HDFS API的高级编程
HDFS的API就两个：FileSystem 和Configuration
1、文件的上传和下载
1 package com.ghgj.hdfs.api; 2 3 import org.apache.hadoop.conf.Configuration; 4 import org.apache.hadoop.fs.FileSystem; 5 import org.apache.hadoop.fs.Path; 6 7 public class HDFS_GET_AND_PUT { 8 9 public static void main(String[] args) throws Exception {10 11 12 Configuration conf = new Configuration();13 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");14 conf.set("dfs.replication", "2");15 FileSystem fs = FileSystem.get(conf);16 17 18 /**19 * 更改操作用户有两种方式:20 * 21 * 1、直接设置运行换种的用户名为hadoop22 * 23 * VM arguments ; -DHADOOP_USER_NAME=hadoop24 * 25 * 2、在代码中进行声明26 * 27 * System.setProperty("HADOOP_USER_NAME", "hadoop");28 */29 System.setProperty("HADOOP_USER_NAME", "hadoop");30 31 // 上传32 fs.copyFromLocalFile(new Path("c:/sss.txt"), new Path("/a/ggg.txt"));33 34 35 36 /**37 * .crc ： 校验文件38 * 39 * 每个块的元数据信息都只会记录合法数据的起始偏移量： qqq.txt blk_41838 : 0 - 1100byte40 * 41 * 如果进行非法的数据追加。最终是能够下载合法数据。42 * 由于你在数据的中间， 也就是说在 0 -1100 之间的范围进行了数据信息的更改。 造成了采用CRC算法计算出来校验值，和最初存入进HDFS的校验值43 * 不一致。HDFS就认为当前这个文件被损坏了。44 */45 46 47 // 下载 48 fs.copyToLocalFile(new Path("/a/qqq.txt"), new Path("c:/qqq3.txt"));49 50 51 /**52 * 上传和下载的API的底层封装其实就是 ： FileUtil.copy(....)53 */54 55 fs.close();56 }57 } 2、配置文件conf
1 package com.exam.hdfs; 2 3 import java.io.IOException; 4 import java.util.Iterator; 5 import java.util.Map.Entry; 6 7 import org.apache.hadoop.conf.Configuration; 8 import org.apache.hadoop.fs.FileSystem; 9 10 public class TestConf1 {11 12 public static void main(String[] args) throws Exception {13 14 15 /**16 * 底层会加载一堆的配置文件：17 * 18 * core-default.xml19 * hdfs-default.xml20 * mapred-default.xml21 * yarn-default.xml22 */23 Configuration conf = new Configuration();24 // conf.addResource("hdfs-default.xml");25 26 /**27 * 当前这个hdfs-site.xml文件就放置在这个项目中的src下。也就是classpath路径下。28 * 所以 FS在初始化的时候，会把hdfs-site.xml这个文件中的name-value对解析到conf中29 * 30 * 31 * 但是：32 * 33 * 1、如果hdfs-site.xml 不在src下， 看是否能加载？？？ 不能34 * 35 * 2、如果文件名不叫做 hdfs-default.xml 或者 hdsf-site.xml 看是否能自动加载？？？ 不能36 * 37 * 得出的结论：38 * 39 * 如果需要项目代码自动加载配置文件中的信息，那么就必须把配置文件改成-default.xml或者-site.xml的名称40 * 而且必须放置在src下41 * 42 * 那如果不叫这个名，或者不在src下，也需要加载这些配置文件中的参数：43 * 44 * 必须使用conf对象提供的一些方法去手动加载45 */46 // conf.addResource("hdfs-site.xml");47 conf.set("dfs.replication", "1");48 conf.addResource("myconfig/hdfs-site.xml");49 50 51 /**52 * 依次加载的参数信息的顺序是：53 * 54 * 1、加载 core/hdfs/mapred/yarn-default.xml55 * 56 * 2、加载通过conf.addResources()加载的配置文件57 * 58 * 3、加载conf.set(name, value)59 */60 61 FileSystem fs = FileSystem.get(conf);62 63 System.out.println(conf.get("dfs.replication"));64 65 66 Iterator > iterator = conf.iterator();67 while(iterator.hasNext()){68 Entry e = iterator.next();69 System.out.println(e.getKey() + "\t" + e.getValue());70 }71 }72 } 输出结果
1 log4j:WARN No appenders could be found for logger (org.apache.hadoop.metrics2.lib.MutableMetricsFactory). 2 log4j:WARN Please initialize the log4j system properly. 3 log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info. 4 1 5 hadoop.security.groups.cache.secs 300 6 dfs.datanode.cache.revocation.timeout.ms 900000 7 dfs.namenode.resource.check.interval 5000 8 s3.client-write-packet-size 65536 9 dfs.client.https.need-auth false 10 dfs.replication 1 11 hadoop.security.group.mapping.ldap.directory.search.timeout 10000 12 dfs.datanode.available-space-volume-choosing-policy.balanced-space-threshold 10737418240 13 hadoop.work.around.non.threadsafe.getpwuid false 14 dfs.namenode.write-lock-reporting-threshold-ms 5000 15 fs.ftp.host.port 21 16 dfs.namenode.avoid.read.stale.datanode false 17 dfs.journalnode.rpc-address 0.0.0.0:8485 18 hadoop.security.kms.client.encrypted.key.cache.expiry 43200000 19 ipc.client.connection.maxidletime 10000 20 hadoop.registry.zk.session.timeout.ms 60000 21 tfile.io.chunk.size 1048576 22 fs.automatic.close true 23 ha.health-monitor.sleep-after-disconnect.ms 1000 24 io.map.index.interval 128 25 dfs.namenode.https-address 0.0.0.0:50470 26 dfs.mover.max-no-move-interval 60000 27 io.seqfile.sorter.recordlimit 1000000 28 fs.s3n.multipart.uploads.enabled false 29 hadoop.util.hash.type murmur 30 dfs.namenode.replication.min 1 31 dfs.datanode.directoryscan.threads 1 32 dfs.namenode.fs-limits.min-block-size 1048576 33 dfs.datanode.directoryscan.interval 21600 34 fs.AbstractFileSystem.file.impl org.apache.hadoop.fs.local.LocalFs 35 dfs.namenode.acls.enabled false 36 dfs.client.short.circuit.replica.stale.threshold.ms 1800000 37 net.topology.script.number.args 100 38 hadoop.http.authentication.token.validity 36000 39 fs.s3.block.size 67108864 40 dfs.namenode.resource.du.reserved 104857600 41 ha.failover-controller.graceful-fence.rpc-timeout.ms 5000 42 s3native.bytes-per-checksum 512 43 dfs.namenode.datanode.registration.ip-hostname-check true 44 dfs.namenode.path.based.cache.block.map.allocation.percent 0.25 45 dfs.namenode.backup.http-address 0.0.0.0:50105 46 hadoop.security.group.mapping org.apache.hadoop.security.JniBasedUnixGroupsMappingWithFallback 47 dfs.namenode.edits.noeditlogchannelflush false 48 dfs.datanode.cache.revocation.polling.ms 500 49 dfs.namenode.audit.loggers default 50 hadoop.security.groups.cache.warn.after.ms 5000 51 io.serializations org.apache.hadoop.io.serializer.WritableSerialization,org.apache.hadoop.io.serializer.avro.AvroSpecificSerialization,org.apache.hadoop.io.serializer.avro.AvroReflectSerialization 52 dfs.namenode.lazypersist.file.scrub.interval.sec 300 53 fs.s3a.threads.core 15 54 hadoop.security.crypto.buffer.size 8192 55 hadoop.http.cross-origin.allowed-methods GET,POST,HEAD 56 hadoop.registry.zk.retry.interval.ms 1000 57 dfs.http.policy HTTP_ONLY 58 hadoop.registry.secure false 59 dfs.namenode.replication.interval 3 60 dfs.namenode.safemode.min.datanodes 0 61 dfs.client.file-block-storage-locations.num-threads 10 62 nfs.dump.dir /tmp/.hdfs-nfs 63 dfs.namenode.secondary.https-address 0.0.0.0:50091 64 hadoop.kerberos.kinit.command kinit 65 dfs.block.access.token.lifetime 600 66 dfs.webhdfs.enabled true 67 dfs.client.use.datanode.hostname false 68 dfs.namenode.delegation.token.max-lifetime 604800000 69 fs.trash.interval 0 70 dfs.datanode.drop.cache.behind.writes false 71 dfs.namenode.avoid.write.stale.datanode false 72 dfs.namenode.num.extra.edits.retained 1000000 73 s3.blocksize 67108864 74 ipc.client.connect.max.retries.on.timeouts 45 75 dfs.datanode.data.dir /home/hadoop/data/hadoopdata/data 76 fs.s3.buffer.dir ${hadoop.tmp.dir}/s3 77 fs.s3n.block.size 67108864 78 nfs.exports.allowed.hosts * rw 79 ha.health-monitor.connect-retry-interval.ms 1000 80 hadoop.security.instrumentation.requires.admin false 81 hadoop.registry.zk.retry.ceiling.ms 60000 82 nfs.rtmax 1048576 83 dfs.client.mmap.cache.size 256 84 dfs.datanode.data.dir.perm 700 85 io.file.buffer.size 4096 86 dfs.namenode.backup.address 0.0.0.0:50100 87 dfs.client.datanode-restart.timeout 30 88 dfs.datanode.readahead.bytes 4194304 89 dfs.namenode.xattrs.enabled true 90 io.mapfile.bloom.size 1048576 91 ipc.client.connect.retry.interval 1000 92 dfs.client-write-packet-size 65536 93 dfs.namenode.checkpoint.txns 1000000 94 dfs.datanode.bp-ready.timeout 20 95 dfs.datanode.transfer.socket.send.buffer.size 131072 96 hadoop.security.kms.client.authentication.retry-count 1 97 dfs.client.block.write.retries 3 98 fs.swift.impl org.apache.hadoop.fs.swift.snative.SwiftNativeFileSystem 99 ha.failover-controller.graceful-fence.connection.retries 1100 hadoop.registry.zk.connection.timeout.ms 15000101 dfs.namenode.safemode.threshold-pct 0.999f102 dfs.cachereport.intervalMsec 10000103 hadoop.security.java.secure.random.algorithm SHA1PRNG104 ftp.blocksize 67108864105 dfs.namenode.list.cache.directives.num.responses 100106 dfs.namenode.kerberos.principal.pattern *107 file.stream-buffer-size 4096108 dfs.datanode.dns.nameserver default109 fs.s3a.max.total.tasks 1000110 dfs.namenode.replication.considerLoad true111 nfs.allow.insecure.ports true112 dfs.namenode.edits.journal-plugin.qjournal org.apache.hadoop.hdfs.qjournal.client.QuorumJournalManager113 dfs.client.write.exclude.nodes.cache.expiry.interval.millis 600000114 dfs.client.mmap.cache.timeout.ms 3600000115 ipc.client.idlethreshold 4000116 io.skip.checksum.errors false117 ftp.stream-buffer-size 4096118 fs.s3a.fast.upload false119 dfs.client.failover.connection.retries.on.timeouts 0120 file.blocksize 67108864121 ftp.replication 3122 dfs.namenode.replication.work.multiplier.per.iteration 2123 hadoop.security.authorization false124 hadoop.http.authentication.simple.anonymous.allowed true125 s3native.client-write-packet-size 65536126 hadoop.rpc.socket.factory.class.default org.apache.hadoop.net.StandardSocketFactory127 file.bytes-per-checksum 512128 dfs.datanode.slow.io.warning.threshold.ms 300129 fs.har.impl.disable.cache true130 rpc.engine.org.apache.hadoop.hdfs.protocolPB.ClientNamenodeProtocolPB org.apache.hadoop.ipc.ProtobufRpcEngine131 io.seqfile.lazydecompress true132 dfs.namenode.reject-unresolved-dn-topology-mapping false133 hadoop.common.configuration.version 0.23.0134 hadoop.security.authentication simple135 dfs.datanode.drop.cache.behind.reads false136 dfs.image.compression.codec org.apache.hadoop.io.compress.DefaultCodec137 dfs.client.read.shortcircuit.streams.cache.size 256138 file.replication 1139 dfs.namenode.top.num.users 10140 dfs.namenode.accesstime.precision 3600000141 dfs.namenode.fs-limits.max-xattrs-per-inode 32142 dfs.image.transfer.timeout 60000143 io.mapfile.bloom.error.rate 0.005144 nfs.wtmax 1048576145 hadoop.security.kms.client.encrypted.key.cache.size 500146 dfs.namenode.edit.log.autoroll.check.interval.ms 300000147 fs.s3a.multipart.purge false148 dfs.namenode.support.allow.format true149 hadoop.hdfs.configuration.version 1150 fs.s3a.connection.establish.timeout 5000151 hadoop.security.group.mapping.ldap.search.attr.member member152 dfs.secondary.namenode.kerberos.internal.spnego.principal ${dfs.web.authentication.kerberos.principal}153 dfs.stream-buffer-size 4096154 hadoop.ssl.client.conf ssl-client.xml155 dfs.namenode.invalidate.work.pct.per.iteration 0.32f156 fs.s3a.multipart.purge.age 86400157 dfs.journalnode.https-address 0.0.0.0:8481158 dfs.namenode.top.enabled true159 hadoop.security.kms.client.encrypted.key.cache.low-watermark 0.3f160 dfs.namenode.max.objects 0161 hadoop.user.group.static.mapping.overrides dr.who=;162 fs.s3a.fast.buffer.size 1048576163 dfs.bytes-per-checksum 512164 dfs.datanode.max.transfer.threads 4096165 dfs.block.access.key.update.interval 600166 ipc.maximum.data.length 67108864167 tfile.fs.input.buffer.size 262144168 ha.failover-controller.new-active.rpc-timeout.ms 60000169 dfs.client.cached.conn.retry 3170 dfs.client.read.shortcircuit false171 hadoop.ssl.hostname.verifier DEFAULT172 dfs.datanode.hdfs-blocks-metadata.enabled false173 dfs.datanode.directoryscan.throttle.limit.ms.per.sec 0174 dfs.image.transfer.chunksize 65536175 hadoop.http.authentication.type simple176 dfs.namenode.list.encryption.zones.num.responses 100177 dfs.client.https.keystore.resource ssl-client.xml178 s3native.blocksize 67108864179 net.topology.impl org.apache.hadoop.net.NetworkTopology180 dfs.client.failover.sleep.base.millis 500181 io.seqfile.compress.blocksize 1000000182 dfs.namenode.path.based.cache.refresh.interval.ms 30000183 dfs.namenode.decommission.interval 30184 dfs.permissions.superusergroup supergroup185 dfs.namenode.fs-limits.max-directory-items 1048576186 hadoop.registry.zk.retry.times 5187 dfs.ha.log-roll.period 120188 fs.AbstractFileSystem.ftp.impl org.apache.hadoop.fs.ftp.FtpFs189 ftp.bytes-per-checksum 512190 dfs.user.home.dir.prefix /user191 dfs.namenode.checkpoint.edits.dir ${dfs.namenode.checkpoint.dir}192 dfs.client.socket.send.buffer.size 131072193 ipc.client.fallback-to-simple-auth-allowed false194 dfs.blockreport.initialDelay 0195 dfs.namenode.inotify.max.events.per.rpc 1000196 dfs.namenode.heartbeat.recheck-interval 300000197 dfs.namenode.safemode.extension 30000198 dfs.client.failover.sleep.max.millis 15000199 dfs.namenode.delegation.key.update-interval 86400000200 dfs.datanode.transfer.socket.recv.buffer.size 131072201 hadoop.rpc.protection authentication202 fs.permissions.umask-mode 022203 fs.s3.sleepTimeSeconds 10204 dfs.namenode.fs-limits.max-xattr-size 16384205 ha.health-monitor.rpc-timeout.ms 45000206 hadoop.http.staticuser.user dr.who207 dfs.datanode.http.address 0.0.0.0:50075208 fs.s3a.connection.maximum 15209 fs.s3a.paging.maximum 5000210 fs.AbstractFileSystem.viewfs.impl org.apache.hadoop.fs.viewfs.ViewFs211 dfs.namenode.blocks.per.postponedblocks.rescan 10000212 fs.ftp.host 0.0.0.0213 dfs.lock.suppress.warning.interval 10s214 hadoop.http.authentication.kerberos.keytab ${user.home}/hadoop.keytab215 fs.s3a.impl org.apache.hadoop.fs.s3a.S3AFileSystem216 hadoop.registry.zk.root /registry217 hadoop.jetty.logs.serve.aliases true218 dfs.namenode.fs-limits.max-blocks-per-file 1048576219 dfs.balancer.keytab.enabled false220 dfs.client.block.write.replace-datanode-on-failure.enable true221 hadoop.http.cross-origin.max-age 1800222 io.compression.codec.bzip2.library system-native223 dfs.namenode.checkpoint.dir file://${hadoop.tmp.dir}/dfs/namesecondary224 dfs.client.use.legacy.blockreader.local false225 dfs.namenode.top.windows.minutes 1,5,25226 ipc.ping.interval 60000227 net.topology.node.switch.mapping.impl org.apache.hadoop.net.ScriptBasedMapping228 nfs.mountd.port 4242229 dfs.storage.policy.enabled true230 dfs.namenode.list.cache.pools.num.responses 100231 fs.df.interval 60000232 nfs.server.port 2049233 ha.zookeeper.parent-znode /hadoop-ha234 hadoop.http.cross-origin.allowed-headers X-Requested-With,Content-Type,Accept,Origin235 dfs.datanode.block-pinning.enabled false236 dfs.namenode.num.checkpoints.retained 2237 fs.s3a.attempts.maximum 10238 s3native.stream-buffer-size 4096239 io.seqfile.local.dir ${hadoop.tmp.dir}/io/local240 fs.s3n.multipart.copy.block.size 5368709120241 dfs.encrypt.data.transfer.cipher.key.bitlength 128242 dfs.client.mmap.retry.timeout.ms 300000243 dfs.datanode.sync.behind.writes false244 dfs.namenode.fslock.fair true245 hadoop.ssl.keystores.factory.class org.apache.hadoop.security.ssl.FileBasedKeyStoresFactory246 dfs.permissions.enabled true247 fs.AbstractFileSystem.hdfs.impl org.apache.hadoop.fs.Hdfs248 dfs.blockreport.split.threshold 1000000249 dfs.datanode.balance.bandwidthPerSec 1048576250 dfs.block.scanner.volume.bytes.per.second 1048576251 hadoop.security.random.device.file.path /dev/urandom252 fs.s3.maxRetries 4253 hadoop.http.filter.initializers org.apache.hadoop.http.lib.StaticUserWebFilter254 dfs.namenode.stale.datanode.interval 30000255 ipc.client.rpc-timeout.ms 0256 fs.client.resolve.remote.symlinks true257 dfs.default.chunk.view.size 32768258 hadoop.ssl.enabled.protocols TLSv1259 dfs.namenode.decommission.blocks.per.interval 500000260 dfs.namenode.handler.count 10261 dfs.image.transfer.bandwidthPerSec 0262 rpc.metrics.quantile.enable false263 hadoop.ssl.enabled false264 dfs.replication.max 512265 dfs.namenode.name.dir /home/hadoop/data/hadoopdata/name266 dfs.namenode.read-lock-reporting-threshold-ms 5000267 dfs.datanode.https.address 0.0.0.0:50475268 dfs.datanode.failed.volumes.tolerated 0269 ipc.client.kill.max 10270 fs.s3a.threads.max 256271 ipc.server.listen.queue.size 128272 dfs.client.domain.socket.data.traffic false273 dfs.block.access.token.enable false274 dfs.blocksize 134217728275 fs.s3a.connection.timeout 50000276 fs.s3a.threads.keepalivetime 60277 file.client-write-packet-size 65536278 dfs.datanode.address 0.0.0.0:50010279 ha.failover-controller.cli-check.rpc-timeout.ms 20000280 ha.zookeeper.acl world:anyone:rwcda281 ipc.client.connect.max.retries 10282 dfs.encrypt.data.transfer false283 dfs.namenode.write.stale.datanode.ratio 0.5f284 ipc.client.ping true285 dfs.datanode.shared.file.descriptor.paths /dev/shm,/tmp286 dfs.short.circuit.shared.memory.watcher.interrupt.check.ms 60000287 hadoop.tmp.dir /home/hadoop/data/hadoopdata288 dfs.datanode.handler.count 10289 dfs.client.failover.max.attempts 15290 dfs.balancer.max-no-move-interval 60000291 dfs.client.read.shortcircuit.streams.cache.expiry.ms 300000292 dfs.namenode.block-placement-policy.default.prefer-local-node true293 hadoop.ssl.require.client.cert false294 hadoop.security.uid.cache.secs 14400295 dfs.client.read.shortcircuit.skip.checksum false296 dfs.namenode.resource.checked.volumes.minimum 1297 hadoop.registry.rm.enabled false298 dfs.namenode.quota.init-threads 4299 dfs.namenode.max.extra.edits.segments.retained 10000300 dfs.webhdfs.user.provider.user.pattern ^[A-Za-z_][A-Za-z0-9._-]*[$]?$301 dfs.client.mmap.enabled true302 dfs.client.file-block-storage-locations.timeout.millis 1000303 dfs.datanode.block.id.layout.upgrade.threads 12304 dfs.datanode.use.datanode.hostname false305 hadoop.fuse.timer.period 5306 dfs.client.context default307 fs.trash.checkpoint.interval 0308 dfs.journalnode.http-address 0.0.0.0:8480309 dfs.balancer.address 0.0.0.0:0310 dfs.namenode.lock.detailed-metrics.enabled false311 dfs.namenode.delegation.token.renew-interval 86400000312 ha.health-monitor.check-interval.ms 1000313 dfs.namenode.retrycache.heap.percent 0.03f314 ipc.client.connect.timeout 20000315 dfs.reformat.disabled false316 dfs.blockreport.intervalMsec 21600000317 fs.s3a.multipart.threshold 2147483647318 dfs.https.server.keystore.resource ssl-server.xml319 hadoop.http.cross-origin.enabled false320 io.map.index.skip 0321 dfs.balancer.block-move.timeout 0322 io.native.lib.available true323 s3.replication 3324 dfs.namenode.kerberos.internal.spnego.principal ${dfs.web.authentication.kerberos.principal}325 fs.AbstractFileSystem.har.impl org.apache.hadoop.fs.HarFs326 hadoop.security.kms.client.encrypted.key.cache.num.refill.threads 2327 fs.s3n.multipart.uploads.block.size 67108864328 dfs.image.compress false329 dfs.datanode.dns.interface default330 dfs.datanode.available-space-volume-choosing-policy.balanced-space-preference-fraction 0.75f331 tfile.fs.output.buffer.size 262144332 fs.du.interval 600000333 dfs.client.failover.connection.retries 0334 dfs.namenode.edit.log.autoroll.multiplier.threshold 2.0335 hadoop.security.group.mapping.ldap.ssl false336 dfs.namenode.top.window.num.buckets 10337 fs.s3a.buffer.dir ${hadoop.tmp.dir}/s3a338 dfs.namenode.checkpoint.check.period 60339 fs.defaultFS hdfs://hadoop1:9000340 fs.s3a.multipart.size 104857600341 dfs.client.slow.io.warning.threshold.ms 30000342 dfs.datanode.max.locked.memory 0343 dfs.namenode.retrycache.expirytime.millis 600000344 hadoop.security.group.mapping.ldap.search.attr.group.name cn345 dfs.client.block.write.replace-datanode-on-failure.best-effort false346 dfs.ha.fencing.ssh.connect-timeout 30000347 dfs.datanode.scan.period.hours 504348 hadoop.registry.zk.quorum localhost:2181349 dfs.namenode.fs-limits.max-component-length 255350 hadoop.http.cross-origin.allowed-origins *351 dfs.namenode.enable.retrycache true352 dfs.datanode.du.reserved 0353 dfs.datanode.ipc.address 0.0.0.0:50020354 hadoop.registry.system.acls sasl:yarn@, sasl:mapred@, sasl:hdfs@355 dfs.namenode.path.based.cache.retry.interval.ms 30000356 hadoop.security.crypto.cipher.suite AES/CTR/NoPadding357 dfs.client.block.write.replace-datanode-on-failure.policy DEFAULT358 dfs.namenode.http-address 0.0.0.0:50070359 hadoop.security.crypto.codec.classes.aes.ctr.nopadding org.apache.hadoop.crypto.OpensslAesCtrCryptoCodec,org.apache.hadoop.crypto.JceAesCtrCryptoCodec360 dfs.ha.tail-edits.period 60361 hadoop.security.groups.negative-cache.secs 30362 hadoop.ssl.server.conf ssl-server.xml363 hadoop.registry.jaas.context Client364 s3native.replication 3365 hadoop.security.group.mapping.ldap.search.filter.group (objectClass=group)366 hadoop.http.authentication.kerberos.principal HTTP/_HOST@LOCALHOST367 dfs.namenode.startup.delay.block.deletion.sec 0368 hadoop.security.group.mapping.ldap.search.filter.user (&(objectClass=user)(sAMAccountName={0}))369 dfs.namenode.edits.dir ${dfs.namenode.name.dir}370 dfs.namenode.checkpoint.max-retries 3371 s3.stream-buffer-size 4096372 ftp.client-write-packet-size 65536373 dfs.datanode.fsdatasetcache.max.threads.per.volume 4374 hadoop.security.sensitive-config-keys password$,fs.s3.*[Ss]ecret.?[Kk]ey,fs.azure.account.key.*,dfs.webhdfs.oauth2.[a-z]+.token,hadoop.security.sensitive-config-keys375 dfs.namenode.decommission.max.concurrent.tracked.nodes 100376 dfs.namenode.name.dir.restore false377 ipc.server.log.slow.rpc false378 dfs.heartbeat.interval 3379 dfs.namenode.secondary.http-address hadoop3:50090380 ha.zookeeper.session-timeout.ms 5000381 s3.bytes-per-checksum 512382 fs.s3a.connection.ssl.enabled true383 hadoop.http.authentication.signature.secret.file ${user.home}/hadoop-http-auth-signature-secret384 hadoop.fuse.connection.timeout 300385 dfs.namenode.checkpoint.period 3600386 ipc.server.max.connections 0387 dfs.ha.automatic-failover.enabled false 3、列出指定目录下的文件以及块的信息
1 package com.exam.hdfs; 2 3 import org.apache.hadoop.conf.Configuration; 4 import org.apache.hadoop.fs.BlockLocation; 5 import org.apache.hadoop.fs.FileSystem; 6 import org.apache.hadoop.fs.LocatedFileStatus; 7 import org.apache.hadoop.fs.Path; 8 import org.apache.hadoop.fs.RemoteIterator; 9 10 public class TestHDFS1 {11 12 public static void main(String[] args) throws Exception {13 14 Configuration conf = new Configuration();15 System.setProperty("HADOOP_USER_NAME", "hadoop");16 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");17 FileSystem fs = FileSystem.get(conf);18 19 /**20 * 列出指定的目录下的所有文件21 */22 RemoteIterator listFiles = fs.listFiles(new Path("/"), true);23 while(listFiles.hasNext()){24 LocatedFileStatus file = listFiles.next();25 26 27 System.out.println(file.getPath()+"\t");28 System.out.println(file.getPath().getName()+"\t");29 System.out.println(file.getLen()+"\t");30 System.out.println(file.getReplication()+"\t");31 32 /**33 * blockLocations的长度是几？ 是什么意义？34 * 35 * 块的数量36 */37 BlockLocation[] blockLocations = file.getBlockLocations();38 System.out.println(blockLocations.length+"\t");39 40 for(BlockLocation bl : blockLocations){41 String[] hosts = bl.getHosts();42 43 System.out.print(hosts[0] + "-" + hosts[1]+"\t");44 }45 System.out.println();46 47 }48 49 50 }51 } 输出结果
1 hdfs://hadoop1:9000/aa/bb/cc/hadoop.tar.gz 2 hadoop.tar.gz 3 199007110 4 2 5 3 6 hadoop3-hadoop1 hadoop1-hadoop2 hadoop1-hadoop4
4、上传文件
1 package com.exam.hdfs; 2 3 import java.io.File; 4 import java.io.FileInputStream; 5 import java.io.InputStream; 6 7 import org.apache.hadoop.conf.Configuration; 8 import org.apache.hadoop.fs.FSDataOutputStream; 9 import org.apache.hadoop.fs.FileSystem;10 import org.apache.hadoop.fs.Path;11 import org.apache.hadoop.io.IOUtils;12 13 public class UploadDataByStream {14 15 public static void main(String[] args) throws Exception {16 17 18 Configuration conf = new Configuration();19 System.setProperty("HADOOP_USER_NAME", "hadoop");20 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");21 FileSystem fs = FileSystem.get(conf);22 23 24 InputStream in = new FileInputStream(new File("d:/abc.tar.gz"));25 FSDataOutputStream out = fs.create(new Path("/aa/abc.tar.gz"));26 27 28 IOUtils.copyBytes(in, out, 4096, true);29 30 fs.close();31 32 }33 } 5、下载文件
1 package com.exam.hdfs; 2 3 import java.io.File; 4 import java.io.FileOutputStream; 5 import java.io.OutputStream; 6 7 import org.apache.hadoop.conf.Configuration; 8 import org.apache.hadoop.fs.FSDataInputStream; 9 import org.apache.hadoop.fs.FileSystem;10 import org.apache.hadoop.fs.Path;11 import org.apache.hadoop.io.IOUtils;12 13 public class DownloadDataByStream {14 15 16 public static void main(String[] args) throws Exception {17 18 Configuration conf = new Configuration();19 System.setProperty("HADOOP_USER_NAME", "hadoop");20 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");21 FileSystem fs = FileSystem.get(conf);22 23 24 FSDataInputStream in = fs.open(new Path("/aa/abc.tar.gz"));25 OutputStream out = new FileOutputStream(new File("D:/abc.sh"));26 27 28 IOUtils.copyBytes(in, out, 4096, true);29 30 fs.close();31 32 }33 } 6、删除某个路径下特定类型的文件，比如class类型文件，比如txt类型文件
1 package com.exam.hdfs; 2 3 import java.net.URI; 4 5 import org.apache.hadoop.conf.Configuration; 6 import org.apache.hadoop.fs.FileStatus; 7 import org.apache.hadoop.fs.FileSystem; 8 import org.apache.hadoop.fs.Path; 9 10 public class HDFS_DELETE_CLASS {11 12 public static final String FILETYPE = "tar.gz";13 public static final String DELETE_PATH = "/aa";14 15 public static void main(String[] args) throws Exception {16 17 new HDFS_DELETE_CLASS().rmrClassFile(new Path(DELETE_PATH));18 }19 20 public void rmrClassFile(Path path) throws Exception{21 22 // 首先获取集群必要的信息，以得到FileSystem的示例对象fs23 Configuration conf = new Configuration();24 FileSystem fs = FileSystem.get(new URI("hdfs://hadoop1:9000"), conf, "hadoop");25 26 // 首先检查path本身是文件夹还是目录27 FileStatus fileStatus = fs.getFileStatus(path);28 boolean directory = fileStatus.isDirectory();29 30 // 根据该目录是否是文件或者文件夹进行相应的操作31 if(directory){32 // 如果是目录33 checkAndDeleteDirectory(path, fs);34 }else{35 // 如果是文件，检查该文件名是不是FILETYPE类型的文件36 checkAndDeleteFile(path, fs);37 }38 }39 40 // 处理目录41 public static void checkAndDeleteDirectory(Path path, FileSystem fs) throws Exception{42 // 查看该path目录下一级子目录和子文件的状态43 FileStatus[] listStatus = fs.listStatus(path);44 for(FileStatus fStatus: listStatus){45 Path p = fStatus.getPath();46 // 如果是文件，并且是以FILETYPE结尾，则删掉，否则继续遍历下一级目录47 if(fStatus.isFile()){48 checkAndDeleteFile(p, fs);49 }else{50 checkAndDeleteDirectory(p, fs);51 }52 }53 }54 55 // 檢查文件是否符合刪除要求，如果符合要求則刪除，不符合要求则不做处理56 public static void checkAndDeleteFile(Path path, FileSystem fs) throws Exception{57 String name = path.getName();58 System.out.println(name);59 /*// 直接判断有没有FILETYPE这个字符串,不是特别稳妥，并且会有误操作，所以得判断是不是以FILETYPE结尾60 if(name.indexOf(FILETYPE) != -1){61 fs.delete(path, true);62 }*/63 // 判断是不是以FILETYPE结尾64 int startIndex = name.length() - FILETYPE.length();65 int endIndex = name.length();66 // 求得文件后缀名67 String fileSuffix = name.substring(startIndex, endIndex);68 if(fileSuffix.equals(FILETYPE)){69 fs.delete(path, true);70 }71 }72 } 7、删除HDFS集群中的所有空文件和空目录
1 public class DeleteEmptyDirAndFile { 2 3 static FileSystem fs = null; 4 5 public static void main(String[] args) throws Exception { 6 7 initFileSystem(); 8 9 // 创建测试数据 10 // makeTestData(); 11 12 // 删除测试数据 13 // deleteTestData(); 14 15 // 删除指定文件夹下的空文件和空文件夹 16 deleteEmptyDirAndFile(new Path("/aa")); 17 } 18 19 /** 20 * 删除指定文件夹下的 空文件 和 空文件夹 21 * @throws Exception 22 */ 23 public static void deleteEmptyDirAndFile(Path path) throws Exception { 24 25 //当是空文件夹时 26 FileStatus[] listStatus = fs.listStatus(path); 27 if(listStatus.length == 0){ 28 fs.delete(path, true); 29 return; 30 } 31 32 // 该方法的结果：包括指定目录的 文件 和 文件夹 33 RemoteIterator listLocatedStatus = fs.listLocatedStatus(path); 34 35 while (listLocatedStatus.hasNext()) { 36 LocatedFileStatus next = listLocatedStatus.next(); 37 38 Path currentPath = next.getPath(); 39 // 获取父目录 40 Path parent = next.getPath().getParent(); 41 42 // 如果是文件夹，继续往下遍历，删除符合条件的文件（空文件夹） 43 if (next.isDirectory()) { 44 45 // 如果是空文件夹 46 if(fs.listStatus(currentPath).length == 0){ 47 // 删除掉 48 fs.delete(currentPath, true); 49 }else{ 50 // 不是空文件夹，那么则继续遍历 51 if(fs.exists(currentPath)){ 52 deleteEmptyDirAndFile(currentPath); 53 } 54 } 55 56 // 如果是文件 57 } else { 58 // 获取文件的长度 59 long fileLength = next.getLen(); 60 // 当文件是空文件时， 删除 61 if(fileLength == 0){ 62 fs.delete(currentPath, true); 63 } 64 } 65 66 // 当空文件夹或者空文件删除时，有可能导致父文件夹为空文件夹， 67 // 所以每次删除一个空文件或者空文件的时候都需要判断一下，如果真是如此，那么就需要把该文件夹也删除掉 68 int length = fs.listStatus(parent).length; 69 if(length == 0){ 70 fs.delete(parent, true); 71 } 72 } 73 } 74 75 /** 76 * 初始化FileSystem对象之用 77 */ 78 public static void initFileSystem() throws Exception{ 79 Configuration conf = new Configuration(); 80 System.setProperty("HADOOP_USER_NAME", "hadoop"); 81 conf.addResource("config/core-site.xml"); 82 conf.addResource("config/hdfs-site.xml"); 83 fs = FileSystem.get(conf); 84 } 85 86 /** 87 * 创建 测试 数据之用 88 */ 89 public static void makeTestData() throws Exception { 90 91 String emptyFilePath = "D:\\bigdata\\1704mr_test\\empty.txt"; 92 String notEmptyFilePath = "D:\\bigdata\\1704mr_test\\notEmpty.txt"; 93 94 // 空文件夹 和 空文件 的目录 95 String path1 = "/aa/bb1/cc1/dd1/"; 96 fs.mkdirs(new Path(path1)); 97 fs.mkdirs(new Path("/aa/bb1/cc1/dd2/")); 98 fs.copyFromLocalFile(new Path(emptyFilePath), new Path(path1)); 99 fs.copyFromLocalFile(new Path(notEmptyFilePath), new Path(path1));100 101 // 空文件 的目录102 String path2 = "/aa/bb1/cc2/dd2/";103 fs.mkdirs(new Path(path2));104 fs.copyFromLocalFile(new Path(emptyFilePath), new Path(path2));105 106 // 非空文件 的目录107 String path3 = "/aa/bb2/cc3/dd3";108 fs.mkdirs(new Path(path3));109 fs.copyFromLocalFile(new Path(notEmptyFilePath), new Path(path3));110 111 // 空 文件夹112 String path4 = "/aa/bb2/cc4/dd4";113 fs.mkdirs(new Path(path4));114 115 System.out.println("测试数据创建成功");116 }117 118 /**119 * 删除 指定文件夹120 * @throws Exception 121 */122 public static void deleteTestData() throws Exception {123 boolean delete = fs.delete(new Path("/aa"), true);124 System.out.println(delete ? "删除数据成功" : "删除数据失败");125 }126 127 } 8、手动拷贝某个特定的数据块（比如某个文件的第二个数据块）
1 /** 2 * 手动拷贝某个特定的数据块（比如某个文件的第二个数据块） 3 * */ 4 public static void copyBlock(String str,int num) { 5 6 Path path = new Path(str); 7 8 BlockLocation[] localtions = new BlockLocation[0] ; 9 10 try {11 FileStatus fileStatus = fs.getFileStatus(path);12 13 localtions = fs.getFileBlockLocations(fileStatus, 0, fileStatus.getLen());14 15 /*for(int i=0;i 9、编写程序统计出HDFS文件系统中文件大小小于HDFS集群中的默认块大小的文件占比
1 import org.apache.hadoop.conf.Configuration; 2 import org.apache.hadoop.fs.FileSystem; 3 import org.apache.hadoop.fs.LocatedFileStatus; 4 import org.apache.hadoop.fs.Path; 5 import org.apache.hadoop.fs.RemoteIterator; 6 7 /** 8 * 9 * 编写程序统计出HDFS文件系统中文件大小小于HDFS集群中的默认块大小的文件占比10 * 比如：大于等于128M的文件个数为98，小于128M的文件总数为2，所以答案是2%11 */12 public class Exam1_SmallFilePercent {13 14 private static int DEFAULT_BLOCKSIZE = 128 * 1024 * 1024;15 16 public static void main(String[] args) throws Exception {17 18 19 Configuration conf = new Configuration();20 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");21 System.setProperty("HADOOP_USER_NAME", "hadoop");22 FileSystem fs = FileSystem.get(conf);23 24 25 Path path = new Path("/");26 float smallFilePercent = getSmallFilePercent(fs, path);27 System.out.println(smallFilePercent);28 29 30 fs.close();31 }32 33 /**34 * 该方法求出指定目录下的小文件和总文件数的对比35 * @throws Exception 36 */37 private static float getSmallFilePercent(FileSystem fs, Path path) throws Exception {38 // TODO Auto-generated method stub39 40 int smallFile = 0;41 int totalFile = 0;42 43 RemoteIterator listFiles = fs.listFiles(path, false);44 while(listFiles.hasNext()){45 totalFile++;46 LocatedFileStatus next = listFiles.next();47 long len = next.getLen();48 if(len < DEFAULT_BLOCKSIZE){49 smallFile++;50 }51 }52 System.out.println(smallFile+" : "+totalFile);53 54 return smallFile * 1f /totalFile;55 }56 57 } 10、编写程序统计出HDFS文件系统中的平均数据块数（数据块总数/文件总数）
1 import org.apache.hadoop.conf.Configuration; 2 import org.apache.hadoop.fs.FileSystem; 3 import org.apache.hadoop.fs.LocatedFileStatus; 4 import org.apache.hadoop.fs.Path; 5 import org.apache.hadoop.fs.RemoteIterator; 6 7 /** 8 * 9 * 编写程序统计出HDFS文件系统中的平均数据块数（数据块总数/文件总数）10 * 比如：一个文件有5个块，一个文件有3个块，那么平均数据块数为411 * 如果还有一个文件，并且数据块就1个，那么整个HDFS的平均数据块数就是312 */13 public class Exam2_HDSFAvgBlocks {14 15 public static void main(String[] args) throws Exception {16 17 18 Configuration conf = new Configuration();19 conf.set("fs.defaultFS", "hdfs://hadoop1:9000");20 System.setProperty("HADOOP_USER_NAME", "hadoop");21 FileSystem fs = FileSystem.get(conf);22 23 24 Path path = new Path("/");25 float avgHDFSBlocks = getHDFSAvgBlocks(fs, path);26 System.out.println("HDFS的平均数据块个数为：" + avgHDFSBlocks);27 28 29 fs.close();30 }31 32 /**33 * 求出指定目录下的所有文件的平均数据块个数34 */35 private static float getHDFSAvgBlocks(FileSystem fs, Path path) throws Exception {36 // TODO Auto-generated method stub37 38 int totalFiles = 0; // 总文件数39 int totalBlocks = 0; // 总数据块数40 41 RemoteIterator listFiles = fs.listFiles(path, false);42 43 while(listFiles.hasNext()){44 LocatedFileStatus next = listFiles.next();45 int length = next.getBlockLocations().length;46 totalBlocks += length;47 if(next.getLen() != 0){48 totalFiles++;49 }50 }51 System.out.println(totalBlocks+" : "+totalFiles);52 53 return totalBlocks * 1f / totalFiles;54 }55 56 } 11、编写程序统计出HDFS文件系统中的平均副本数（副本总数/总数据块数）
1 import org.apache.hadoop.conf.Configuration; 2 import org.apache.hadoop.fs.FileSystem; 3 import org.apache.hadoop.fs.LocatedFileStatus; 4 import org.apache.hadoop.fs.Path; 5 import org.apache.hadoop.fs.RemoteIterator; 6 7 /** 8 * 编写程序统计出HDFS文件系统中的平均副本数（副本总数/总数据块数） 9 * 比如：总共两个文件，一个文件5个数据块，每个数据块3个副本，第二个文件2个数据块，每个文件2个副本，最终的平均副本数 = （3*3 + 2*2）/（3+2）= 2.810 */11 public class Exam3_HDSFAvgBlockCopys {12 13 public static void main(String[] args) throws Exception {14 15 16 Configuration conf = new Configuration();17 conf.set("fs.defaultFS", "hdfs://hadoop02:9000");18 System.setProperty("HADOOP_USER_NAME", "hadoop");19 FileSystem fs = FileSystem.get(conf);20 21 22 Path path = new Path("/");23 float avgHDFSBlockCopys = getHDFSAvgBlockCopys(fs, path);24 System.out.println("HDFS的平均数据块个数为：" + avgHDFSBlockCopys);25 26 27 fs.close();28 }29 30 /**31 * 求出指定目录下的所有文件的平均数据块个数32 */33 private static float getHDFSAvgBlockCopys(FileSystem fs, Path path) throws Exception {34 // TODO Auto-generated method stub35 36 int totalCopy = 0; // 总副本数37 int totalBlocks = 0; // 总数据块数38 39 RemoteIterator listFiles = fs.listFiles(path, false);40 41 while(listFiles.hasNext()){42 LocatedFileStatus next = listFiles.next();43 44 int length = next.getBlockLocations().length;45 short replication = next.getReplication();46 47 totalBlocks += length;48 totalCopy += length * replication;49 }50 System.out.println(totalCopy+" : "+totalBlocks);51 52 return totalCopy * 1f / totalBlocks;53 }54 55 } 12、统计HDFS整个文件系统中的不足指定数据块大小的数据块的比例
1 import java.io.IOException; 2 3 import org.apache.hadoop.conf.Configuration; 4 import org.apache.hadoop.fs.BlockLocation; 5 import org.apache.hadoop.fs.FileSystem; 6 import org.apache.hadoop.fs.LocatedFileStatus; 7 import org.apache.hadoop.fs.Path; 8 import org.apache.hadoop.fs.RemoteIterator; 9 10 /**11 * 统计HDFS整个文件系统中的不足指定数据块大小的数据块的比例12 * 比如指定的数据块大小是128M，总数据块有100个，不是大小为完整的128M的数据块有5个，那么不足指定数据块大小的数据块的比例就为5%13 * 注意：千万注意考虑不同文件的指定数据块大小可能不一致。所以千万不能用默认的128M一概而论14 */15 public class Exam4_LTBlockSize {16 17 public static void main(String[] args) throws Exception {18 19 Configuration conf = new Configuration();20 conf.set("fs.defaultFS", "hdfs://hadoop02:9000");21 System.setProperty("HADOOP_USER_NAME", "hadoop");22 FileSystem fs = FileSystem.get(conf);23 24 Path path = new Path("/");25 float avgHDFSBlockCopys = getLessThanBlocksizeBlocks(fs, path);26 System.out.println("HDFS的不足指定数据块大小的数据块数目为：" + avgHDFSBlockCopys);27 28 fs.close();29 }30 31 private static float getLessThanBlocksizeBlocks(FileSystem fs, Path path) throws Exception {32 // TODO Auto-generated method stub33 34 int totalBlocks = 0; // 总副本数35 int lessThenBlocksizeBlocks = 0; // 总数据块数36 37 RemoteIterator listFiles = fs.listFiles(path, false);38 39 while(listFiles.hasNext()){40 LocatedFileStatus next = listFiles.next();41 42 BlockLocation[] blockLocations = next.getBlockLocations();43 int length = blockLocations.length;44 45 if(length != 0){46 totalBlocks += length;47 long lastBlockSize = blockLocations[length - 1].getLength();48 long blockSize = next.getBlockSize();49 if(lastBlockSize < blockSize){50 lessThenBlocksizeBlocks++;51 }52 }53 }54 System.out.println(lessThenBlocksizeBlocks+" : "+totalBlocks);55 56 return lessThenBlocksizeBlocks * 1f / totalBlocks;57 }58 } 13、统计出一个给定数组的蓄水总量（把数组的每个位置的数看是做地势高低）
1 /** 2 统计出一个给定数组的蓄水总量（把数组的每个位置的数看是做地势高低） 3 比如：int[] intArray = new int[]{4,3,2,5,6,4,4,7} 4 能蓄水：[0,1,2,0,0,2,2,0] 所以总量是：7 5 6 核心思路：把数组切成很多个 01数组，每一层一个01数组，统计每个01数组中的合法0的总个数（数组的左边第一个1的中间区间中的0的个数）即可 7 */ 8 public class Exam5_WaterStoreOfArray { 9 10 public static void main(String[] args) { 11 12 // int[] intArray = new int[]{4,3,2,5,6,4,4,7}; 13 // int[] intArray = new int[]{1,2,3,4,5,6}; 14 int[] intArray = new int[]{3,1,2,7,3,8,4,9,5,6}; 15 16 int totalWater = getArrayWater(intArray); 17 System.out.println(totalWater); 18 } 19 20 /** 21 * 求出数组中的水数 22 */ 23 private static int getArrayWater(int[] intArray) { 24 25 int findMaxValueOfArray = findMaxValueOfArray(intArray); 26 int findMinValueOfArray = findMinValueOfArray(intArray); 27 int length = intArray.length; 28 29 int totalWater = 0; 30 31 // 循环次数就是最大值和最小值的差 32 for(int i=findMinValueOfArray; i i){ 37 tempArray[j] = 1; 38 }else{ 39 tempArray[j] = 0; 40 } 41 } 42 // 获取每一个01数组的合法0个数 43 int waterOfOneZeroArray = getWaterOfOneZeroArray(tempArray); 44 totalWater += waterOfOneZeroArray; 45 } 46 return totalWater; 47 } 48 49 50 /** 51 * 寻找逻辑是：从左右开始各找一个1，然后这两个1之间的所有0的个数，就是水数 52 */ 53 private static int getWaterOfOneZeroArray(int[] tempArray) { 54 55 int length = tempArray.length; 56 int toatalWater = 0; 57 58 // 找左边的1 59 int i = 0; 60 while(i < length){ 61 if(tempArray[i] == 1){ 62 break; 63 } 64 i++; 65 } 66 67 // 从右边开始找1 68 int j=length-1; 69 while(j >= i){ 70 if(tempArray[j] == 1){ 71 break; 72 } 73 j--; 74 } 75 76 // 找以上两个1之间的0的个数。 77 if(i == j || i + 1 == j){ 78 return 0; 79 }else{ 80 for(int k=i+1; k max){103 max = intArray[i];104 }105 }106 return max;107 }108 }109 110 /**111 * 找出一个数组中的最小值112 */113 public static int findMinValueOfArray(int[] intArray){114 int length = intArray.length;115 if(length == 0){116 return 0;117 }else if(length == 1){118 return intArray[0];119 }else{120 int min = intArray[0];121 for(int i=1; i