hadoop/hadoop-hdfs-project/hadoop-hdfs/CHANGES-HDFS-EC-7285.txt

394 lines
15 KiB
Plaintext
Raw Normal View History

BREAKDOWN OF HDFS-7285 SUBTASKS AND RELATED JIRAS
HDFS-7347. Configurable erasure coding policy for individual files and
directories ( Zhe Zhang via vinayakumarb )
HDFS-7339. Representing striped block groups in NameNode with hierarchical
naming protocol ( Zhe Zhang )
HDFS-7652. Process block reports for erasure coded blocks (Zhe Zhang)
HDFS-7716. Erasure Coding: extend BlockInfo to handle EC info (Jing Zhao)
HDFS-7749. Erasure Coding: Add striped block support in INodeFile (Jing Zhao)
HDFS-7837. Erasure Coding: allocate and persist striped blocks in NameNode
(Jing Zhao via Zhe Zhang)
HDFS-7872. Erasure Coding: INodeFile.dumpTreeRecursively() supports to print
striped blocks (Takuya Fukudome via jing9)
HDFS-7853. Erasure coding: extend LocatedBlocks to support reading from
striped files (Jing Zhao)
HDFS-7826. Erasure Coding: Update INodeFile quota computation for striped
blocks ( Kai Sasaki via jing9 )
HDFS-7912. Erasure Coding: track BlockInfo instead of Block in
UnderReplicatedBlocks and PendingReplicationBlocks (Jing Zhao)
HDFS-7369. Erasure coding: distribute recovery work for striped blocks to
DataNode (Zhe Zhang)
HDFS-7864. Erasure Coding: Update safemode calculation for striped blocks
(GAO Rui via jing9)
HDFS-7827. Erasure Coding: support striped blocks in non-protobuf fsimage
( Hui Zheng via jing9 )
HDFS-7616. Add a test for BlockGroup support in FSImage.
(Takuya Fukudome via szetszwo)
HDFS-7907. Erasure Coding: track invalid, corrupt, and under-recovery striped
blocks in NameNode (Jing Zhao)
HDFS-8005. Erasure Coding: simplify striped block recovery work computation
and add tests (Jing Zhao)
HDFS-7617. Add unit tests for editlog transactions for EC
(Hui Zheng via Zhe Zhang)
HDFS-7839. Erasure coding: implement facilities in NameNode to create and
manage EC zones (Zhe Zhang)
HDFS-7969. Erasure coding: NameNode support for lease recovery of striped
block groups. (Zhe Zhang)
HDFS-7782. Erasure coding: pread from files in striped layout.
(Zhe Zhang and Jing Zhao via Zhe Zhang)
HDFS-8023. Erasure Coding: retrieve eraure coding schema for a file from
NameNode (vinayakumarb)
HDFS-8074. Define a system-wide default EC schema. (Kai Zheng)
HDFS-8077. Erasure coding: fix bugs in EC zone and symlinks.
(Jing Zhao and Zhe Zhang via Jing Zhao)
HDFS-8104. Make hard-coded values consistent with the system default schema first before remove them. (Kai Zheng)
HDFS-7889. Subclass DFSOutputStream to support writing striping layout files. (Li Bo via Kai Zheng)
HDFS-8090. Erasure Coding: Add RPC to client-namenode to list all
ECSchemas loaded in Namenode. (vinayakumarb)
HDFS-8122. Erasure Coding: Support specifying ECSchema during creation of ECZone.
(Vinayakumar B via Zhe Zhang)
HDFS-8114. Erasure coding: Add auditlog FSNamesystem#createErasureCodingZone if this
operation fails. (Rakesh R via Zhe Zhang)
HDFS-8123. Erasure Coding: Better to move EC related proto messages to a
separate erasurecoding proto file (Rakesh R via vinayakumarb)
HDFS-7349. Support DFS command for the EC encoding (vinayakumarb)
HDFS-8120. Erasure coding: created util class to analyze striped block groups.
(Contributed by Zhe Zhang and Li Bo via Jing Zhao)
HDFS-7994. Detect if resevered EC Block ID is already used during namenode
startup. (Hui Zheng via szetszwo)
HDFS-8167. BlockManager.addBlockCollectionWithCheck should check if the block is a striped block. (Hui Zheng via zhz).
HDFS-8166. DFSStripedOutputStream should not create empty blocks. (Jing Zhao)
HDFS-7937. Erasure Coding: INodeFile quota computation unit tests.
(Kai Sasaki via Jing Zhao)
HDFS-8145. Fix the editlog corruption exposed by failed TestAddStripedBlocks.
(Jing Zhao)
HDFS-8146. Protobuf changes for BlockECRecoveryCommand and its fields for
making it ready for transfer to DN (Uma Maheswara Rao G via vinayakumarb)
HDFS-8181. createErasureCodingZone sets retryCache state as false always
(Uma Maheswara Rao G via vinayakumarb)
HDFS-8190. StripedBlockUtil.getInternalBlockLength may have overflow error.
(szetszwo)
HDFS-8216. TestDFSStripedOutputStream should use BlockReaderTestUtil to
create BlockReader. (szetszwo via Zhe Zhang)
HDFS-8212. DistributedFileSystem.createErasureCodingZone should pass schema
in FileSystemLinkResolver. (szetszwo via Zhe Zhang)
HDFS-8024. Erasure Coding: ECworker frame, basics, bootstraping and configuration.
(umamahesh)
HDFS-8156. Add/implement necessary APIs even we just have the system default
schema. (Kai Zheng via Zhe Zhang)
HDFS-8136. Client gets and uses EC schema when reads and writes a stripping
file. (Kai Sasaki via Kai Zheng)
HDFS-8233. Fix DFSStripedOutputStream#getCurrentBlockGroupBytes when the last
stripe is at the block group boundary. (jing9)
HDFS-8223. Should calculate checksum for parity blocks in DFSStripedOutputStream.
(Yi Liu via jing9)
HDFS-8228. Erasure Coding: SequentialBlockGroupIdGenerator#nextValue may cause
block id conflicts (Jing Zhao via Zhe Zhang)
HDFS-8033. Erasure coding: stateful (non-positional) read from files in
striped layout (Zhe Zhang)
HDFS-8230. Erasure Coding: Ignore DatanodeProtocol#DNA_ERASURE_CODING_RECOVERY
commands from standbynode if any (vinayakumarb)
HDFS-8189. ClientProtocol#createErasureCodingZone API was wrongly annotated
as Idempotent (vinayakumarb)
HDFS-8235. Erasure Coding: Create DFSStripedInputStream in DFSClient#open.
(Kai Sasaki via jing9)
HDFS-8272. Erasure Coding: simplify the retry logic in DFSStripedInputStream
(stateful read). (Jing Zhao via Zhe Zhang)
HDFS-8282. Erasure coding: move striped reading logic to StripedBlockUtil.
(Zhe Zhang)
HDFS-8183. Erasure Coding: Improve DFSStripedOutputStream closing of
datastreamer threads. (Rakesh R via Zhe Zhang)
HDFS-8308. Erasure Coding: NameNode may get blocked in waitForLoadingFSImage()
when loading editlog. (jing9)
HDFS-7949. WebImageViewer need support file size calculation with striped
blocks. (Rakesh R via Zhe Zhang)
HDFS-8316. Erasure coding: refactor EC constants to be consistent with HDFS-8249.
(Zhe Zhang via jing9)
HDFS-8281. Erasure Coding: implement parallel stateful reading for striped layout.
(jing9)
HDFS-8137. Send the EC schema to DataNode via EC encoding/recovering command(umamahesh)
HDFS-8242. Erasure Coding: XML based end-to-end test for ECCli commands
(Rakesh R via vinayakumarb)
HDFS-8324. Add trace info to DFSClient#getErasureCodingZoneInfo(..) (vinayakumarb via
umamahesh)
HDFS-7672. Handle write failure for stripping blocks and refactor the
existing code in DFSStripedOutputStream and StripedDataStreamer. (szetszwo)
HDFS-7348. Erasure Coding: DataNode reconstruct striped blocks.
(Yi Liu via Zhe Zhang)
HADOOP-11921. Enhance tests for erasure coders. (Kai Zheng)
HDFS-8334. Erasure coding: rename DFSStripedInputStream related test
classes. (Zhe Zhang)
HDFS-8129. Erasure Coding: Maintain consistent naming for Erasure Coding related classes - EC/ErasureCoding
(umamahesh)
HDFS-8203. Erasure Coding: Seek and other Ops in DFSStripedInputStream.
(Yi Liu via jing9)
HDFS-8289. Erasure Coding: add ECSchema to HdfsFileStatus. (Yong Zhang via
jing9)
HDFS-8355. Erasure Coding: Refactor BlockInfo and BlockInfoUnderConstruction.
(Tsz Wo Nicholas Sze via jing9)
HDFS-7678. Erasure coding: DFSInputStream with decode functionality (pread).
(Zhe Zhang)
HDFS-8372. Erasure coding: compute storage type quotas for striped files,
to be consistent with HDFS-8327. (Zhe Zhang via jing9)
HDFS-8368. Erasure Coding: DFS opening a non-existent file need to be
handled properly (Rakesh R via zhz)
HDFS-8363. Erasure Coding: DFSStripedInputStream#seekToNewSource. (yliu)
HDFS-8195. Erasure coding: Fix file quota change when we complete/commit
the striped blocks. (Takuya Fukudome via zhz)
HDFS-8364. Erasure coding: fix some minor bugs in EC CLI
(Walter Su via vinayakumarb)
HDFS-8391. NN should consider current EC tasks handling count from DN while
assigning new tasks. (umamahesh)
HDFS-8367. BlockInfoStriped uses EC schema. (Kai Sasaki via Kai Zheng)
HDFS-8352. Erasure Coding: test webhdfs read write stripe file. (waltersu4549)
HDFS-8417. Erasure Coding: Pread failed to read data starting from not-first stripe.
(Walter Su via jing9)
HDFS-8418. Fix the isNeededReplication calculation for Striped block in NN.
(Yi Liu via jing9)
HDFS-8320. Erasure coding: consolidate striping-related terminologies. (zhz)
HDFS-8366. Erasure Coding: Make the timeout parameter of polling blocking queue
configurable in DFSStripedOutputStream. (Li Bo)
HDFS-8378. Erasure Coding: Few improvements for the erasure coding worker.
(Rakesh R via waltersu4549)
HDFS-8375. Add cellSize as an XAttr to ECZone. ( Vinayakumar B via zhz).
HDFS-8428. Erasure Coding: Fix the NullPointerException when deleting file.
(Yi Liu via zhz).
HDFS-8323. Bump GenerationStamp for write faliure in DFSStripedOutputStream.
(Tsz Wo Nicholas Sze via jing9)
HDFS-8427. Remove dataBlockNum and parityBlockNum from BlockInfoStriped.
(Kai Sasaki via jing9)
HDFS-8186. Erasure coding: Make block placement policy for EC file configurable.
(Walter Su via zhz)
HDFS-8294. Erasure Coding: Fix Findbug warnings present in erasure coding.
(Rakesh R via zhz)
HDFS-8441. Erasure Coding: make condition check earlier for setReplication.
(waltersu4549)
HDFS-7768. Change fsck to support EC files. (Takanobu Asanuma via szetszwo)
HDFS-8382. Remove chunkSize and initialize from erasure coder. (Kai Zheng)
HDFS-8408. Revisit and refactor ErasureCodingInfo (vinayakumarb)
HDFS-8479. Erasure coding: fix striping related logic in FSDirWriteFileOp to
sync with HDFS-8421. (Zhe Zhang via jing9)
HDFS-8481. Erasure coding: remove workarounds in client side stripped blocks
recovering. (zhz)
HDFS-8336. Expose some administrative erasure coding operations to HdfsAdmin
(Uma Maheswara Rao G via vinayakumarb)
HDFS-8444. Erasure Coding: fix cannot rename a zone dir
(Walter Su via vinayakumarb)
HDFS-8517. Fix a decoding issue in stripped block recovering in client side.
(Kai Zheng via jing9)
HDFS-8453. Erasure coding: properly handle start offset for internal blocks
in a block group. (Zhe Zhang via jing9)
HDFS-7621. Erasure Coding: update the Balancer/Mover data migration logic.
(Walter Su via zhz)
HDFS-8328. Follow-on to update decode for DataNode striped blocks
reconstruction. (yliu)
HDFS-8319. Erasure Coding: support decoding for stateful read.
(Jing Zhao via zhz)
HDFS-8460. Erasure Coding: stateful read result doesn't match data
occasionally because of flawed test. (Walter Su via zhz)
HDFS-8556. Erasure Coding: Fix usage of 'createZone' (vinayakumarb)
HDFS-8571. Fix TestErasureCodingCli test (Vinayakumar B via waltersu4549)
HDFS-8450. Erasure Coding: Consolidate erasure coding zone related
implementation into a single class (Rakesh R via vinayakumarb)
HDFS-8585. Erasure Coding: Remove dataBlockNum and parityBlockNum from
StripedBlockProto. (Yi Liu via jing9)
HDFS-8559. Erasure Coding: fix non-protobuf fsimage for striped blocks.
(Jing Zhao via yliu)
HDFS-8580. Erasure coding: Persist cellSize in BlockInfoStriped and
StripedBlocksFeature. (Walter Su via jing9)
HDFS-8466. Refactor BlockInfoContiguous and fix NPE in
TestBlockInfo#testCopyConstructor() (vinayakumarb)
HDFS-8254. Avoid assigning a leading streamer in StripedDataStreamer to
tolerate datanode failure. (Tsz Wo Nicholas Sze via jing9)
HDFS-8543. Erasure Coding: processOverReplicatedBlock() handles striped block.
(Walter Su via jing9)
HDFS-8602. Erasure Coding: Client can't read(decode) the EC files which have
corrupt blocks. (jing9 and Kai Sasaki)
HDFS-8567. Erasure Coding: SafeMode handles file smaller than a full stripe.
(Walter Su via jing9)
HDFS-8253. DFSStripedOutputStream.closeThreads releases cellBuffers
multiple times. (Kai Sasaki via szetszwo)
HDFS-8468. 2 RPC calls for every file read in DFSClient#open(..) resulting in
double Audit log entries (vinayakumarb)
HDFS-8684. Erasure Coding: fix some block number calculation for striped
block. (yliu)
HDFS-8461. Erasure coding: fix priority level of UnderReplicatedBlocks for
striped block. (Walter Su via jing9)
HDFS-8719. Erasure Coding: client generates too many small packets when
writing parity data. (Li Bo via waltersu4549)
HDFS-8563. Erasure Coding: fsck handles file smaller than a full stripe.
(Walter Su via jing9)
HDFS-8484. Erasure coding: Two contiguous blocks occupy IDs belong to same
striped group. (Walter Su via jing9)
HDFS-8744. Erasure Coding: the number of chunks in packet is not updated
when writing parity data. (Li Bo)
HDFS-8669. Erasure Coding: handle missing internal block locations in
DFSStripedInputStream. (jing9)
HDFS-8702. Erasure coding: update BlockManager.blockHasEnoughRacks(..) logic
for striped block. (Kai Sasaki via jing9)
HDFS-8734. Erasure Coding: fix one cell need two packets. (Walter Su via
jing9)
HDFS-8619. Erasure Coding: revisit replica counting for striped blocks.
(Jing Zhao via yliu)
HDFS-8058. Erasure coding: use BlockInfo[] for both striped and contiguous
blocks in INodeFile. (Zhe Zhang and Yi Liu via zhz)
HDFS-8787. Erasure coding: rename BlockInfoContiguousUC and BlockInfoStripedUC
to be consistent with trunk. (zhz)
HDFS-8433. Erasure coding: set blockToken in LocatedStripedBlock.(waltersu4549)
HDFS-8760. Erasure Coding: reuse BlockReader when reading the same block in pread.
(jing9)
HDFS-8781. Erasure Coding: Correctly handle BlockManager#InvalidateBlocks for
striped block. (Yi Liu via jing9)
HDFS-8813. Erasure Coding: Client no need to decode missing parity blocks.
(Walter Su via jing9)
HDFS-8798. Erasure Coding: fix DFSStripedInputStream/DFSStripedOutputStream
re-fetch token when expired. (Walter Su via jing9)
HDFS-8769. Erasure coding: unit test for SequentialBlockGroupIdGenerator.
(Rakesh R via waltersu4549)
HDFS-8202. Improve end to end stirpping file test to add erasure recovering
test. (Xinwei Qin via zhz)
HDFS-8804. Erasure Coding: use DirectBufferPool in DFSStripedInputStream for
buffer allocation. (jing9)
HDFS-8399. Erasure Coding: unit test the behaviour of BlockManager recovery
work for the deleted blocks. (Rakesh R via zhz)
HDFS-8857. Erasure Coding: Fix ArrayIndexOutOfBoundsException in
TestWriteStripedFileWithFailure. (Li Bo)