List of usage examples for org.apache.hadoop.fs FileSystem getScheme
public String getScheme()
From source file:cascading.flow.tez.planner.Hadoop2TezFlowStepJob.java
License:Open Source License
private Path prepareEnsureStagingDir(TezConfiguration workingConf) throws IOException { String stepStagingPath = createStepStagingPath(); workingConf.set(TezConfiguration.TEZ_AM_STAGING_DIR, stepStagingPath); Path stagingDir = new Path(stepStagingPath); FileSystem fileSystem = FileSystem.get(workingConf); stagingDir = fileSystem.makeQualified(stagingDir); TokenCache.obtainTokensForNamenodes(new Credentials(), new Path[] { stagingDir }, workingConf); TezClientUtils.ensureStagingDirExists(workingConf, stagingDir); if (fileSystem.getScheme().startsWith("file:/")) new File(stagingDir.toUri()).mkdirs(); return stagingDir; }
From source file:com.datatorrent.contrib.hive.HiveOperator.java
License:Apache License
/** * Override this method to change the FileSystem instance that is used by the * operator./*from www. j a v a 2s. c o m*/ * * @return A FileSystem object. * @throws IOException */ protected FileSystem getHDFSInstance() throws IOException { FileSystem tempFS = FileSystem.newInstance(new Path(store.filepath).toUri(), new Configuration()); if (!tempFS.getScheme().equalsIgnoreCase("hdfs")) { localString = " local"; } return tempFS; }
From source file:com.thinkbiganalytics.kylo.hadoop.FileSystemUtil.java
License:Apache License
/** * Adds Hadoop {@link FileSystem} classes to the Hadoop configuration. *//*from ww w. j a v a 2 s . c o m*/ public static void registerFileSystems(@Nonnull final Iterable<FileSystem> fileSystems, @Nonnull final Configuration conf) { for (final FileSystem fs : fileSystems) { try { final String scheme = fs.getScheme(); final Class clazz = fs.getClass(); log.debug("Found {} FileSystem using class: {}", scheme, clazz); conf.setClass("fs." + scheme + ".impl", clazz, FileSystem.class); } catch (final Exception e) { log.warn("Cannot load FileSystem using class: {}: {}", fs.getClass().getName(), e, e); } } }
From source file:com.uber.hoodie.common.table.log.HoodieLogFormatWriter.java
License:Apache License
/** * @param fs/*from w w w.j a va 2s. co m*/ * @param logFile * @param bufferSize * @param replication * @param sizeThreshold */ HoodieLogFormatWriter(FileSystem fs, HoodieLogFile logFile, Integer bufferSize, Short replication, Long sizeThreshold, String logWriteToken, String rolloverLogWriteToken) throws IOException, InterruptedException { this.fs = fs; this.logFile = logFile; this.sizeThreshold = sizeThreshold; this.bufferSize = bufferSize; this.replication = replication; this.logWriteToken = logWriteToken; this.rolloverLogWriteToken = rolloverLogWriteToken; Path path = logFile.getPath(); if (fs.exists(path)) { boolean isAppendSupported = StorageSchemes.isAppendSupported(fs.getScheme()); if (isAppendSupported) { log.info(logFile + " exists. Appending to existing file"); try { this.output = fs.append(path, bufferSize); } catch (RemoteException e) { log.warn("Remote Exception, attempting to handle or recover lease", e); handleAppendExceptionOrRecoverLease(path, e); } catch (IOException ioe) { if (ioe.getMessage().toLowerCase().contains("not supported")) { // may still happen if scheme is viewfs. isAppendSupported = false; } else { throw ioe; } } } if (!isAppendSupported) { this.logFile = logFile.rollOver(fs, rolloverLogWriteToken); log.info("Append not supported.. Rolling over to " + logFile); createNewFile(); } } else { log.info(logFile + " does not exist. Create a new file"); // Block size does not matter as we will always manually autoflush createNewFile(); } }
From source file:eu.stratosphere.yarn.Client.java
License:Apache License
public void run(String[] args) throws Exception { if (UserGroupInformation.isSecurityEnabled()) { throw new RuntimeException("Stratosphere YARN client does not have security support right now." + "File a bug, we will fix it asap"); }//from w w w .j a va 2 s .c o m //Utils.logFilesInCurrentDirectory(LOG); // // Command Line Options // Options options = new Options(); options.addOption(VERBOSE); options.addOption(STRATOSPHERE_CONF_DIR); options.addOption(STRATOSPHERE_JAR); options.addOption(JM_MEMORY); options.addOption(TM_MEMORY); options.addOption(TM_CORES); options.addOption(CONTAINER); options.addOption(GEN_CONF); options.addOption(QUEUE); options.addOption(QUERY); options.addOption(SHIP_PATH); CommandLineParser parser = new PosixParser(); CommandLine cmd = null; try { cmd = parser.parse(options, args); } catch (MissingOptionException moe) { System.out.println(moe.getMessage()); printUsage(); System.exit(1); } if (System.getProperty("log4j.configuration") == null) { Logger root = Logger.getRootLogger(); root.removeAllAppenders(); PatternLayout layout = new PatternLayout("%d{HH:mm:ss,SSS} %-5p %-60c %x - %m%n"); ConsoleAppender appender = new ConsoleAppender(layout, "System.err"); root.addAppender(appender); if (cmd.hasOption(VERBOSE.getOpt())) { root.setLevel(Level.DEBUG); LOG.debug("CLASSPATH: " + System.getProperty("java.class.path")); } else { root.setLevel(Level.INFO); } } // Jar Path Path localJarPath; if (cmd.hasOption(STRATOSPHERE_JAR.getOpt())) { String userPath = cmd.getOptionValue(STRATOSPHERE_JAR.getOpt()); if (!userPath.startsWith("file://")) { userPath = "file://" + userPath; } localJarPath = new Path(userPath); } else { localJarPath = new Path( "file://" + Client.class.getProtectionDomain().getCodeSource().getLocation().getPath()); } if (cmd.hasOption(GEN_CONF.getOpt())) { LOG.info("Placing default configuration in current directory"); File outFile = generateDefaultConf(localJarPath); LOG.info("File written to " + outFile.getAbsolutePath()); System.exit(0); } // Conf Path Path confPath = null; String confDirPath = ""; if (cmd.hasOption(STRATOSPHERE_CONF_DIR.getOpt())) { confDirPath = cmd.getOptionValue(STRATOSPHERE_CONF_DIR.getOpt()) + "/"; File confFile = new File(confDirPath + CONFIG_FILE_NAME); if (!confFile.exists()) { LOG.fatal("Unable to locate configuration file in " + confFile); System.exit(1); } confPath = new Path(confFile.getAbsolutePath()); } else { System.out.println("No configuration file has been specified"); // no configuration path given. // -> see if there is one in the current directory File currDir = new File("."); File[] candidates = currDir.listFiles(new FilenameFilter() { @Override public boolean accept(final File dir, final String name) { return name != null && name.endsWith(".yaml"); } }); if (candidates == null || candidates.length == 0) { System.out.println( "No configuration file has been found in current directory.\n" + "Copying default."); File outFile = generateDefaultConf(localJarPath); confPath = new Path(outFile.toURI()); } else { if (candidates.length > 1) { System.out.println("Multiple .yaml configuration files were found in the current directory\n" + "Please specify one explicitly"); System.exit(1); } else if (candidates.length == 1) { confPath = new Path(candidates[0].toURI()); } } } List<File> shipFiles = new ArrayList<File>(); // path to directory to ship if (cmd.hasOption(SHIP_PATH.getOpt())) { String shipPath = cmd.getOptionValue(SHIP_PATH.getOpt()); File shipDir = new File(shipPath); if (shipDir.isDirectory()) { shipFiles = new ArrayList<File>(Arrays.asList(shipDir.listFiles(new FilenameFilter() { @Override public boolean accept(File dir, String name) { return !(name.equals(".") || name.equals("..")); } }))); } else { LOG.warn("Ship directory is not a directory!"); } } boolean hasLog4j = false; //check if there is a log4j file if (confDirPath.length() > 0) { File l4j = new File(confDirPath + "/log4j.properties"); if (l4j.exists()) { shipFiles.add(l4j); hasLog4j = true; } } // queue String queue = "default"; if (cmd.hasOption(QUEUE.getOpt())) { queue = cmd.getOptionValue(QUEUE.getOpt()); } // JobManager Memory int jmMemory = 512; if (cmd.hasOption(JM_MEMORY.getOpt())) { jmMemory = Integer.valueOf(cmd.getOptionValue(JM_MEMORY.getOpt())); } // Task Managers memory int tmMemory = 1024; if (cmd.hasOption(TM_MEMORY.getOpt())) { tmMemory = Integer.valueOf(cmd.getOptionValue(TM_MEMORY.getOpt())); } // Task Managers vcores int tmCores = 1; if (cmd.hasOption(TM_CORES.getOpt())) { tmCores = Integer.valueOf(cmd.getOptionValue(TM_CORES.getOpt())); } Utils.getStratosphereConfiguration(confPath.toUri().getPath()); int jmPort = GlobalConfiguration.getInteger(ConfigConstants.JOB_MANAGER_IPC_PORT_KEY, 0); if (jmPort == 0) { LOG.warn("Unable to find job manager port in configuration!"); jmPort = ConfigConstants.DEFAULT_JOB_MANAGER_IPC_PORT; } conf = Utils.initializeYarnConfiguration(); // intialize HDFS LOG.info("Copy App Master jar from local filesystem and add to local environment"); // Copy the application master jar to the filesystem // Create a local resource to point to the destination jar path final FileSystem fs = FileSystem.get(conf); if (fs.getScheme().startsWith("file")) { LOG.warn("The file system scheme is '" + fs.getScheme() + "'. This indicates that the " + "specified Hadoop configuration path is wrong and the sytem is using the default Hadoop configuration values." + "The Stratosphere YARN client needs to store its files in a distributed file system"); } // Create yarnClient final YarnClient yarnClient = YarnClient.createYarnClient(); yarnClient.init(conf); yarnClient.start(); // Query cluster for metrics if (cmd.hasOption(QUERY.getOpt())) { showClusterMetrics(yarnClient); } if (!cmd.hasOption(CONTAINER.getOpt())) { LOG.fatal("Missing required argument " + CONTAINER.getOpt()); printUsage(); yarnClient.stop(); System.exit(1); } // TM Count final int taskManagerCount = Integer.valueOf(cmd.getOptionValue(CONTAINER.getOpt())); System.out.println("Using values:"); System.out.println("\tContainer Count = " + taskManagerCount); System.out.println("\tJar Path = " + localJarPath.toUri().getPath()); System.out.println("\tConfiguration file = " + confPath.toUri().getPath()); System.out.println("\tJobManager memory = " + jmMemory); System.out.println("\tTaskManager memory = " + tmMemory); System.out.println("\tTaskManager cores = " + tmCores); // Create application via yarnClient YarnClientApplication app = yarnClient.createApplication(); GetNewApplicationResponse appResponse = app.getNewApplicationResponse(); Resource maxRes = appResponse.getMaximumResourceCapability(); if (tmMemory > maxRes.getMemory() || tmCores > maxRes.getVirtualCores()) { LOG.fatal("The cluster does not have the requested resources for the TaskManagers available!\n" + "Maximum Memory: " + maxRes.getMemory() + ", Maximum Cores: " + tmCores); yarnClient.stop(); System.exit(1); } if (jmMemory > maxRes.getMemory()) { LOG.fatal("The cluster does not have the requested resources for the JobManager available!\n" + "Maximum Memory: " + maxRes.getMemory()); yarnClient.stop(); System.exit(1); } int totalMemoryRequired = jmMemory + tmMemory * taskManagerCount; ClusterResourceDescription freeClusterMem = getCurrentFreeClusterResources(yarnClient); if (freeClusterMem.totalFreeMemory < totalMemoryRequired) { LOG.fatal("This YARN session requires " + totalMemoryRequired + "MB of memory in the cluster. " + "There are currently only " + freeClusterMem.totalFreeMemory + "MB available."); yarnClient.stop(); System.exit(1); } if (tmMemory > freeClusterMem.containerLimit) { LOG.fatal("The requested amount of memory for the TaskManagers (" + tmMemory + "MB) is more than " + "the largest possible YARN container: " + freeClusterMem.containerLimit); yarnClient.stop(); System.exit(1); } if (jmMemory > freeClusterMem.containerLimit) { LOG.fatal("The requested amount of memory for the JobManager (" + jmMemory + "MB) is more than " + "the largest possible YARN container: " + freeClusterMem.containerLimit); yarnClient.stop(); System.exit(1); } // respect custom JVM options in the YAML file final String javaOpts = GlobalConfiguration.getString(ConfigConstants.STRATOSPHERE_JVM_OPTIONS, ""); // Set up the container launch context for the application master ContainerLaunchContext amContainer = Records.newRecord(ContainerLaunchContext.class); String amCommand = "$JAVA_HOME/bin/java" + " -Xmx" + Utils.calculateHeapSize(jmMemory) + "M " + javaOpts; if (hasLog4j) { amCommand += " -Dlog.file=\"" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/jobmanager-log4j.log\" -Dlog4j.configuration=file:log4j.properties"; } amCommand += " eu.stratosphere.yarn.ApplicationMaster" + " " + " 1>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/jobmanager-stdout.log" + " 2>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/jobmanager-stderr.log"; amContainer.setCommands(Collections.singletonList(amCommand)); System.err.println("amCommand=" + amCommand); // Set-up ApplicationSubmissionContext for the application ApplicationSubmissionContext appContext = app.getApplicationSubmissionContext(); final ApplicationId appId = appContext.getApplicationId(); // Setup jar for ApplicationMaster LocalResource appMasterJar = Records.newRecord(LocalResource.class); LocalResource stratosphereConf = Records.newRecord(LocalResource.class); Path remotePathJar = Utils.setupLocalResource(conf, fs, appId.toString(), localJarPath, appMasterJar, fs.getHomeDirectory()); Path remotePathConf = Utils.setupLocalResource(conf, fs, appId.toString(), confPath, stratosphereConf, fs.getHomeDirectory()); Map<String, LocalResource> localResources = new HashMap<String, LocalResource>(2); localResources.put("stratosphere.jar", appMasterJar); localResources.put("stratosphere-conf.yaml", stratosphereConf); // setup security tokens (code from apache storm) final Path[] paths = new Path[3 + shipFiles.size()]; StringBuffer envShipFileList = new StringBuffer(); // upload ship files for (int i = 0; i < shipFiles.size(); i++) { File shipFile = shipFiles.get(i); LocalResource shipResources = Records.newRecord(LocalResource.class); Path shipLocalPath = new Path("file://" + shipFile.getAbsolutePath()); paths[3 + i] = Utils.setupLocalResource(conf, fs, appId.toString(), shipLocalPath, shipResources, fs.getHomeDirectory()); localResources.put(shipFile.getName(), shipResources); envShipFileList.append(paths[3 + i]); if (i + 1 < shipFiles.size()) { envShipFileList.append(','); } } paths[0] = remotePathJar; paths[1] = remotePathConf; paths[2] = new Path(fs.getHomeDirectory(), ".stratosphere/" + appId.toString() + "/"); FsPermission permission = new FsPermission(FsAction.ALL, FsAction.ALL, FsAction.ALL); fs.setPermission(paths[2], permission); // set permission for path. Utils.setTokensFor(amContainer, paths, this.conf); amContainer.setLocalResources(localResources); fs.close(); // Setup CLASSPATH for ApplicationMaster Map<String, String> appMasterEnv = new HashMap<String, String>(); Utils.setupEnv(conf, appMasterEnv); // set configuration values appMasterEnv.put(Client.ENV_TM_COUNT, String.valueOf(taskManagerCount)); appMasterEnv.put(Client.ENV_TM_CORES, String.valueOf(tmCores)); appMasterEnv.put(Client.ENV_TM_MEMORY, String.valueOf(tmMemory)); appMasterEnv.put(Client.STRATOSPHERE_JAR_PATH, remotePathJar.toString()); appMasterEnv.put(Client.ENV_APP_ID, appId.toString()); appMasterEnv.put(Client.ENV_CLIENT_HOME_DIR, fs.getHomeDirectory().toString()); appMasterEnv.put(Client.ENV_CLIENT_SHIP_FILES, envShipFileList.toString()); appMasterEnv.put(Client.ENV_CLIENT_USERNAME, UserGroupInformation.getCurrentUser().getShortUserName()); amContainer.setEnvironment(appMasterEnv); // Set up resource type requirements for ApplicationMaster Resource capability = Records.newRecord(Resource.class); capability.setMemory(jmMemory); capability.setVirtualCores(1); appContext.setApplicationName("Stratosphere"); // application name appContext.setAMContainerSpec(amContainer); appContext.setResource(capability); appContext.setQueue(queue); // file that we write into the conf/ dir containing the jobManager address. final File addrFile = new File(confDirPath + CliFrontend.JOBMANAGER_ADDRESS_FILE); Runtime.getRuntime().addShutdownHook(new Thread() { @Override public void run() { try { LOG.info("Killing the Stratosphere-YARN application."); yarnClient.killApplication(appId); LOG.info("Deleting files in " + paths[2]); FileSystem shutFS = FileSystem.get(conf); shutFS.delete(paths[2], true); // delete conf and jar file. shutFS.close(); } catch (Exception e) { LOG.warn("Exception while killing the YARN application", e); } try { addrFile.delete(); } catch (Exception e) { LOG.warn("Exception while deleting the jobmanager address file", e); } LOG.info("YARN Client is shutting down"); yarnClient.stop(); } }); LOG.info("Submitting application master " + appId); yarnClient.submitApplication(appContext); ApplicationReport appReport = yarnClient.getApplicationReport(appId); YarnApplicationState appState = appReport.getYarnApplicationState(); boolean told = false; char[] el = { '/', '|', '\\', '-' }; int i = 0; while (appState != YarnApplicationState.FINISHED && appState != YarnApplicationState.KILLED && appState != YarnApplicationState.FAILED) { if (!told && appState == YarnApplicationState.RUNNING) { System.err .println("Stratosphere JobManager is now running on " + appReport.getHost() + ":" + jmPort); System.err.println("JobManager Web Interface: " + appReport.getTrackingUrl()); // write jobmanager connect information PrintWriter out = new PrintWriter(addrFile); out.println(appReport.getHost() + ":" + jmPort); out.close(); addrFile.setReadable(true, false); // readable for all. told = true; } if (!told) { System.err.print(el[i++] + "\r"); if (i == el.length) { i = 0; } Thread.sleep(500); // wait for the application to switch to RUNNING } else { Thread.sleep(5000); } appReport = yarnClient.getApplicationReport(appId); appState = appReport.getYarnApplicationState(); } LOG.info("Application " + appId + " finished with" + " state " + appState + " at " + appReport.getFinishTime()); if (appState == YarnApplicationState.FAILED || appState == YarnApplicationState.KILLED) { LOG.warn("Application failed. Diagnostics " + appReport.getDiagnostics()); } }
From source file:gobblin.util.filesystem.FileSystemInstrumentation.java
License:Apache License
public FileSystemInstrumentation(FileSystem underlying) { super(underlying.getScheme(), underlying.getScheme()); this.underlyingFs = underlying; }
From source file:gobblin.util.filesystem.InstrumentedFileSystem.java
License:Apache License
public InstrumentedFileSystem(String scheme, FileSystem underlyingFileSystem) { super(scheme, underlyingFileSystem.getScheme()); this.underlyingFs = underlyingFileSystem; }
From source file:io.druid.indexer.JobHelper.java
License:Apache License
public static DataSegment serializeOutIndex(final DataSegment segmentTemplate, final Configuration configuration, final Progressable progressable, final TaskAttemptID taskAttemptID, final File mergedBase, final Path segmentBasePath) throws IOException { final FileSystem outputFS = FileSystem.get(segmentBasePath.toUri(), configuration); final Path tmpPath = new Path(segmentBasePath, String.format("index.zip.%d", taskAttemptID.getId())); final AtomicLong size = new AtomicLong(0L); final DataPusher zipPusher = (DataPusher) RetryProxy.create(DataPusher.class, new DataPusher() { @Override//w w w . j a v a2s .c om public long push() throws IOException { try (OutputStream outputStream = outputFS.create(tmpPath, true, DEFAULT_FS_BUFFER_SIZE, progressable)) { size.set(zipAndCopyDir(mergedBase, outputStream, progressable)); outputStream.flush(); } catch (IOException | RuntimeException exception) { log.error(exception, "Exception in retry loop"); throw exception; } return -1; } }, RetryPolicies.exponentialBackoffRetry(NUM_RETRIES, SECONDS_BETWEEN_RETRIES, TimeUnit.SECONDS)); zipPusher.push(); log.info("Zipped %,d bytes to [%s]", size.get(), tmpPath.toUri()); final Path finalIndexZipFilePath = new Path(segmentBasePath, "index.zip"); final URI indexOutURI = finalIndexZipFilePath.toUri(); final ImmutableMap<String, Object> loadSpec; // TODO: Make this a part of Pushers or Pullers switch (outputFS.getScheme()) { case "hdfs": loadSpec = ImmutableMap.<String, Object>of("type", "hdfs", "path", indexOutURI.toString()); break; case "s3": case "s3n": loadSpec = ImmutableMap.<String, Object>of("type", "s3_zip", "bucket", indexOutURI.getHost(), "key", indexOutURI.getPath().substring(1) // remove the leading "/" ); break; case "file": loadSpec = ImmutableMap.<String, Object>of("type", "local", "path", indexOutURI.getPath()); break; default: throw new IAE("Unknown file system scheme [%s]", outputFS.getScheme()); } final DataSegment finalSegment = segmentTemplate.withLoadSpec(loadSpec).withSize(size.get()) .withBinaryVersion(SegmentUtils.getVersionFromDir(mergedBase)); if (!renameIndexFiles(outputFS, tmpPath, finalIndexZipFilePath)) { throw new IOException(String.format("Unable to rename [%s] to [%s]", tmpPath.toUri().toString(), finalIndexZipFilePath.toUri().toString())); } writeSegmentDescriptor(outputFS, finalSegment, new Path(segmentBasePath, "descriptor.json"), progressable); return finalSegment; }
From source file:io.druid.indexer.JobHelper.java
License:Apache License
public static Path makeSegmentOutputPath(Path basePath, FileSystem fileSystem, String dataSource, String version, Interval interval, int partitionNum) { Path outputPath = new Path(prependFSIfNullScheme(fileSystem, basePath), "./" + dataSource); if ("hdfs".equals(fileSystem.getScheme())) { outputPath = new Path(outputPath, String.format("./%s_%s", interval.getStart().toString(ISODateTimeFormat.basicDateTime()), interval.getEnd().toString(ISODateTimeFormat.basicDateTime()))); outputPath = new Path(outputPath, version.replace(":", "_")); } else {/* ww w . j a v a 2s .c o m*/ outputPath = new Path(outputPath, String.format("./%s_%s", interval.getStart().toString(), interval.getEnd().toString())); outputPath = new Path(outputPath, String.format("./%s", version)); } outputPath = new Path(outputPath, Integer.toString(partitionNum)); return outputPath; }
From source file:io.druid.storage.hdfs.HdfsDataSegmentFinder.java
License:Apache License
@Override public Set<DataSegment> findSegments(String workingDirPathStr, boolean updateDescriptor) throws SegmentLoadingException { final Set<DataSegment> segments = Sets.newHashSet(); final Path workingDirPath = new Path(workingDirPathStr); FileSystem fs; try {/*from w w w . j a v a2 s . c o m*/ fs = workingDirPath.getFileSystem(config); log.info(fs.getScheme()); log.info("FileSystem URI:" + fs.getUri().toString()); if (!fs.exists(workingDirPath)) { throw new SegmentLoadingException("Working directory [%s] doesn't exist.", workingDirPath); } if (!fs.isDirectory(workingDirPath)) { throw new SegmentLoadingException("Working directory [%s] is not a directory!?", workingDirPath); } final RemoteIterator<LocatedFileStatus> it = fs.listFiles(workingDirPath, true); while (it.hasNext()) { final LocatedFileStatus locatedFileStatus = it.next(); final Path path = locatedFileStatus.getPath(); if (path.getName().endsWith("descriptor.json")) { final Path indexZip; final String descriptorParts[] = path.getName().split("_"); if (descriptorParts.length == 2 && descriptorParts[1].equals("descriptor.json") && org.apache.commons.lang.StringUtils.isNumeric(descriptorParts[0])) { indexZip = new Path(path.getParent(), StringUtils.format("%s_index.zip", descriptorParts[0])); } else { indexZip = new Path(path.getParent(), "index.zip"); } if (fs.exists(indexZip)) { final DataSegment dataSegment = mapper.readValue(fs.open(path), DataSegment.class); log.info("Found segment [%s] located at [%s]", dataSegment.getIdentifier(), indexZip); final Map<String, Object> loadSpec = dataSegment.getLoadSpec(); final String pathWithoutScheme = indexZip.toUri().getPath(); if (!loadSpec.get("type").equals(HdfsStorageDruidModule.SCHEME) || !loadSpec.get("path").equals(pathWithoutScheme)) { loadSpec.put("type", HdfsStorageDruidModule.SCHEME); loadSpec.put("path", pathWithoutScheme); if (updateDescriptor) { log.info("Updating loadSpec in descriptor.json at [%s] with new path [%s]", path, pathWithoutScheme); mapper.writeValue(fs.create(path, true), dataSegment); } } segments.add(dataSegment); } else { throw new SegmentLoadingException( "index.zip didn't exist at [%s] while descripter.json exists!?", indexZip); } } } } catch (IOException e) { throw new SegmentLoadingException(e, "Problems interacting with filesystem[%s].", workingDirPath); } return segments; }