Hadoop 2.6.1 Release Notes

• YARN-4047. Major bug reported by Jason Lowe and fixed by Jason Lowe (resourcemanager)
  ClientRMService getApplications has high scheduler lock contention
  

• YARN-3999. Major bug reported by Jian He and fixed by Jian He
  RM hangs on draining events
  

• YARN-3990. Critical bug reported by Rohith Sharma K S and fixed by Bibin A Chundatt (resourcemanager)
  AsyncDispatcher may overloaded with RMAppNodeUpdateEvent when Node is connected/disconnected
  

• YARN-3978. Major improvement reported by Eric Payne and fixed by Eric Payne (timelineserver , yarn)
  Configurably turn off the saving of container info in Generic AHS
  

• YARN-3850. Blocker bug reported by Varun Saxena and fixed by Varun Saxena (log-aggregation , nodemanager)
  NM fails to read files from full disks which can lead to container logs being lost and other issues
  

• YARN-3832. Critical bug reported by Ranga Swamy and fixed by Brahma Reddy Battula (nodemanager)
  Resource Localization fails on a cluster due to existing cache directories
  

• YARN-3733. Blocker bug reported by Bibin A Chundatt and fixed by Rohith Sharma K S (resourcemanager)
  Fix DominantRC#compare() does not work as expected if cluster resource is empty
  

• YARN-3725. Blocker bug reported by Zhijie Shen and fixed by Zhijie Shen (resourcemanager , timelineserver)
  App submission via REST API is broken in secure mode due to Timeline DT service address is empty
  

• YARN-3700. Major sub-task reported by Xuan Gong and fixed by Xuan Gong (resourcemanager , webapp , yarn)
  ATS Web Performance issue at load time when large number of jobs
  

• YARN-3585. Critical bug reported by Peng Zhang and fixed by Rohith Sharma K S
  NodeManager cannot exit on SHUTDOWN event triggered and NM recovery is enabled
  

• YARN-3526. Major bug reported by Weiwei Yang and fixed by Weiwei Yang (resourcemanager , webapp)
  ApplicationMaster tracking URL is incorrectly redirected on a QJM cluster
  

• YARN-3493. Critical bug reported by Sumana Sathish and fixed by Jian He (yarn)
  RM fails to come up with error "Failed to load/recover state" when mem settings are changed
  

• YARN-3487. Critical sub-task reported by Jason Lowe and fixed by Jason Lowe (capacityscheduler)
  CapacityScheduler scheduler lock obtained unnecessarily when calling getQueue
  

• YARN-3464. Critical bug reported by zhihai xu and fixed by zhihai xu (nodemanager)
  Race condition in LocalizerRunner kills localizer before localizing all resources
  

• YARN-3393. Critical bug reported by Zhijie Shen and fixed by Zhijie Shen (timelineserver)
  Getting application(s) goes wrong when app finishes before starting the attempt
  

• YARN-3369. Blocker bug reported by Giovanni Matteo Fumarola and fixed by Brahma Reddy Battula (resourcemanager)
  Missing NullPointer check in AppSchedulingInfo causes RM to die
  

• YARN-3287. Major bug reported by Jonathan Eagles and fixed by Daryn Sharp
  TimelineClient kerberos authentication failure uses wrong login context.
  

• YARN-3267. Major bug reported by Prakash Ramachandran and fixed by Chang Li
  Timelineserver applies the ACL rules after applying the limit on the number of records
  

• YARN-3251. Blocker bug reported by Jason Lowe and fixed by Craig Welch
  Fix CapacityScheduler deadlock when computing absolute max avail capacity (short term fix for 2.6.1)
  

• YARN-3249. Minor improvement reported by Ryu Kobayashi and fixed by Ryu Kobayashi (resourcemanager)
  Add a "kill application" button to Resource Manager's Web UI
  

• YARN-3242. Critical bug reported by zhihai xu and fixed by zhihai xu (resourcemanager)
  Asynchrony in ZK-close can lead to ZKRMStateStore watcher receiving events for old client
  

• YARN-3239. Major bug reported by Hitesh Shah and fixed by Jian He
  WebAppProxy does not support a final tracking url which has query fragments and params
  

• YARN-3238. Blocker bug reported by Jason Lowe and fixed by Jason Lowe
  Connection timeouts to nodemanagers are retried at multiple levels
  

• YARN-3231. Critical bug reported by Siqi Li and fixed by Siqi Li
  FairScheduler: Changing queueMaxRunningApps interferes with pending jobs
  

• YARN-3230. Major improvement reported by Jian He and fixed by Jian He
  Clarify application states on the web UI
  

• YARN-3227. Critical bug reported by Jonathan Eagles and fixed by Zhijie Shen
  Timeline renew delegation token fails when RM user's TGT is expired
  

• YARN-3222. Critical bug reported by Rohith Sharma K S and fixed by Rohith Sharma K S (resourcemanager)
  RMNodeImpl#ReconnectNodeTransition should send scheduler events in sequential order
  

• YARN-3207. Major bug reported by Prakash Ramachandran and fixed by Zhijie Shen (timelineserver)
  secondary filter matches entites which do not have the key being filtered for.
  

• YARN-3124. Major sub-task reported by Wangda Tan and fixed by Wangda Tan (api , client , resourcemanager)
  Capacity Scheduler LeafQueue/ParentQueue should use QueueCapacities to track capacities-by-label
  

• YARN-3103. Blocker bug reported by Jason Lowe and fixed by Jason Lowe (client)
  AMRMClientImpl does not update AMRM token properly
  

• YARN-3099. Major sub-task reported by Wangda Tan and fixed by Wangda Tan (api , client , resourcemanager)
  Capacity Scheduler LeafQueue/ParentQueue should use ResourceUsage to track used-resources-by-label.
  

• YARN-3098. Major sub-task reported by Wangda Tan and fixed by Wangda Tan (capacityscheduler)
  Create common QueueCapacities class in Capacity Scheduler to track capacities-by-labels of queues
  

• YARN-3094. Major bug reported by Jun Gong and fixed by Jun Gong (resourcemanager)
  reset timer for liveness monitors after RM recovery
  

• YARN-3092. Major sub-task reported by Wangda Tan and fixed by Wangda Tan (api , client , resourcemanager)
  Create common ResourceUsage class to track labeled resource usages in Capacity Scheduler
  

• YARN-3055. Blocker bug reported by Yi Liu and fixed by Daryn Sharp (security)
  The token is not renewed properly if it's shared by jobs (oozie) in DelegationTokenRenewer
  

• YARN-3024. Major bug reported by Chengbing Liu and fixed by Chengbing Liu (nodemanager)
  LocalizerRunner should give DIE action when all resources are localized
  

• YARN-3011. Major sub-task reported by Wang Hao and fixed by Varun Saxena (nodemanager)
  NM dies because of the failure of resource localization
  

• YARN-2997. Major bug reported by Chengbing Liu and fixed by Chengbing Liu (nodemanager)
  NM keeps sending already-sent completed containers to RM until containers are removed from context
  

• YARN-2992. Blocker bug reported by Karthik Kambatla and fixed by Karthik Kambatla (resourcemanager)
  ZKRMStateStore crashes due to session expiry
  

• YARN-2978. Critical bug reported by Jason Tufo and fixed by Varun Saxena
  ResourceManager crashes with NPE while getting queue info
  

• YARN-2964. Blocker bug reported by Daryn Sharp and fixed by Jian He (resourcemanager)
  RM prematurely cancels tokens for jobs that submit jobs (oozie)
  

• YARN-2952. Major bug reported by Jian He and fixed by Rohith Sharma K S
  Incorrect version check in RMStateStore
  

• YARN-2922. Major bug reported by Jason Tufo and fixed by Rohith Sharma K S (capacityscheduler , resourcemanager , scheduler)
  ConcurrentModificationException in CapacityScheduler's LeafQueue
  

• YARN-2920. Major sub-task reported by Wangda Tan and fixed by Wangda Tan
  CapacityScheduler should be notified when labels on nodes changed
  

• YARN-2918. Major sub-task reported by Rohith Sharma K S and fixed by Wangda Tan (resourcemanager)
  Don't fail RM if queue's configured labels are not existed in cluster-node-labels
  

• YARN-2917. Critical bug reported by Rohith Sharma K S and fixed by Rohith Sharma K S (resourcemanager)
  Potential deadlock in AsyncDispatcher when system.exit called in AsyncDispatcher#dispatch and AsyscDispatcher#serviceStop from shutdown hook
  

• YARN-2910. Major bug reported by Wilfred Spiegelenburg and fixed by Wilfred Spiegelenburg (fairscheduler)
  FSLeafQueue can throw ConcurrentModificationException
  

• YARN-2906. Major bug reported by Jason Lowe and fixed by Jason Lowe (capacityscheduler)
  CapacitySchedulerPage shows HTML tags for a queue's Active Users
  

• YARN-2905. Blocker bug reported by Jason Lowe and fixed by Varun Saxena
  AggregatedLogsBlock page can infinitely loop if the aggregated log file is corrupted
  

• YARN-2900. Major sub-task reported by Jonathan Eagles and fixed by Mit Desai (timelineserver)
  Application (Attempt and Container) Not Found in AHS results in Internal Server Error (500)
  

• YARN-2894. Major bug reported by Rohith Sharma K S and fixed by Rohith Sharma K S (resourcemanager)
  When ACL's are enabled, if RM switches then application can not be viewed from web.
  

• YARN-2890. Major bug reported by Mit Desai and fixed by Mit Desai
  MiniYarnCluster should turn on timeline service if configured to do so
  

• YARN-2874. Blocker bug reported by Naganarasimha G R and fixed by Naganarasimha G R (resourcemanager)
  Dead lock in "DelegationTokenRenewer" which blocks RM to execute any further apps
  

• YARN-2865. Critical bug reported by Rohith Sharma K S and fixed by Rohith Sharma K S (resourcemanager)
  Application recovery continuously fails with "Application with id already present. Cannot duplicate"
  

• YARN-2856. Critical bug reported by Rohith Sharma K S and fixed by Rohith Sharma K S (resourcemanager)
  Application recovery throw InvalidStateTransitonException: Invalid event: ATTEMPT_KILLED at ACCEPTED
  

• YARN-2816. Major bug reported by zhihai xu and fixed by zhihai xu (nodemanager)
  NM fail to start with NPE during container recovery
  

• YARN-2766. Major sub-task reported by Robert Kanter and fixed by Robert Kanter (timelineserver)
  ApplicationHistoryManager is expected to return a sorted list of apps/attempts/containers
  

• YARN-2694. Major sub-task reported by Wangda Tan and fixed by Wangda Tan (capacityscheduler , resourcemanager)
  Ensure only single node labels specified in resource request / host, and node label expression only specified when resourceName=ANY
  

• YARN-2637. Critical bug reported by Wangda Tan and fixed by Craig Welch (resourcemanager)
  maximum-am-resource-percent could be respected for both LeafQueue/User when trying to activate applications.
  

• YARN-2414. Major bug reported by Zhijie Shen and fixed by Wangda Tan (webapp)
  RM web UI: app page will crash if app is failed before any attempt has been created
  

• YARN-2340. Critical bug reported by Nishan Shetty and fixed by Rohith Sharma K S (resourcemanager , scheduler)
  NPE thrown when RM restart after queue is STOPPED. There after RM can not recovery application's and remain in standby
  

• YARN-2301. Major improvement reported by Jian He and fixed by Naganarasimha G R
  Improve yarn container command
  

• YARN-2246. Major bug reported by Devaraj K and fixed by Devaraj K (webapp)
  Job History Link in RM UI is redirecting to the URL which contains Job Id twice
  

• YARN-1984. Major sub-task reported by Jason Lowe and fixed by Varun Saxena
  LeveldbTimelineStore does not handle db exceptions properly
  

• YARN-1809. Major sub-task reported by Zhijie Shen and fixed by Xuan Gong
  Synchronize RM and Generic History Service Web-UIs
  

• MAPREDUCE-6361. Critical bug reported by Junping Du and fixed by Junping Du
  NPE issue in shuffle caused by concurrent issue between copySucceeded() in one thread and copyFailed() in another thread on the same host
  

• MAPREDUCE-6324. Blocker bug reported by Jason Lowe and fixed by Jason Lowe (mr-am)
  Uber jobs fail to update AMRM token when it rolls over
  

• MAPREDUCE-6303. Blocker bug reported by Jason Lowe and fixed by Jason Lowe
  Read timeout when retrying a fetch error can be fatal to a reducer
  

• MAPREDUCE-6300. Minor bug reported by Siqi Li and fixed by Siqi Li
  Task list sort by task id broken
  

• MAPREDUCE-6267. Minor improvement reported by Chris Trezzo and fixed by Chris Trezzo
  Refactor JobSubmitter#copyAndConfigureFiles into it's own class
  

• MAPREDUCE-6238. Critical bug reported by zhihai xu and fixed by zhihai xu (mrv2)
  MR2 can't run local jobs with -libjars command options which is a regression from MR1
  

• MAPREDUCE-6237. Major bug reported by Kannan Rajah and fixed by Kannan Rajah (mrv2)
  Multiple mappers with DBInputFormat don't work because of reusing conections
  

• MAPREDUCE-6230. Blocker bug reported by Jason Lowe and fixed by Jason Lowe (mr-am)
  MR AM does not survive RM restart if RM activated a new AMRM secret key
  

• MAPREDUCE-6166. Major bug reported by Eric Payne and fixed by Eric Payne (mrv2)
  Reducers do not validate checksum of map outputs when fetching directly to disk
  

• MAPREDUCE-5649. Major bug reported by stanley shi and fixed by Gera Shegalov (mrv2)
  Reduce cannot use more than 2G memory for the final merge
  

• HDFS-8863. Critical bug reported by Kihwal Lee and fixed by Kihwal Lee
  The remaining space check in BlockPlacementPolicyDefault is flawed
  

• HDFS-8846. Major bug reported by Zhe Zhang and fixed by Zhe Zhang (namenode)
  Add a unit test for INotify functionality across a layout version upgrade
  

• HDFS-8486. Blocker bug reported by Daryn Sharp and fixed by Daryn Sharp (datanode)
  DN startup may cause severe data loss
  

  Public service notice: - Every restart of a 2.6.x or 2.7.0 DN incurs a risk of unwanted block deletion. - Apply this patch if you are running a pre-2.7.1 release.

• HDFS-8480. Critical bug reported by Zhe Zhang and fixed by Zhe Zhang
  Fix performance and timeout issues in HDFS-7929 by using hard-links to preserve old edit logs instead of copying them
  

• HDFS-8431. Critical bug reported by Sumana Sathish and fixed by Anu Engineer (HDFS)
  hdfs crypto class not found in Windows
  

• HDFS-8404. Major bug reported by Nathan Roberts and fixed by Nathan Roberts (namenode)
  Pending block replication can get stuck using older genstamp
  

• HDFS-8384. Minor improvement reported by Tsz Wo Nicholas Sze and fixed by Jing Zhao (namenode)
  Allow NN to startup if there are files having a lease but are not under construction
  

• HDFS-8270. Major bug reported by Andrey Stepachev and fixed by J.Andreina (hdfs-client)
  create() always retried with hardcoded timeout when file already exists with open lease
  

  Proxy level retries will not be done on AlreadyBeingCreatedExeption for create() op.

• HDFS-8245. Major bug reported by Rushabh S Shah and fixed by Rushabh S Shah
  Standby namenode doesn't process DELETED_BLOCK if the add block request is in edit log.
  

• HDFS-8219. Major bug reported by Peter Shi and fixed by Surendra Singh Lilhore
  setStoragePolicy with folder behavior is different after cluster restart
  

• HDFS-8127. Blocker bug reported by Jing Zhao and fixed by Jing Zhao (ha)
  NameNode Failover during HA upgrade can cause DataNode to finalize upgrade
  

• HDFS-8072. Major bug reported by Arpit Agarwal and fixed by Arpit Agarwal (datanode)
  Reserved RBW space is not released if client terminates while writing block
  

• HDFS-8070. Blocker bug reported by Gopal V and fixed by Colin Patrick McCabe (caching)
  Pre-HDFS-7915 DFSClient cannot use short circuit on post-HDFS-7915 DataNode
  

• HDFS-8046. Major bug reported by Kihwal Lee and fixed by Kihwal Lee
  Allow better control of getContentSummary
  

• HDFS-7999. Major bug reported by zhouyingchao and fixed by zhouyingchao
  FsDatasetImpl#createTemporary sometimes holds the FSDatasetImpl lock for a very long time
  

• HDFS-7980. Major bug reported by Hui Zheng and fixed by Walter Su
  Incremental BlockReport will dramatically slow down the startup of a namenode
  

• HDFS-7960. Critical bug reported by Lei (Eddy) Xu and fixed by Colin Patrick McCabe
  The full block report should prune zombie storages even if they're not empty
  

• HDFS-7930. Blocker bug reported by Konstantin Shvachko and fixed by Yi Liu (namenode)
  commitBlockSynchronization() does not remove locations
  

• HDFS-7929. Major bug reported by Zhe Zhang and fixed by Zhe Zhang
  inotify unable fetch pre-upgrade edit log segments once upgrade starts
  

• HDFS-7915. Major bug reported by Colin Patrick McCabe and fixed by Colin Patrick McCabe
  The DataNode can sometimes allocate a ShortCircuitShm slot and fail to tell the DFSClient about it because of a network error
  

• HDFS-7894. Critical bug reported by Kihwal Lee and fixed by Brahma Reddy Battula
  Rolling upgrade readiness is not updated in jmx until query command is issued.
  

• HDFS-7885. Critical bug reported by vitthal (Suhas) Gogate and fixed by Tsz Wo Nicholas Sze (datanode)
  Datanode should not trust the generation stamp provided by client
  

• HDFS-7884. Blocker bug reported by Tsz Wo Nicholas Sze and fixed by Brahma Reddy Battula (datanode)
  NullPointerException in BlockSender
  

• HDFS-7871. Critical bug reported by Jing Zhao and fixed by Jing Zhao
  NameNodeEditLogRoller can keep printing "Swallowing exception" message
  

• HDFS-7830. Major bug reported by Lei (Eddy) Xu and fixed by Lei (Eddy) Xu (datanode)
  DataNode does not release the volume lock when adding a volume fails.
  

• HDFS-7788. Blocker bug reported by Kihwal Lee and fixed by Rushabh S Shah
  Post-2.6 namenode may not start up with an image containing inodes created with an old release.
  

• HDFS-7763. Major bug reported by Liang Xie and fixed by Liang Xie (ha)
  fix zkfc hung issue due to not catching exception in a corner case
  

• HDFS-7742. Major bug reported by Nathan Roberts and fixed by Nathan Roberts (namenode)
  favoring decommissioning node for replication can cause a block to stay underreplicated for long periods
  

• HDFS-7733. Major bug reported by Arpit Agarwal and fixed by Arpit Agarwal (nfs)
  NFS: readdir/readdirplus return null directory attribute on failure
  

• HDFS-7714. Major bug reported by Chris Nauroth and fixed by Vinayakumar B (datanode)
  Simultaneous restart of HA NameNodes and DataNode can cause DataNode to register successfully with only one NameNode.
  

• HDFS-7707. Major bug reported by Yongjun Zhang and fixed by Yongjun Zhang (namenode)
  Edit log corruption due to delayed block removal again
  

• HDFS-7610. Major bug reported by Lei (Eddy) Xu and fixed by Lei (Eddy) Xu (datanode)
  Fix removal of dynamically added DN volumes
  

• HDFS-7609. Critical bug reported by Carrey Zhan and fixed by Ming Ma (namenode)
  Avoid retry cache collision when Standby NameNode loading edits
  

• HDFS-7596. Major bug reported by Arpit Agarwal and fixed by Arpit Agarwal (namenode)
  NameNode should prune dead storages from storageMap
  

• HDFS-7587. Blocker bug reported by Kihwal Lee and fixed by Jing Zhao (namenode)
  Edit log corruption can happen if append fails with a quota violation
  

• HDFS-7579. Minor improvement reported by Charles Lamb and fixed by Charles Lamb (datanode)
  Improve log reporting during block report rpc failure
  

• HDFS-7575. Critical bug reported by Lars Francke and fixed by Arpit Agarwal
  Upgrade should generate a unique storage ID for each volume
  

• HDFS-7552. Major bug reported by Liang Xie and fixed by Liang Xie (datanode , test)
  change FsVolumeList toString() to fix TestDataNodeVolumeFailureToleration
  

• HDFS-7533. Major bug reported by Kihwal Lee and fixed by Eric Payne
  Datanode sometimes does not shutdown on receiving upgrade shutdown command
  

• HDFS-7531. Major improvement reported by Lei (Eddy) Xu and fixed by Lei (Eddy) Xu (datanode)
  Improve the concurrent access on FsVolumeList
  

• HDFS-7503. Major bug reported by Arpit Agarwal and fixed by Arpit Agarwal (namenode)
  Namenode restart after large deletions can cause slow processReport (due to logging)
  

• HDFS-7489. Critical bug reported by Noah Lorang and fixed by Noah Lorang (datanode)
  Incorrect locking in FsVolumeList#checkDirs can hang datanodes
  

• HDFS-7470. Major bug reported by zhaoyunjiong and fixed by zhaoyunjiong (namenode)
  SecondaryNameNode need twice memory when calling reloadFromImageFile
  

• HDFS-7446. Major improvement reported by Colin Patrick McCabe and fixed by Colin Patrick McCabe (hdfs-client)
  HDFS inotify should have the ability to determine what txid it has read up to
  

• HDFS-7443. Blocker bug reported by Kihwal Lee and fixed by Colin Patrick McCabe
  Datanode upgrade to BLOCKID_BASED_LAYOUT fails if duplicate block files are present in the same volume
  

• HDFS-7425. Minor bug reported by Chris Nauroth and fixed by Chris Nauroth (namenode)
  NameNode block deletion logging uses incorrect appender.
  

• HDFS-7314. Major improvement reported by Ming Ma and fixed by Ming Ma
  When the DFSClient lease cannot be renewed, abort open-for-write files rather than the entire DFSClient
  

• HDFS-7278. Major improvement reported by Colin Patrick McCabe and fixed by Colin Patrick McCabe (datanode)
  Add a command that allows sysadmins to manually trigger full block reports from a DN
  

• HDFS-7263. Major bug reported by Konstantin Shvachko and fixed by Tao Luo (hdfs-client)
  Snapshot read can reveal future bytes for appended files.
  

• HDFS-7235. Major bug reported by Yongjun Zhang and fixed by Yongjun Zhang (datanode , namenode)
  DataNode#transferBlock should report blocks that don't exist using reportBadBlock
  

• HDFS-7225. Major bug reported by Zhe Zhang and fixed by Zhe Zhang (namenode)
  Remove stale block invalidation work when DN re-registers with different UUID
  

• HDFS-7213. Critical bug reported by Daryn Sharp and fixed by Eric Payne (namenode)
  processIncrementalBlockReport performance degradation
  

• HDFS-7182. Major improvement reported by Ming Ma and fixed by Ming Ma
  JMX metrics aren't accessible when NN is busy
  

• HDFS-7035. Major sub-task reported by Lei (Eddy) Xu and fixed by Lei (Eddy) Xu (datanode)
  Make adding a new data directory to the DataNode an atomic operation and improve error handling
  

• HDFS-7009. Major bug reported by Ming Ma and fixed by Ming Ma (datanode)
  Active NN and standby NN have different live nodes
  

• HDFS-4882. Critical bug reported by Zesheng Wu and fixed by Ravi Prakash (hdfs-client , namenode)
  Prevent the Namenode's LeaseManager from looping forever in checkLeases
  

• HDFS-3443. Major bug reported by suja s and fixed by Vinayakumar B (auto-failover , ha)
  Fix NPE when namenode transition to active during startup by adding checkNNStartup() in NameNodeRpcServer
  

• HADOOP-12280. Minor improvement reported by Masatake Iwasaki and fixed by Masatake Iwasaki (test)
  Skip unit tests based on maven profile rather than NativeCodeLoader.isNativeCodeLoaded
  

• HADOOP-11934. Blocker bug reported by Mike Yoder and fixed by Larry McCay (security)
  Use of JavaKeyStoreProvider in LdapGroupsMapping causes infinite loop
  

• HADOOP-11932. Critical bug reported by Jian He and fixed by Brahma Reddy Battula
  MetricsSinkAdapter hangs when being stopped
  

• HADOOP-11812. Blocker improvement reported by Gera Shegalov and fixed by Gera Shegalov (fs)
  Implement listLocatedStatus for ViewFileSystem to speed up split calculation
  

• HADOOP-11802. Major bug reported by Eric Payne and fixed by Colin Patrick McCabe
  DomainSocketWatcher thread terminates sometimes after there is an I/O error during requestShortCircuitShm
  

• HADOOP-11730. Major bug reported by Takenori Sato and fixed by Takenori Sato (fs/s3)
  Regression: s3n read failure recovery broken
  

• HADOOP-11710. Critical sub-task reported by Sean Busbey and fixed by Sean Busbey (fs)
  Make CryptoOutputStream behave like DFSOutputStream wrt synchronization
  

• HADOOP-11674. Critical bug reported by Sean Busbey and fixed by Sean Busbey (io)
  oneByteBuf in CryptoInputStream and CryptoOutputStream should be non static
  

• HADOOP-11604. Critical bug reported by Liang Xie and fixed by Chris Nauroth (net)
  Prevent ConcurrentModificationException while closing domain sockets during shutdown of DomainSocketWatcher thread.
  

• HADOOP-11506. Major improvement reported by Dmitriy V. Ryaboy and fixed by Gera Shegalov (conf)
  Configuration variable expansion regex expensive for long values
  

• HADOOP-11491. Critical bug reported by Gera Shegalov and fixed by Brahma Reddy Battula (fs)
  HarFs incorrectly declared as requiring an authority
  

• HADOOP-11482. Major bug reported by Arun Suresh and fixed by Arun Suresh
  Use correct UGI when KMSClientProvider is called by a proxy user
  

• HADOOP-11466. Minor improvement reported by Suman Somasundar and fixed by Suman Somasundar (io , performance , util)
  FastByteComparisons: do not use UNSAFE_COMPARER on the SPARC architecture because it is slower there
  

• HADOOP-11368. Major bug reported by Arun Suresh and fixed by Arun Suresh (kms)
  Fix SSLFactory truststore reloader thread leak in KMSClientProvider
  

• HADOOP-11350. Major bug reported by Benoy Antony and fixed by Benoy Antony (security)
  The size of header buffer of HttpServer is too small when HTTPS is enabled
  

• HADOOP-11343. Blocker bug reported by Jerry Chen and fixed by Jerry Chen (security)
  Overflow is not properly handled in caclulating final iv for AES CTR
  

• HADOOP-11333. Major bug reported by zhaoyunjiong and fixed by zhaoyunjiong
  Fix deadlock in DomainSocketWatcher when the notification pipe is full
  

• HADOOP-11316. Blocker bug reported by Tsuyoshi Ozawa and fixed by Tsuyoshi Ozawa
  "mvn package -Pdist,docs -DskipTests -Dtar" fails because of non-ascii characters
  

• HADOOP-11295. Major bug reported by Ming Ma and fixed by Ming Ma
  RPC Server Reader thread can't shutdown if RPCCallQueue is full
  

• HADOOP-11238. Minor bug reported by Chris Li and fixed by Chris Li
  Update the NameNode's Group Cache in the background when possible
  

• HADOOP-10786. Major bug reported by Tobi Vollebregt and fixed by Stephen Chu (security)
  Fix UGI#reloginFromKeytab on Java 8
  

• HADOOP-8151. Major bug reported by Todd Lipcon and fixed by Matt Foley (io , native)
  Error handling in snappy decompressor throws invalid exceptions
  

• HADOOP-7139. Major improvement reported by Stephen Rose and fixed by Kanaka Kumar Avvaru (io)
  Allow appending to existing SequenceFiles
  

  Existing sequence files can be appended.