHairong Kuang
HADOOP-3460SequenceFileAsBinaryOutputFormatMinor.Koji NoguchiKoji Noguchi
HADOOP-3336Direct a subset of namenode RPC events for audit loggingMajor.Chris DouglasChris Douglas
HADOOP-3307Archives in Hadoop.MajorfsMahadev konarMahadev konar
HADOOP-3250Extend FileSystem API to allow appending to filesMajorfsdhruba borthakurTsz Wo Nicholas Sze
HADOOP-3246FTP client over HDFSMajorutilAnkurAnkur
HADOOP-3230Add command line access to named countersMajorscriptsTom WhiteTom White
HADOOP-3221Need a “LineBasedTextInputFormat”Major.Milind BhandarkarAmareshwari Sriramadasu
HADOOP-3188compaction utility for directoriesMajor.Robert ChanslerRobert Chansler
HADOOP-3187Quotas for name space managementMajor.Robert ChanslerHairong Kuang
HADOOP-3074URLStreamHandler for the DFSMajorutilChristophe TatonChristophe Taton
HADOOP-3061Writable for single byte and doubleMajorioAndrzej BialeckiAndrzej Bialecki
HADOOP-3022Fast Cluster RestartMajor.Robert ChanslerKonstantin Shvachko
HADOOP-1915adding counters methods using String (as opposed to Enum)Minor.Alejandro AbdelnurTom White
HADOOP-1328Hadoop Streaming needs to provide a way for the stream plugin to update global countersMajor.Runping QiTom White
HADOOP-930Add support for reading regular (non-block-based) files from S3 in S3FileSystemMajorfsTom WhiteTom White

IMPROVEMENTS:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3599The new setCombineOnceOnly shouldn't take a JobConf, since it is a method on JobConfMajor.Owen O'MalleyOwen O'Malley
HADOOP-3572setQuotas usage interface has some minor bugs.Minor.Mahadev konarHairong Kuang
HADOOP-3547Improve documentation about distributing native libraries via DistributedCacheBlockerdocumentationArun C MurthyAmareshwari Sriramadasu
HADOOP-3535IOUtils.close needs better documentationBlockerioOwen O'MalleyOwen O'Malley
HADOOP-3532Create build targets to create api change reports using jdiffMajorbuildOwen O'MalleyOwen O'Malley
HADOOP-3501deprecate InMemoryFileSystemMajorfsDoug CuttingDoug Cutting
HADOOP-3492add forrest documentation for user archivesBlocker.Mahadev konarMahadev konar
HADOOP-3467The javadoc for FileSystem.deleteOnExit should have more descriptionBlockerdocumentationTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3464[HOD] HOD can improve error messages by reporting failures on compute nodes back to hod clientMajorcontrib/hodVinod Kumar VavilapalliVinod Kumar Vavilapalli
HADOOP-3455IPC.Client synchronisation looks weakMajoripcSteve LoughranHairong Kuang
HADOOP-3448Add some more hints of the problem when datanode and namenode don't matchMinor.Steve LoughranSteve Loughran
HADOOP-3434Retain cause of bind failure in Server.bindMajor.Steve LoughranSteve Loughran
HADOOP-3429Increase the buffersize for the streaming parent java process's streamsMajor.Devaraj DasAmareshwari Sriramadasu
HADOOP-3406Document controls for profiling maps & reducesBlockerdocumentationArun C MurthyAmareshwari Sriramadasu
HADOOP-3400Facilitate creation of temporary files in HDFSMajor.dhruba borthakurdhruba borthakur
HADOOP-3398ReduceTask::closestPowerOf2 is inefficientTrivial.Chris DouglasChris Douglas
HADOOP-3377Use StringUtils#replaceAll instead ofTrivial.Brice ArnouldBrice Arnould
HADOOP-3369Fast block processing during name-node startup.Major.Konstantin ShvachkoKonstantin Shvachko
HADOOP-3366Shuffle/Merge improvementsMajor.Arun C MurthyArun C Murthy
HADOOP-3364Faster image and log edits loading.Major.Konstantin ShvachkoKonstantin Shvachko
HADOOP-3355Configuration should accept decimal and hexadecimal valuesMajorconfAmareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3350distcp should permit users to limit the number of mapsMajor.Chris DouglasChris Douglas
HADOOP-3345Enhance the hudson-test-patch targetMinorbuildNigel DaleyNigel Daley
HADOOP-3334Move lease handling codes out from FSNamesystemMajor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3332improving the logging during shufflingBlocker.Runping QiDevaraj Das
HADOOP-3308Improve QuickSort by excluding values eq the pivot from the partitionMajor.Chris DouglasChris Douglas
HADOOP-3297The way in which ReduceTask/TaskTracker gets completion events during shuffle can be improvedMajor.Devaraj DasDevaraj Das
HADOOP-3295Allow TextOutputFormat to use configurable separatorsMinorioZheng ShaoZheng Shao
HADOOP-3277hod should better errors message when deallocate is fired on non allocated directory.Minorcontrib/hodKaram Singh
HADOOP-3274The default constructor of BytesWritable should not create a 100-byte array.MinorioTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3270Constant DatanodeCommand should be stored in static fianl immutable variables.Minor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3254FSNamesystem.gotHeartbeat(..., Object[] xferResults, Object[] deleteList) should not use Object[] as pass-by-reference parametersMajor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3177Expose DFSOutputStream.fsync API though the FileSystem interfaceMajor.dhruba borthakurTsz Wo Nicholas Sze
HADOOP-3164Use FileChannel.transferTo() when data is read from DataNode.Major.Raghu AngadiRaghu Angadi
HADOOP-3144better fault tolerance for corrupted text filesMajor.Joydeep Sen SarmaZheng Shao
HADOOP-3096Improve documentation about the Task Execution Environment in the Map-Reduce tutorialBlockerdocumentationArun C MurthyAmareshwari Sriramadasu
HADOOP-3058Hadoop DFS to report more replication metricsMinormetricsMarco NicosiaLohit Vijayarenu
HADOOP-3013fsck to show (checksum) corrupted filesMajor.Koji NoguchiLohit Vijayarenu
HADOOP-2987Keep two generations of fsimageMajor.Robert ChanslerKonstantin Shvachko
HADOOP-2910Throttle IPC Client/Server during bursts of requests or server slowdownMajoripcHairong KuangHairong Kuang
HADOOP-2867Add a task‘s cwd to it’s LD_LIBRARY_PATHMajor.Arun C MurthyAmareshwari Sriramadasu
HADOOP-2857libhdfs: no way to set JVM args other than classpathMinor.Craig MacdonaldCraig Macdonald
HADOOP-2799Replace org.apache.hadoop.io.Closeable with java.io.CloseableMinorioTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-2762Better documentation of controls for memory limits on hadoop daemons and Map-Reduce tasksBlockerdocumentation, scriptsArun C MurthyAmareshwari Sriramadasu
HADOOP-2661Replicator log should include block idMinor.Raghu AngadiRaghu Angadi
HADOOP-2461Configuration should trim property names and accept decimal, hexadecimal, and octal numbersMinorconfTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-2154Non-interleaved checksums would optimize block transfers.Major.Konstantin ShvachkoRaghu Angadi
HADOOP-2019DistributedFileCache should support .tgz files in addition to jars and zip filesMajor.Owen O'MalleyAmareshwari Sriramadasu
HADOOP-4tool to mount dfs on linuxMajor.John XingPete Wyckoff

BUG FIXES:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3947TaskTrackers fail to connect back upon a re-init actionBlocker.Amar KamatAmareshwari Sriramadasu
HADOOP-3901CLASSPATH in bin/hadoop script is set incorrectly for cygwinBlockerscriptsTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3897SecondaryNameNode fails with NullPointerExceptionMajor.Lohit VijayarenuLohit Vijayarenu
HADOOP-3884eclipse plugin build is broken with current eclipse versionsBlockercontrib/eclipse-pluginDoug CuttingDoug Cutting
HADOOP-3865SecondaryNameNode runs out of memoryBlocker.Lohit VijayarenuLohit Vijayarenu
HADOOP-3855Fix import of MiniDFSCluster in TestCompressedEmptyMapOutputs.javaBlockertestArun C MurthyArun C Murthy
HADOOP-3827Jobs with empty map-outputs and intermediate compression failBlocker.Arun C MurthyArun C Murthy
HADOOP-3806Remove debug message from QuicksortTrivial.Chris Douglas
HADOOP-3798compile-core-test fails to compileBlockerbuildMukund MadhugiriMukund Madhugiri
HADOOP-3794KFS implementation needs to return directory modification timeMinor.Sriram RaoSriram Rao
HADOOP-3776NPE in NameNode with unknown blocksBlocker.Raghu AngadiRaghu Angadi
HADOOP-3774Typos in shell outputBlockerfsTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3762Task tracker died due to OOMBlocker.Runping QiDoug Cutting
HADOOP-3755the gridmix scripts do not work with hod 0.4Major.Runping QiRunping Qi
HADOOP-3752Audit logging fails to record renameBlocker.Chris DouglasChris Douglas
HADOOP-3743-libjars, -files and -archives options do not work with 0.18Blocker.Mahadev konarAmareshwari Sriramadasu
HADOOP-3737CompressedWritable throws OutOfMemoryErrorMajorioGrant GlouserGrant Glouser
HADOOP-3724Namenode does not start due to exception throw while saving ImageBlocker.Lohit Vijayarenudhruba borthakur
HADOOP-3718KFS: write(int v) API writes out an integer rather than a byteMinor.Sriram RaoSriram Rao
HADOOP-3716KFS listStatus() returns NULL on empty directoriesMinor.Sriram RaoSriram Rao
HADOOP-3706CompositeInputFormat: Unable to wrap custom InputFormatsMajor.Jingkei LyJingkei Ly
HADOOP-3693Fix documentation for Archives, distcp and native librariesBlockerdocumentationAmareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3692Fix documentation for Cluster setup and Quick start guidesBlockerdocumentationAmareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3691Fix mapred docsBlockerdocumentationAmareshwari SriramadasuJothi Padmanabhan
HADOOP-3688Fix up HDFS docsBlocker.Robert ChanslerRobert Chansler
HADOOP-3677Problems with generation stamp upgradeBlocker.Konstantin ShvachkoRaghu Angadi
HADOOP-3673Deadlock in Datanode RPC serversBlocker.dhruba borthakurTsz Wo Nicholas Sze
HADOOP-3670JobTracker running out of heap spaceBlocker.Christian KunzAmareshwari Sriramadasu
HADOOP-3668Clean up HOD documentationBlockercontrib/hodVinod Kumar VavilapalliVinod Kumar Vavilapalli
HADOOP-3653test-patch target not working on hudson.zones.apache.org due to HADOOP-3480Blocker.Nigel DaleyBrice Arnould
HADOOP-3649ArrayIndexOutOfBounds in FSNamesystem.getBlockLocationsInternalBlocker.Arun C MurthyLohit Vijayarenu
HADOOP-3647Corner-case in IFile leads to failed tasksBlocker.Arun C MurthyArun C Murthy
HADOOP-3645MetricsTimeVaryingRate returns wrong value for metric_avg_timeBlockermetricsLohit VijayarenuLohit Vijayarenu
HADOOP-3639Exception when closing DFSClient while multiple files are openBlocker.Benjamin GuflerBenjamin Gufler
HADOOP-3635Uncaught exception in DataBlockScannerBlocker.Koji NoguchiTsz Wo Nicholas Sze
HADOOP-3630CompositeRecordReader: key and values can be in uninitialized state if files being joined have no recordsMajor.Jingkei LyChris Douglas
HADOOP-3615DatanodeProtocol.versionID should be 16LBlocker.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3606Update streaming documentationBlockerdocumentationAmareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3604Reduce stuck at shuffling phaseBlocker.Runping QiArun C Murthy
HADOOP-3603Setting spill threshold to 100% fails to detect spill for recordsBlocker.Chris DouglasChris Douglas
HADOOP-3597SortValidator always uses the default file system irrespective of the actual inputMajortestJothi PadmanabhanJothi Padmanabhan
HADOOP-3593Update MapRed tutorialBlockerdocumentationDevaraj DasDevaraj Das
HADOOP-3590Null pointer exception in JobTracker when the task tracker is not yet resolvedBlocker.Amar KamatAmar Kamat
HADOOP-3588Bug report for archivesBlocker.Hairong KuangMahadev konar
HADOOP-3586keep combiner backward compatible with earlier versions of hadoopBlocker.Olga NatkovichChris Douglas
HADOOP-3580Using a har file as input for the Sort example failsBlocker.Jothi PadmanabhanMahadev konar
HADOOP-3576hadoop dfs -mv throws NullPointerExceptionBlocker.Lohit VijayarenuTsz Wo Nicholas Sze
HADOOP-3575clover target broken after src restructuringMinorbuildNigel DaleyNigel Daley
HADOOP-3571ArrayIndexOutOfBoundsException in BlocksMap$BlockInfo.setPreviousBlocker.Tsz Wo Nicholas SzeKonstantin Shvachko
HADOOP-3564Sometime after successful hod allocation datanode fails to come up with java.net.BindException for dfs.datanode.ipc.addressBlockercontrib/hodKaram SinghVinod Kumar Vavilapalli
HADOOP-3561With trash enabled, ‘hadoop fs -rmr .’ still fully deletes the working dirBlocker.Chris DouglasChris Douglas
HADOOP-3560Archvies sometimes create empty part files.Blocker.Mahadev konarMahadev konar
HADOOP-3559test-libhdfs fails on linuxBlocker.Mukund MadhugiriLohit Vijayarenu
HADOOP-3552forrest doc for hadoop commandsBlockerdocumentationSharad AgarwalSharad Agarwal
HADOOP-3548The tools.jar is not included in the distributionBlockerbuildOwen O'MalleyOwen O'Malley
HADOOP-3546TaskTracker re-initialization gets stuck in cleaning upBlocker.Amareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3545archive is failing with “Illegal Capacity” errorBlocker.Jothi PadmanabhanMahadev konar
HADOOP-3544The command “archive” is missing in the example in docs/hadoop_archives.html (and pdf)BlockerdocumentationJothi PadmanabhanJothi Padmanabhan
HADOOP-3542Hadoop archives should not create _logs file in the final archive directory.Blocker.Mahadev konarMahadev konar
HADOOP-3539Cygwin: cygpath displays an error message in running bin/hadoop scriptBlockerscriptsTsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3537Datanode not starting up with java.lang.StringIndexOutOfBoundsException in NetworkTopology.removeBlocker.Koji NoguchiHairong Kuang
HADOOP-3534The namenode ignores ioexceptions in closeBlocker.Owen O'MalleyTsz Wo Nicholas Sze
HADOOP-3533The api to JobTracker and TaskTracker have changed incompatiblyBlocker.Owen O'MalleyOwen O'Malley
HADOOP-3531Hod does not report job tracker failure on hod client side when job tracker fails to come upBlockercontrib/hodKaram SinghHemanth Yamijala
HADOOP-3528Metrics FilesCreated and files_deleted metrics do not match.BlockermetricsLohit VijayarenuLohit Vijayarenu
HADOOP-3523[HOD] If a job does not exist in Torque's list of jobs, HOD allocate on previously allocated directory fails.Blockercontrib/hodHemanth YamijalaHemanth Yamijala
HADOOP-3521Hadoop mapreduce task metrics, unable to send metrics data.Blocker.Eric YangArun C Murthy
HADOOP-3520Generation stamp upgrade fails TestDFSUpgradeFromImageBlocker.Konstantin Shvachkodhruba borthakur
HADOOP-3519NPE in DFS FileSystem renameBlocker.Tom WhiteHairong Kuang
HADOOP-3517The last InMemory merge may be missedBlocker.Devaraj DasArun C Murthy
HADOOP-3516TestHarFileSystem.testArchives fails with NullPointerExceptionBlockertestNigel DaleySubru Krishnan
HADOOP-3513Improve NNThroughputBenchmark log messages.MajortestKonstantin ShvachkoKonstantin Shvachko
HADOOP-3511Namenode should not restore the root's quota if the quota was not in the imageBlocker.Hairong KuangHairong Kuang
HADOOP-3509FSNamesystem.close() throws NullPointerExceptionMajor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3505omissions in HOD documentationBlockercontrib/hod, documentationAri RabkinVinod Kumar Vavilapalli
HADOOP-3503Race condition when client and namenode start block recovery simultaneouslyMajor.dhruba borthakurdhruba borthakur
HADOOP-3496TestHarFileSystem.testArchives failsBlockerfsAmareshwari SriramadasuTom White
HADOOP-3493TestStreamingFailure fails.Major.Amareshwari SriramadasuLohit Vijayarenu
HADOOP-3491Name-node shutdown causes InterruptedException in ResolutionMonitorMajor.Konstantin ShvachkoLohit Vijayarenu
HADOOP-3489NPE in SafeModeMonitorMajor.Konstantin ShvachkoLohit Vijayarenu
HADOOP-3487Balancer should not allocate a thread per block moveBlocker.Hairong KuangHairong Kuang
HADOOP-3480Need to update Eclipse template to reflect current trunkBlockerbuildTsz Wo Nicholas SzeBrice Arnould
HADOOP-3476Code cleanup needed in fuse-dfsMajor.Tsz Wo Nicholas SzePete Wyckoff
HADOOP-3471TestIndexedSort sometimes failsMajortestChris DouglasChris Douglas
HADOOP-3468Compile error: FTPFileSystem.java:26: cannot access org.apache.commons.net.ftp.FTPBlockerfsTsz Wo Nicholas SzeAnkur
HADOOP-3463hadoop scripts don't change directory to hadoop_homeCriticalscriptsOwen O'MalleyOwen O'Malley
HADOOP-3454Text.find incorrectly searches beyond the end of the bufferMajor.Chad WhipkeyChad Whipkey
HADOOP-3453ipc.Client.close() throws NullPointerExceptionMajoripcTsz Wo Nicholas SzeHairong Kuang
HADOOP-3451test-libhdfs fails on LinuxBlockertestMukund MadhugiriLohit Vijayarenu
HADOOP-3443map outputs should not be renamed between partitionsCritical.Owen O'MalleyOwen O'Malley
HADOOP-3440TaskRunner creates a symlink with name ‘null’ if a file is added to DistributedCache without fragmentMinor.Abhijit BagriDevaraj Das
HADOOP-3435test-patch fail if sh != bashMajor.Brice ArnouldBrice Arnould
HADOOP-3427In ReduceTask::fetchOutputs, wait for result can be improved slightlyMajor.Devaraj DasDevaraj Das
HADOOP-3424the value returned from getPartition should be checked to make sure it is in the range 0..#reduces-1Major.Owen O'MalleyChris Douglas
HADOOP-3419TestFsck fails once in a while on WINDOWS/LINUXMajor.Lohit VijayarenuLohit Vijayarenu
HADOOP-3418NameNode does not restart if parent directory of a “FileUnderConstruction” is deleted.Blocker.Raghu AngadiTsz Wo Nicholas Sze
HADOOP-3413SequenceFile.Reader doesn't use the Serialization frameworkCriticalioArun C MurthyTom White
HADOOP-3410KFS implementation needs to return file modification timeMinor.Sriram RaoSriram Rao
HADOOP-3409NameNode should save the root inode into fsimageMajor.Hairong KuangHairong Kuang
HADOOP-3408Change FSNamesytem status metrics to IntValueMajormetricsLohit VijayarenuLohit Vijayarenu
HADOOP-3403Job tracker's ExpireTackers thread gets NullPointerException if a tasktracker is lost.Blocker.Amareshwari SriramadasuArun C Murthy
HADOOP-3401Update FileBench to use the “work” directory for SequenceFileOutputFormatMajortestChris DouglasChris Douglas
HADOOP-3399Debug log not removed in ipc clientTrivialipcRaghu AngadiRaghu Angadi
HADOOP-3396Unit test TestDatanodeBlockScanner fails on WindowsCritical.Mukund MadhugiriLohit Vijayarenu
HADOOP-3393TestHDFSServerPorts fails on LINUX (NFS mounted directory) and on WINDOWSMajor.Lohit VijayarenuLohit Vijayarenu
HADOOP-3391HADOOP-3248 introduced a findbugs warning.Minor.Raghu AngadiRaghu Angadi
HADOOP-3388TestDatanodeBlockScanner failed while trying to corrupt replicasMajortestdhruba borthakurdhruba borthakur
HADOOP-3381INode interlinks can multiply effect of memory leaksMajor.Raghu AngadiRaghu Angadi
HADOOP-3376[HOD] HOD should have a way to detect and deal with clusters that violate/exceed resource manager limitsMajorcontrib/hodVinod Kumar VavilapalliVinod Kumar Vavilapalli
HADOOP-3375Lease paths are sometimes not removed from LeaseManager.sortedLeasesByPathBlocker.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3371MBeanUtil dumps stacktrace from registerMBeanMinormetricsLohit VijayarenuLohit Vijayarenu
HADOOP-3365SequenceFile.Sorter.MergeQueue.next does an unnecessary copy of the keyMajorioArun C MurthyDevaraj Das
HADOOP-3363HDFS throws a InconsistentFSStateException when the name node starts up on a directory that isnt formattedBlocker.Steve LoughranKonstantin Shvachko
HADOOP-3351Fix history viewerMajor.Amareshwari SriramadasuAmareshwari Sriramadasu
HADOOP-3349FSNamesystem.changeLease(src, dst) incorrectly updates the paths inside a leaseMajor.Tsz Wo Nicholas SzeTsz Wo Nicholas Sze
HADOOP-3348TestUrlStreamHandler hangs on LINUXMajorfsLohit VijayarenuLohit Vijayarenu
HADOOP-3340hadoop dfs metrics shows 0MajormetricsEric YangLohit Vijayarenu
HADOOP-3339DFS Write pipeline does not detect defective datanode correctly if it times out.Major.Raghu AngadiRaghu Angadi
HADOOP-3338trunk doesn't compile after HADOOP-544 was committedBlocker.Nigel DaleyChristophe Taton
HADOOP-3337Name-node fails to start because DatanodeInfo format changed.Blocker.Konstantin ShvachkoTsz Wo Nicholas Sze
HADOOP-3335‘make clean’ in src/c++/libhdfs does ‘rm -rf /*’CriticalbuildDoug CuttingDoug Cutting
HADOOP-3333job failing because of reassigning same tasktracker to failing tasksBlocker.Christian KunzJothi Padmanabhan
HADOOP-3326ReduceTask should not sleep for 200 ms while waiting for merge to finishMajor.Owen O'MalleySharad Agarwal
HADOOP-3320NPE in NetworkTopology.getDistance()Blocker.Konstantin ShvachkoHairong Kuang
HADOOP-3318Hadoop streaming doesn't recognize “Darwin” as an OS but Soylatte (OpenJDK port to Mac) reports that rather than “Mac OS X”Major.Sam PullaraSam Pullara
HADOOP-3313RPC::Invoker makes unnecessary calls to System.currentTimeMillisMinor.Chris DouglasChris Douglas
HADOOP-3309Unit test fails on Windows: org.apache.hadoop.mapred.TestMiniMRDFSSort.unknownMajor.Mukund MadhugiriLohit Vijayarenu
HADOOP-3301Misleading error message when S3 URI contains hostname containing an underscoreMajorfs/s3Tom WhiteTom White
HADOOP-3300FindBugs warnings in NetworkTopologyMajor.Tom WhiteTom White
HADOOP-3299org.apache.hadoop.mapred.join.CompositeInputFormat does not initialize TextInput format files with the configuration resulting in an NullPointerExceptionMajorioJasonChris Douglas
HADOOP-3296Some levels are skipped while creating the task cache in JobInProgressMajor.Amar KamatAmar Kamat
HADOOP-3282TestCheckpoint occasionally fails because of the port issues.MajortestKonstantin ShvachkoKonstantin Shvachko
HADOOP-3272Reduce redundant copy of Block object in BlocksMap.map hash mapMajor.Lohit VijayarenuLohit Vijayarenu
HADOOP-3269NameNode doesn't startup when restarted after running an MR jobBlocker.Devaraj DasTsz Wo Nicholas Sze
HADOOP-3268TestUrlStreamHandler.testFileUrls fails on WindowsMajortestNigel DaleyChristophe Taton
HADOOP-3259Configuration.substituteVars() needs to handle security exceptionsTrivialconfSteve LoughranEdward J. Yoon
HADOOP-3248Improve Namenode startup performanceMajor.girish vaitheeswarandhruba borthakur
HADOOP-3240TestJobShell should not create files in the current directoryBlockertestTsz Wo Nicholas SzeMahadev konar
HADOOP-3232Datanodes time outCritical.Johan OskarssonJohan Oskarsson
HADOOP-3203TaskTracker::localizeJob doesn't provide the correct size to LocalDirAllocatorMajor.Chris DouglasAmareshwari Sriramadasu
HADOOP-3176Change lease record when a open-for-write-file gets renamedMajor.dhruba borthakurdhruba borthakur
HADOOP-3160remove exists() from ClientProtocol and NameNodeMajor.Lohit VijayarenuLohit Vijayarenu
HADOOP-3151Hod should have better error messages.Majorcontrib/hodArkady BorkovskyVinod Kumar Vavilapalli
HADOOP-3135if the ‘mapred.system.dir’ in the client jobconf is different from the JobTracker's value job submission failsCritical.Alejandro AbdelnurSubru Krishnan
HADOOP-3130Shuffling takes too long to get the last map output.Major.Runping QiAmar Kamat
HADOOP-3127rm /user/<username>/.Trash/____ only moves it back to .TrashMinorfsKoji NoguchiBrice Arnould
HADOOP-3119Text.getBytes()Trivial.Andrew GudkovTim Nelson
HADOOP-3109RPC should accepted connections even when rpc queue is full (ie undo part of HADOOP-2910)Blocker.Sanjay RadiaHairong Kuang
HADOOP-3101‘bin/hadoop job’ should display the help and silently exitMinor.Amar KamatEdward J. Yoon
HADOOP-3085pushMetric() method of various metric util classes should catch exceptionsMajormetricsRunping QiChris Douglas
HADOOP-3076[HOD] If a cluster directory is specified as a relative path, an existing script.exitcode file will not be deleted.Blockercontrib/hodHemanth YamijalaVinod Kumar Vavilapalli
HADOOP-2961[HOD] Hod expects port info though external host is not mentioned.Minorcontrib/hodHemanth YamijalaVinod Kumar Vavilapalli
HADOOP-2930make {start,stop}-balancer.sh work even if hadoop-daemon.sh isn't in the PATHTrivialscriptsSpiros PapadimitriouSpiros Papadimitriou
HADOOP-2928Remove deprecated methods getContentLength() in ClientProtocol, NameNode, FileSystem, DistributedFileSystem and DFSClientBlocker.Tsz Wo Nicholas SzeLohit Vijayarenu
HADOOP-2905fsck -move triggers NPE in namenodeMajor.Michael BieniosekLohit Vijayarenu
HADOOP-2793Links for worst performing shuffle tasks are broken in Analyze Job.Minor.Amar KamatAmareshwari Sriramadasu
HADOOP-2669DFS client lost lease during writing into DFS filesMajor.Runping Qidhruba borthakur
HADOOP-2565DFSPath cache of FileStatus can become staleMajor.Doug CuttingTsz Wo Nicholas Sze
HADOOP-2427Cleanup of mapred.local.dir after maptask is completeMajor.Lohit VijayarenuAmareshwari Sriramadasu
HADOOP-2393TaskTracker locks up removing job files within a synchronized methodCritical.Joydeep Sen SarmaAmareshwari Sriramadasu
HADOOP-2294In hdfs.h , the comment says you release the result of a hdfsListDirectory with a freehdfsFileInfo, but should say hdfsFreeFileInfoTrivial.Dick KingCraig Macdonald
HADOOP-2132Killing successfully completed jobs moves them to failedCritical.Srikanth KakaniJothi Padmanabhan
HADOOP-2095Reducer failed due to Out ofMemoryMajor.Runping QiArun C Murthy
HADOOP-1318Do not fail completed maps on lost tasktrackers if ‘-reducer NONE’ is specifiedMinor.Arun C MurthyAmareshwari Sriramadasu
HADOOP-236job tracker should refuse connection from a task tracker with a different version numberMajor.Hairong KuangSharad Agarwal

TESTS:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3100Develop tests to test the DFS command line interfaceMajortestMukund MadhugiriMukund Madhugiri

SUB-TASKS:

JIRASummaryPriorityComponentReporterContributor

OTHER:

JIRASummaryPriorityComponentReporterContributor
HADOOP-3541Namespace recovery from the secondary image should be documented.BlockerdocumentationKonstantin ShvachkoKonstantin Shvachko
HADOOP-2984Distcp should have forrest documentationBlockerutilOwen O'MalleyChris Douglas
HADOOP-2632Discussion of fsck operation in the permissions regimeMajor.Robert ChanslerRobert Chansler