@Override public void configureJobConf(TableDesc tableDesc, JobConf jobConf) throws IOException { TableMapReduceUtil.addDependencyJars(jobConf, getClass()); }
addDependencyJars(job); addDependencyJarsForClasses(job.getConfiguration(), MetricRegistry.class);
@Override public void configureJobConf(TableDesc tableDesc, JobConf jobConf) throws IOException { super.configureJobConf(tableDesc, jobConf); TableMapReduceUtil.addDependencyJars(jobConf, keyClass); }
TextOutputFormat.setOutputPath(job, new Path(inputDir.getParent(), "outputs")); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.addDependencyJarsForClasses(job.getConfiguration(), Histogram.class, // yammer metrics
public int runGenerator(int numMappers, long numNodes, Path tmpOutput, Integer width, Integer wrapMultiplier, Integer numWalkers) throws Exception { LOG.info("Running Generator with numMappers=" + numMappers +", numNodes=" + numNodes); createSchema(); job = Job.getInstance(getConf()); job.setJobName("Link Generator"); job.setNumReduceTasks(0); job.setJarByClass(getClass()); FileInputFormat.setInputPaths(job, tmpOutput); job.setInputFormatClass(OneFilePerMapperSFIF.class); job.setOutputKeyClass(NullWritable.class); job.setOutputValueClass(NullWritable.class); setJobConf(job, numMappers, numNodes, width, wrapMultiplier, numWalkers); setMapperForGenerator(job); job.setOutputFormatClass(NullOutputFormat.class); job.getConfiguration().setBoolean("mapreduce.map.speculative", false); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.addDependencyJarsForClasses(job.getConfiguration(), AbstractHBaseTool.class); TableMapReduceUtil.initCredentials(job); boolean success = jobCompletion(job); return success ? 0 : 1; }
CellSerialization.class.getName()); if (addDependencyJars) { addDependencyJars(job);
protected Job doLoad(Configuration conf, HTableDescriptor htd) throws Exception { Path outputDir = getTestDir(TEST_NAME, "load-output"); LOG.info("Load output dir: " + outputDir); NMapInputFormat.setNumMapTasks(conf, conf.getInt(NUM_MAP_TASKS_KEY, NUM_MAP_TASKS_DEFAULT)); conf.set(TABLE_NAME_KEY, htd.getTableName().getNameAsString()); Job job = Job.getInstance(conf); job.setJobName(TEST_NAME + " Load for " + htd.getTableName()); job.setJarByClass(this.getClass()); setMapperClass(job); job.setInputFormatClass(NMapInputFormat.class); job.setNumReduceTasks(0); setJobScannerConf(job); FileOutputFormat.setOutputPath(job, outputDir); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.addDependencyJarsForClasses(job.getConfiguration(), AbstractHBaseTool.class); TableMapReduceUtil.initCredentials(job); assertTrue(job.waitForCompletion(true)); return job; }
TableMapReduceUtil.addDependencyJars( jobConf, HBaseStorageHandler.class, TableInputFormatBase.class, counterClass); } else { TableMapReduceUtil.addDependencyJars( jobConf, HBaseStorageHandler.class, TableInputFormatBase.class); TableMapReduceUtil.addDependencyJars(jobConf, MetricRegistry.class); TableMapReduceUtil.addDependencyJars(copy); merged.addAll(copy.getConfiguration().getStringCollection("tmpjars")); jobConf.set("tmpjars", StringUtils.arrayToString(merged.toArray(new String[0])));
public static void configureIncrementalLoadMap(Job job, TableDescriptor tableDescriptor) throws IOException { Configuration conf = job.getConfiguration(); job.setOutputKeyClass(ImmutableBytesWritable.class); job.setOutputValueClass(MapReduceExtendedCell.class); job.setOutputFormatClass(HFileOutputFormat2.class); ArrayList<TableDescriptor> singleTableDescriptor = new ArrayList<>(1); singleTableDescriptor.add(tableDescriptor); conf.set(OUTPUT_TABLE_NAME_CONF_KEY, tableDescriptor.getTableName().getNameAsString()); // Set compression algorithms based on column families conf.set(COMPRESSION_FAMILIES_CONF_KEY, serializeColumnFamilyAttribute(compressionDetails, singleTableDescriptor)); conf.set(BLOCK_SIZE_FAMILIES_CONF_KEY, serializeColumnFamilyAttribute(blockSizeDetails, singleTableDescriptor)); conf.set(BLOOM_TYPE_FAMILIES_CONF_KEY, serializeColumnFamilyAttribute(bloomTypeDetails, singleTableDescriptor)); conf.set(BLOOM_PARAM_FAMILIES_CONF_KEY, serializeColumnFamilyAttribute(bloomParamDetails, singleTableDescriptor)); conf.set(DATABLOCK_ENCODING_FAMILIES_CONF_KEY, serializeColumnFamilyAttribute(dataBlockEncodingDetails, singleTableDescriptor)); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.initCredentials(job); LOG.info("Incremental table " + tableDescriptor.getTableName() + " output configured."); }
TableMapReduceUtil.addDependencyJars(job.getConfiguration(), org.apache.hbase.thirdparty.com.google.common.base.Preconditions.class); } else {
TableMapReduceUtil.addDependencyJars(job);
job.setJobName(jobname); job.setJarByClass(ExportSnapshot.class); TableMapReduceUtil.addDependencyJars(job); job.setMapperClass(ExportMapper.class); job.setInputFormatClass(ExportSnapshotInputFormat.class);
configureDataBlockEncoding(tableDescriptor, conf); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.initCredentials(job); LOG.info("Incremental table " + regionLocator.getName() + " output configured.");
public static void configureIncrementalLoadMap(Job job, Table table) throws IOException { Configuration conf = job.getConfiguration(); job.setOutputKeyClass(ImmutableBytesWritable.class); job.setOutputValueClass(KeyValue.class); job.setOutputFormatClass(HFileOutputFormat3.class); // Set compression algorithms based on column families configureCompression(conf, table.getTableDescriptor()); configureBloomType(table.getTableDescriptor(), conf); configureBlockSize(table.getTableDescriptor(), conf); HTableDescriptor tableDescriptor = table.getTableDescriptor(); configureDataBlockEncoding(tableDescriptor, conf); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.initCredentials(job); LOG.info("Incremental table " + table.getName() + " output configured."); }
/** * Tests an MR Scan initialized from properties set in the Configuration. */ protected void testScanFromConfiguration(String start, String stop, String last) throws IOException, InterruptedException, ClassNotFoundException { String jobName = "ScanFromConfig" + (start != null ? start.toUpperCase(Locale.ROOT) : "Empty") + "To" + (stop != null ? stop.toUpperCase(Locale.ROOT) : "Empty"); Configuration c = new Configuration(TEST_UTIL.getConfiguration()); c.set(TableInputFormat.INPUT_TABLE, TABLE_NAME.getNameAsString()); c.set(TableInputFormat.SCAN_COLUMN_FAMILY, Bytes.toString(INPUT_FAMILYS[0]) + ", " + Bytes.toString(INPUT_FAMILYS[1])); c.set(KEY_STARTROW, start != null ? start : ""); c.set(KEY_LASTROW, last != null ? last : ""); if (start != null) { c.set(TableInputFormat.SCAN_ROW_START, start); } if (stop != null) { c.set(TableInputFormat.SCAN_ROW_STOP, stop); } Job job = Job.getInstance(c, jobName); job.setMapperClass(ScanMapper.class); job.setReducerClass(ScanReducer.class); job.setMapOutputKeyClass(ImmutableBytesWritable.class); job.setMapOutputValueClass(ImmutableBytesWritable.class); job.setInputFormatClass(TableInputFormat.class); job.setNumReduceTasks(1); FileOutputFormat.setOutputPath(job, new Path(job.getJobName())); TableMapReduceUtil.addDependencyJars(job); assertTrue(job.waitForCompletion(true)); }
private void doMapReduce(final Class<? extends Test> cmd) throws IOException, InterruptedException, ClassNotFoundException { Configuration conf = getConf(); Path inputDir = writeInputFile(conf); conf.set(EvaluationMapTask.CMD_KEY, cmd.getName()); conf.set(EvaluationMapTask.PE_KEY, getClass().getName()); Job job = Job.getInstance(conf); job.setJarByClass(PerformanceEvaluation.class); job.setJobName("HBase Performance Evaluation"); job.setInputFormatClass(PeInputFormat.class); PeInputFormat.setInputPaths(job, inputDir); job.setOutputKeyClass(LongWritable.class); job.setOutputValueClass(LongWritable.class); job.setMapperClass(EvaluationMapTask.class); job.setReducerClass(LongSumReducer.class); job.setNumReduceTasks(1); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, new Path(inputDir.getParent(), "outputs")); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.initCredentials(job); job.waitForCompletion(true); }
/** * Look for jars we expect to be on the classpath by name. */ @Test public void testAddDependencyJars() throws Exception { Job job = new Job(); TableMapReduceUtil.addDependencyJars(job); String tmpjars = job.getConfiguration().get("tmpjars"); // verify presence of modules assertTrue(tmpjars.contains("hbase-common")); assertTrue(tmpjars.contains("hbase-protocol")); assertTrue(tmpjars.contains("hbase-client")); assertTrue(tmpjars.contains("hbase-hadoop-compat")); assertTrue(tmpjars.contains("hbase-server")); // verify presence of 3rd party dependencies. assertTrue(tmpjars.contains("zookeeper")); assertTrue(tmpjars.contains("netty")); assertTrue(tmpjars.contains("protobuf")); assertTrue(tmpjars.contains("guava")); assertTrue(tmpjars.contains("htrace")); }
job.getConfiguration().setBoolean("mapreduce.reduce.speculative", false); TableMapReduceUtil.initTableReducerJob(COMMON_TABLE_NAME, null, job, null, null, null, null); TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.addDependencyJars(job.getConfiguration(), AbstractHBaseTool.class); TableMapReduceUtil.initCredentials(job); job.setNumReduceTasks(0);
private int doVerify(Path outputDir, int numReducers) throws IOException, InterruptedException, ClassNotFoundException { job = new Job(getConf()); job.setJobName("Link Verifier"); job.setNumReduceTasks(numReducers); job.setJarByClass(getClass()); setJobScannerConf(job); Scan scan = new Scan(); scan.addColumn(FAMILY_NAME, COLUMN_PREV); scan.setCaching(10000); scan.setCacheBlocks(false); String[] split = labels.split(COMMA); scan.setAuthorizations(new Authorizations(split[this.labelIndex * 2], split[(this.labelIndex * 2) + 1])); TableMapReduceUtil.initTableMapperJob(tableName.getName(), scan, VerifyMapper.class, BytesWritable.class, BytesWritable.class, job); TableMapReduceUtil.addDependencyJars(job.getConfiguration(), AbstractHBaseTool.class); job.getConfiguration().setBoolean("mapreduce.map.speculative", false); job.setReducerClass(VerifyReducer.class); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); boolean success = job.waitForCompletion(true); return success ? 0 : 1; }