big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
17,782 Commits 2 Branches 2 Tags 264 MiB
b061215ecf
Commit Graph

478 Commits

Author SHA1 Message Date
Andrew Wang
a12f09ba3c HDFS-12447. Rename AddECPolicyResponse to AddErasureCodingPolicyResponse. Contributed by SammiChen. 2017-09-20 11:51:17 -07:00
Ming Ma
230b85d586 HDFS-12473. Change hosts JSON file format. 2017-09-20 09:03:59 -07:00
Brahma Reddy Battula
fda1221c55 HDFS-11799. Introduce a config to allow setting up write pipeline with fewer nodes than replication factor. Contributed by Brahma Reddy Battula 2017-09-19 11:25:45 +05:30
Kai Zheng
dba7a7dd9d HDFS-12479. Some misuses of lock in DFSStripedOutputStream. Contributed by Huafeng Wang 2017-09-19 17:45:41 +08:00
Lei Xu
fbe06b5880 HDFS-12349. Improve log message when it could not alloc enough blocks for EC. (Lei (Eddy) Xu) 2017-09-15 12:12:42 -07:00
Andrew Wang
390c2b5df0 Revert "HDFS-11156. Add new op GETFILEBLOCKLOCATIONS to WebHDFS REST API. Contributed by Weiwei Yang." 
This reverts commit 7fcc73fc0d.

 Conflicts:
	hadoop-hdfs-project/hadoop-hdfs/src/site/markdown/WebHDFS.md
	hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/web/TestWebHDFS.java
 2017-09-14 15:12:26 -07:00
Arpit Agarwal
65a941008d Revert "HDFS-10391. Always enable NameNode service RPC port. Contributed by Gergely Novak." 
This reverts commit aa4b6fbe75.
 2017-09-14 11:17:08 -07:00
Andrew Wang
f4b6267465 HDFS-12222. Document and test BlockLocation for erasure-coded files. Contributed by Huafeng Wang. 2017-09-12 17:35:30 -07:00
Lei Xu
de9994bd89 Revert "HDFS-12349. Improve log message when it could not alloc enough blocks for EC. (lei)" 
This reverts commit 3e6d0ca2b2.
 2017-09-11 10:06:05 -07:00
Arpit Agarwal
aa4b6fbe75 HDFS-10391. Always enable NameNode service RPC port. Contributed by Gergely Novak. 2017-09-09 08:40:53 -07:00
Lei Xu
3e6d0ca2b2 HDFS-12349. Improve log message when it could not alloc enough blocks for EC. (lei) 2017-09-07 18:01:37 -07:00
Andrew Wang
4e50dc976a HDFS-12218. Addendum. Rename split EC / replicated block metrics in BlockManager. 2017-09-07 16:57:19 -07:00
Andrew Wang
40c2f31f8d HDFS-12218. Rename split EC / replicated block metrics in BlockManager. 2017-09-07 16:56:35 -07:00
Kai Zheng
2adf8bed71 HDFS-12402. Refactor ErasureCodingPolicyManager and related codes. Contributed by Sammi Chen 2017-09-07 20:38:23 +08:00
Kai Zheng
d7f27043ce HDFS-12392. Writing striped file failed due to different cell size. Contributed by Sammi Chen 2017-09-06 13:29:52 +08:00
Andrew Wang
ccd2ac60ec HDFS-11882. Precisely calculate acked length of striped block groups in updatePipeline. 2017-09-05 14:16:03 -07:00
Kai Zheng
5dba54596a HDFS-12388. A bad error message in DFSStripedOutputStream. Contributed by Huafeng Wang 2017-09-05 17:46:07 +08:00
Kai Zheng
7a96033b15 HDFS-11964. Decoding inputs should be correctly prepared in pread. Contributed by Takanobu Asanuma 2017-09-01 17:48:26 +08:00
Mingliang Liu
36f33a1efb HDFS-12380. Simplify dataQueue.wait condition logical operation in DataStreamer::run(). Contributed by liaoyuxiangqin 2017-08-31 23:18:48 -07:00
Rakesh Radhakrishnan
200b11368d HDFS-12258. ec -listPolicies should list all policies in system, no matter it's enabled or disabled. Contributed by Wei Zhou. 2017-08-30 12:58:56 +05:30
Andrew Wang
f29a0fc288 HDFS-12303. Change default EC cell size to 1MB for better performance. Contributed by Wei Zhou. 2017-08-25 14:14:23 -07:00
Kihwal Lee
8455d70756 HDFS-12299. Race Between update pipeline and DN Re-Registration 2017-08-25 14:49:29 -05:00
John Zhuge
3123895db0 HDFS-12318. Fix IOException condition for openInfo in DFSInputStream. Contributed by legend. 2017-08-24 00:26:37 -07:00
Xiao Chen
1000a2af04 HDFS-10899. Add functionality to re-encrypt EDEKs. 2017-08-23 17:06:16 -07:00
John Zhuge
b6bfb2fcb2 HDFS-11738. Hedged pread takes more time when block moved from initial locations. Contributed by Vinayakumar B. 2017-08-21 13:45:30 -07:00
Andrew Wang
96b3a6b972 HDFS-11082. Provide replicated EC policy to replicate files. Contributed by SammiChen. 2017-08-16 22:17:06 -07:00
Akira Ajisaka
08aaa4b36f
HDFS-12269. Better to return a Map rather than HashMap in getErasureCodingCodecs. Contributed by Huafeng Wang. 2017-08-17 13:20:27 +09:00
Yiqun Lin
2e43c28e01 HDFS-11696. Fix warnings from Spotbugs in hadoop-hdfs. Contributed by Yiqun Lin. 2017-08-15 16:48:49 +08:00
Chris Douglas
645a8f2a4d HADOOP-14726. Mark FileStatus::isDir as final 2017-08-14 21:57:20 -07:00
John Zhuge
8b242f09a6 HDFS-11303. Hedged read might hang infinitely if read data from all DN failed . Contributed by Chen Zhang, Wei-chiu Chuang, and John Zhuge. 2017-08-11 19:42:07 -07:00
Kai Zheng
a53b8b6fdc HDFS-11975. Provide a system-default EC policy. Contributed by Huichun Lu 2017-08-09 10:12:58 +08:00
Haohui Mai
71b8dda4f6 HDFS-10326. Disable setting tcp socket send/receive buffers for write pipelines. Contributed by Daryn Sharp. 2017-08-08 14:58:16 -07:00
Chris Douglas
12e44e7bda HDFS-6984. Serialize FileStatus via protobuf. 2017-08-02 12:12:48 -07:00
Lei Xu
9586b0e24f HADOOP-14397. Pull up the builder pattern to FileSystem and add AbstractContractCreateTest for it. (Lei (Eddy) Xu) 2017-07-31 20:12:40 -07:00
Lei Xu
77791e4c36 HDFS-12044. Mismatch between BlockManager.maxReplicationStreams and ErasureCodingWorker.stripedReconstructionPool pool size causes slow and bursty recovery. (Contributed by Lei (Eddy) Xu) 2017-07-28 10:50:49 -07:00
Wei-Chiu Chuang
369f731264 HDFS-12062. removeErasureCodingPolicy needs super user permission. Contributed by Wei-Chiu Chuang. 2017-07-28 00:51:03 -07:00
Mingliang Liu
6436768baf HDFS-12089. Fix ambiguous NN retry log message in WebHDFS. Contributed by Eric Badger 2017-07-05 11:10:57 -07:00
Andrew Wang
441378e7e4 Revert "HDFS-11696. Fix warnings from Spotbugs in hadoop-hdfs. Contributed by Yiqun Lin." 
This reverts commit 89a8edc014.

 Conflicts:
	hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/tools/DFSAdmin.java
 2017-06-29 13:54:16 -07:00
Wei-Chiu Chuang
16c8dbde57 HDFS-11881. NameNode consumes a lot of memory for snapshot diff report generation. Contributed by Manoj Govindassamy. 2017-06-29 06:38:41 -07:00
Wei-Chiu Chuang
f99b6d19de HDFS-11870. Add CLI cmd to enable/disable an erasure code policy. Contributed by lufei. 2017-06-28 10:54:25 -07:00
Ravi Prakash
a9d3412b4c HDFS-11993. Add log info when connect to datanode socket address failed. Contributed by chencan 2017-06-26 13:24:27 -07:00
Arpit Agarwal
6d116ffad2 HDFS-11789. Maintain Short-Circuit Read Statistics. Contributed by Hanisha Koneru. 2017-06-22 13:35:56 -07:00
Vinayakumar B
8dbd53ef9f HDFS-11067. DFS#listStatusIterator(..) should throw FileNotFoundException if the directory deleted before fetching next batch of entries. Contributed by Vinayakumar B. 2017-06-22 17:37:08 +05:30
Kai Zheng
5db3f98468 HDFS-11606. Add CLI cmd to remove an erasure code policy. Contributed by Tim Yao 2017-06-21 14:01:55 +08:00
Kai Zheng
a010b330e7 HDFS-11933. Arguments check for ErasureCodingPolicy->composePolicyName. Contributed by Lu Fei 2017-06-21 13:46:18 +08:00
Lei Xu
45ff4d38e6 HDFS-11647. Add -E option in hdfs "count" command to show erasure policy summarization. Contributed by luhuichun. 2017-06-20 11:55:09 -07:00
Lei Xu
6460df21a0 HADOOP-14395. Provide Builder pattern for DistributedFileSystem.append. Contributed by Lei (Eddy) Xu. 2017-06-16 17:24:00 -07:00
Xiao Chen
4ebc23ba7b HADOOP-14524. Make CryptoCodec Closeable so it can be cleaned up proactively. 2017-06-16 09:49:33 -07:00
Andrew Wang
fb68980959 HDFS-10480. Add an admin command to list currently open files. Contributed by Manoj Govindassamy. 2017-06-15 14:46:55 -07:00
Lei Xu
5e7cfdca7b HADOOP-14394. Provide Builder pattern for DistributedFileSystem.create. (lei) 2017-06-15 10:59:24 -07:00
Lei Xu
999c8fcbef HDFS-10999. Introduce separate stats for Replicated and Erasure Coded Blocks apart from the current Aggregated stats. (Manoj Govindassamy via lei) 2017-06-14 10:44:19 -07:00
Vinayakumar B
70fc6746b3 HDFS-11708. Positional read will fail if replicas moved to different DNs after stream is opened. Contributed by Vinayakumar B. 2017-06-07 10:55:53 +05:30
Yiqun Lin
8d9084eb62 HDFS-11359. DFSAdmin report command supports displaying maintenance state datanodes. Contributed by Yiqun Lin. 2017-06-02 12:48:30 +08:00
Andrew Wang
7101477e47 HDFS-11383. Intern strings in BlockLocation and ExtendedBlock. Contributed by Misha Dmitriev. 2017-06-01 15:20:18 -07:00
Kihwal Lee
29b7df960f HDFS-11856. Ability to re-add Upgrading Nodes to pipeline for future pipeline updates. Contributed by Vinayakumar B. 2017-05-25 13:05:23 -05:00
Kai Zheng
a62be38a5e HDFS-11793. Allow to enable user defined erasure coding policy. Contributed by Sammi Chen 2017-05-24 18:45:52 +08:00
Rakesh Radhakrishnan
1b5451bf05 HDFS-11794. Add ec sub command -listCodec to show currently supported ec codecs. Contributed by SammiChen. 2017-05-23 17:03:28 +05:30
Sunil G
b6f66b0da1 YARN-6584. Correct license headers in hadoop-common, hdfs, yarn and mapreduce. Contributed by Yeliang Cang. 2017-05-22 14:10:06 +05:30
Akira Ajisaka
89a8edc014
HDFS-11696. Fix warnings from Spotbugs in hadoop-hdfs. Contributed by Yiqun Lin. 2017-05-16 12:41:59 -04:00
Andrew Wang
54fd0e44b7 HDFS-11644. Support for querying outputstream capabilities. Contributed by Manoj Govindassamy. 2017-05-08 21:59:49 -07:00
Kihwal Lee
cef2815cf4 HDFS-11702. Remove indefinite caching of key provider uri in DFSClient. Contributed by Rushabh S Shah. 2017-05-08 08:27:37 -05:00
Andrew Wang
a9a3d219fe HDFS-9342. Erasure coding: client should update and commit block based on acknowledged size. Contributed by SammiChen. 2017-05-07 14:45:26 -07:00
Chris Douglas
a3954ccab1 HDFS-9807. Add an optional StorageID to writes. Contributed by Ewan Higgs 2017-05-05 12:01:26 -07:00
Lei Xu
25f5d9ad5e HDFS-11687. Add new public encryption APIs required by Hive. (lei) 
Change-Id: I4a23a00de63ad18022312ceb1f306a87d032d07c
 2017-05-04 12:06:50 -07:00
Andrew Wang
c2a52ef9c2 HDFS-11643. Add shouldReplicate option to create builder. Contributed by SammiChen. 2017-05-04 11:39:14 -07:00
Lei Xu
19a7e94ee4 HDFS-11718. DFSStripedOutputStream hsync/hflush should not throw UnsupportedOperationException. (Manoj Govindassamy via lei) 
Change-Id: I4cc226b80c64a0d900a3b1ce71e51f051cd29c22
 2017-04-28 17:06:14 -07:00
Kai Zheng
cb672a45a0 HDFS-11605. Allow user to customize new erasure code policies. Contributed by Huafeng Wang 2017-04-28 13:18:50 +08:00
Akira Ajisaka
249cb210f5
HDFS-11697. Add javadoc for storage policy and erasure coding policy. Contributed by Kai Sasaki. 2017-04-27 16:12:27 +09:00
Akira Ajisaka
93fa48fcf2
HDFS-11417. Add datanode admin command to get the storage info. Contributed by Surendra Singh Lilhore. 2017-04-26 17:43:29 +09:00
Chris Douglas
2f73396b59 HDFS-6708. StorageType should be encoded in the block token. Contributed by Ewan Higgs 2017-04-25 23:57:00 -07:00
Yongjun Zhang
20e3ae260b HDFS-11402. HDFS Snapshots should capture point-in-time copies of OPEN files. (Manoj Govindassamy via Yongjun Zhang) 2017-04-21 20:35:48 -07:00
Yongjun Zhang
5078df7be3 HDFS-11689. New exception thrown by DFSClient%isHDFSEncryptionEnabled broke hacky hive code. Contributed by Yongjun Zhang. 2017-04-21 13:36:31 -07:00
Kai Zheng
b0803388fc HDFS-11604. Define and parse erasure code policies. Contributed by Lin Zeng 2017-04-21 13:33:33 +08:00
John Zhuge
8c81a16a1f HDFS-11531. Expose hedged read metrics via libHDFS API. Contributed by Sailesh Mukil. 2017-04-18 21:11:07 -07:00
Andrew Wang
c0ca785dbb HDFS-11652. Improve ECSchema and ErasureCodingPolicy toString, hashCode, equals. 2017-04-17 16:30:25 -07:00
Andrew Wang
a7312715a6 HDFS-10996. Ability to specify per-file EC policy at create time. Contributed by SammiChen. 2017-04-12 12:27:34 -07:00
Andrew Wang
966b1b5b44 HDFS-11565. Use compact identifiers for built-in ECPolicies in HdfsFileStatus. 2017-04-12 12:24:32 -07:00
Chris Nauroth
23b1a7bdf1 HDFS-11163. Mover should move the file blocks to default storage once policy is unset. Contributed by Surendra Singh Lilhore. 2017-04-11 22:03:09 -07:00
Andrew Wang
e8bdad7385 HDFS-11623. Move system erasure coding policies into hadoop-hdfs-client. 2017-04-07 16:46:28 -07:00
Xiaoyu Yao
0eacd4c13b HDFS-11608. HDFS write crashed with block size greater than 2 GB. Contributed by Xiaobing Zhou. 2017-04-06 16:11:55 -07:00
Andrew Wang
9e0e430f18 HDFS-11538. Move ClientProtocol HA proxies into hadoop-hdfs-client. Contributed by Huafeng Wang. 2017-04-04 23:05:24 -07:00
Andrew Wang
18432130a7 HADOOP-14104. Client should always ask namenode for kms provider path. Contributed by Rushabh S Shah. 2017-04-04 13:38:25 -07:00
Chris Douglas
6eba79232f HADOOP-14271. Correct spelling of 'occurred' and variants. Contributed by Yeliang Cang 2017-04-03 20:13:14 -07:00
Arpit Agarwal
5485d93bda HDFS-11603. Improve slow mirror/disk warnings in BlockReceiver. 2017-03-31 12:10:27 -07:00
Hanisha Koneru
28cdc5a8dc HDFS-11551. Handle SlowDiskReport from DataNode at the NameNode. Contributed by Hanisha Koneru. 2017-03-30 22:41:26 -07:00
Rakesh Radhakrishnan
84d787b9d5 HDFS-11541. Call RawErasureEncoder and RawErasureDecoder release() methods. Contributed by SammiChen. 2017-03-29 11:41:48 +05:30
Tsuyoshi Ozawa
84ddedc0b2 HADOOP-10101. Update guava dependency to the latest version. (ozawa) 2017-03-25 10:06:52 +09:00
Andrew Wang
332a997e10 HDFS-11170. Add builder-based create API to FileSystem. Contributed by SammiChen and Wei Zhou. 2017-03-24 12:56:46 -07:00
Andrew Wang
52b00600df HADOOP-13715. Add isErasureCoded() API to FileStatus class. Contributed by Manoj Govindassamy. 2017-03-24 11:44:46 -07:00
Zhe Zhang
59d69257a8 HADOOP-9631. ViewFs should use underlying FileSystem's server side defaults. Contributed by Lohit Vijayarenu and Erik Krogen. 2017-03-22 22:22:22 -07:00
Arpit Agarwal
e7c8da614c HDFS-11545. Propagate DataNode's slow disks info to the NameNode via Heartbeats. Contributed by Hanisha Koneru. 2017-03-20 21:54:58 -07:00
Chris Douglas
b104f3a282 HDFS-6648. Order of namenodes in ConfiguredFailoverProxyProvider is undefined. Contributed by Inigo Goiri 2017-03-20 17:15:13 -07:00
Andrew Wang
cc1292e73a HDFS-9705. Refine the behaviour of getFileChecksum when length = 0. Contributed by Kai Zheng and SammiChen. 2017-03-14 16:41:10 -07:00
Andrew Wang
5addacb1e3 HDFS-11152. Start erasure coding policy ID number from 1 instead of 0 to void potential unexpected errors. Contributed by SammiChen. 2017-03-08 08:47:38 -08:00
Arpit Agarwal
b5adc5c301 HDFS-10838. Last full block report received time for each DN should be easily discoverable. Contributed by Surendra Singh Lilhore. 2017-03-06 16:39:53 -08:00
Yiqun Lin
3536ce031c HDFS-8741. Proper error msg to be printed when invalid operation type is given to WebHDFS operations. Contributed by Surendra Singh Lilhore. 2017-03-06 19:04:03 +08:00
Rakesh Radhakrishnan
82ef9accaf HDFS-11428. Change setErasureCodingPolicy to take a required string EC policy name. Contributed by Andrew Wang. 2017-03-01 16:06:02 +05:30
Andrew Wang
55c07bbed2 HDFS-11382. Persist Erasure Coding Policy ID in a new optional field in INodeFile in FSImage. Contributed by Manoj Govindassamy. 2017-02-27 17:07:45 -08:00
Xiao Chen
e24ed47d9a HDFS-11421. Make WebHDFS' ACLs RegEx configurable. Contributed by Harsh J. 2017-02-24 16:49:46 -08:00
Jing Zhao
627da6f717 HDFS-8498. Blocks can be committed with wrong size. Contributed by Jing Zhao. 2017-02-15 10:44:37 -08:00