View Javadoc

1   /**
2    *
3    * Licensed to the Apache Software Foundation (ASF) under one
4    * or more contributor license agreements.  See the NOTICE file
5    * distributed with this work for additional information
6    * regarding copyright ownership.  The ASF licenses this file
7    * to you under the Apache License, Version 2.0 (the
8    * "License"); you may not use this file except in compliance
9    * with the License.  You may obtain a copy of the License at
10   *
11   *     http://www.apache.org/licenses/LICENSE-2.0
12   *
13   * Unless required by applicable law or agreed to in writing, software
14   * distributed under the License is distributed on an "AS IS" BASIS,
15   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16   * See the License for the specific language governing permissions and
17   * limitations under the License.
18   */
19  package org.apache.hadoop.hbase.mapreduce;
20  
21  import java.io.ByteArrayInputStream;
22  import java.io.DataInput;
23  import java.io.DataInputStream;
24  import java.io.DataOutput;
25  import java.io.IOException;
26  import java.lang.reflect.InvocationTargetException;
27  import java.lang.reflect.Method;
28  import java.util.ArrayList;
29  import java.util.Collections;
30  import java.util.List;
31  import java.util.Locale;
32  import java.util.Map;
33  import java.util.TreeMap;
34  import java.util.UUID;
35  
36  import org.apache.commons.logging.Log;
37  import org.apache.commons.logging.LogFactory;
38  import org.apache.hadoop.conf.Configuration;
39  import org.apache.hadoop.fs.FileSystem;
40  import org.apache.hadoop.fs.Path;
41  import org.apache.hadoop.hbase.Cell;
42  import org.apache.hadoop.hbase.CellComparator;
43  import org.apache.hadoop.hbase.CellUtil;
44  import org.apache.hadoop.hbase.HBaseConfiguration;
45  import org.apache.hadoop.hbase.KeyValue;
46  import org.apache.hadoop.hbase.KeyValueUtil;
47  import org.apache.hadoop.hbase.TableName;
48  import org.apache.hadoop.hbase.Tag;
49  import org.apache.hadoop.hbase.ZooKeeperConnectionException;
50  import org.apache.hadoop.hbase.classification.InterfaceAudience;
51  import org.apache.hadoop.hbase.classification.InterfaceStability;
52  import org.apache.hadoop.hbase.client.Connection;
53  import org.apache.hadoop.hbase.client.ConnectionFactory;
54  import org.apache.hadoop.hbase.client.Delete;
55  import org.apache.hadoop.hbase.client.Durability;
56  import org.apache.hadoop.hbase.client.HBaseAdmin;
57  import org.apache.hadoop.hbase.client.Mutation;
58  import org.apache.hadoop.hbase.client.Put;
59  import org.apache.hadoop.hbase.client.RegionLocator;
60  import org.apache.hadoop.hbase.client.Result;
61  import org.apache.hadoop.hbase.client.Table;
62  import org.apache.hadoop.hbase.filter.Filter;
63  import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
64  import org.apache.hadoop.hbase.util.Bytes;
65  import org.apache.hadoop.hbase.zookeeper.ZKClusterId;
66  import org.apache.hadoop.hbase.zookeeper.ZooKeeperWatcher;
67  import org.apache.hadoop.io.RawComparator;
68  import org.apache.hadoop.io.WritableComparable;
69  import org.apache.hadoop.io.WritableComparator;
70  import org.apache.hadoop.mapreduce.Job;
71  import org.apache.hadoop.mapreduce.Partitioner;
72  import org.apache.hadoop.mapreduce.Reducer;
73  import org.apache.hadoop.mapreduce.TaskCounter;
74  import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
75  import org.apache.hadoop.mapreduce.lib.input.SequenceFileInputFormat;
76  import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
77  import org.apache.hadoop.mapreduce.lib.partition.TotalOrderPartitioner;
78  import org.apache.hadoop.util.GenericOptionsParser;
79  import org.apache.zookeeper.KeeperException;
80  
81  
82  /**
83   * Import data written by {@link Export}.
84   */
85  @InterfaceAudience.Public
86  @InterfaceStability.Stable
87  public class Import {
88    private static final Log LOG = LogFactory.getLog(Import.class);
89    final static String NAME = "import";
90    public final static String CF_RENAME_PROP = "HBASE_IMPORTER_RENAME_CFS";
91    public final static String BULK_OUTPUT_CONF_KEY = "import.bulk.output";
92    public final static String FILTER_CLASS_CONF_KEY = "import.filter.class";
93    public final static String FILTER_ARGS_CONF_KEY = "import.filter.args";
94    public final static String TABLE_NAME = "import.table.name";
95    public final static String WAL_DURABILITY = "import.wal.durability";
96    public final static String HAS_LARGE_RESULT= "import.bulk.hasLargeResult";
97  
98    public static class KeyValueWritableComparablePartitioner 
99        extends Partitioner<KeyValueWritableComparable, KeyValue> {
100     private static KeyValueWritableComparable[] START_KEYS = null;
101     @Override
102     public int getPartition(KeyValueWritableComparable key, KeyValue value,
103         int numPartitions) {
104       for (int i = 0; i < START_KEYS.length; ++i) {
105         if (key.compareTo(START_KEYS[i]) <= 0) {
106           return i;
107         }
108       }
109       return START_KEYS.length;
110     }
111   
112   }
113   
114   public static class KeyValueWritableComparable 
115       implements WritableComparable<KeyValueWritableComparable> {
116     
117     private KeyValue kv = null;
118     private CellComparator cellComparator = new CellComparator();
119     
120     static {                                       
121       // register this comparator
122       WritableComparator.define(KeyValueWritableComparable.class, 
123           new KeyValueWritableComparator());
124     }
125     
126     public KeyValueWritableComparable() {
127     }
128     
129     public KeyValueWritableComparable(KeyValue kv) {
130       this.kv = kv;
131     }
132     
133     @Override
134     public void write(DataOutput out) throws IOException {
135       KeyValue.write(kv, out);
136     }
137     
138     @Override
139     public void readFields(DataInput in) throws IOException {
140       kv = KeyValue.create(in);
141     }
142     
143     @Override
144     @edu.umd.cs.findbugs.annotations.SuppressWarnings(value="EQ_COMPARETO_USE_OBJECT_EQUALS",
145       justification="This is wrong, yes, but we should be purging Writables, not fixing them")
146     public int compareTo(KeyValueWritableComparable o) {
147       return cellComparator.compare(this.kv, ((KeyValueWritableComparable)o).kv);
148     }
149     
150     public static class KeyValueWritableComparator extends WritableComparator {
151     
152       @Override
153       public int compare(byte[] b1, int s1, int l1, byte[] b2, int s2, int l2) {
154         try {
155           KeyValueWritableComparable kv1 = new KeyValueWritableComparable();
156           kv1.readFields(new DataInputStream(new ByteArrayInputStream(b1, s1, l1)));
157           KeyValueWritableComparable kv2 = new KeyValueWritableComparable();
158           kv2.readFields(new DataInputStream(new ByteArrayInputStream(b2, s2, l2)));
159           return compare(kv1, kv2);
160         } catch (IOException e) {
161           throw new RuntimeException(e);
162         } 
163       }
164       
165     }
166   
167   }
168   
169   public static class KeyValueReducer
170       extends
171       Reducer<KeyValueWritableComparable, KeyValue, ImmutableBytesWritable, KeyValue> {
172     protected void reduce(
173         KeyValueWritableComparable row,
174         Iterable<KeyValue> kvs,
175         Reducer<KeyValueWritableComparable,
176           KeyValue, ImmutableBytesWritable, KeyValue>.Context context)
177         throws java.io.IOException, InterruptedException {
178       int index = 0;
179       for (KeyValue kv : kvs) {
180         context.write(new ImmutableBytesWritable(kv.getRowArray()), kv);
181         if (++index % 100 == 0)
182           context.setStatus("Wrote " + index + " KeyValues, "
183               + "and the rowkey whose is being wrote is " + Bytes.toString(kv.getRowArray())); 
184       }
185     }
186   }
187   
188   public static class KeyValueSortImporter 
189       extends TableMapper<KeyValueWritableComparable, KeyValue> {
190     private Map<byte[], byte[]> cfRenameMap;
191     private Filter filter;
192     private static final Log LOG = LogFactory.getLog(KeyValueImporter.class);
193     
194     /**
195      * @param row  The current table row key.
196      * @param value  The columns.
197      * @param context  The current context.
198      * @throws IOException When something is broken with the data.
199      */
200     @Override
201     public void map(ImmutableBytesWritable row, Result value, Context context) 
202         throws IOException {
203       try {
204         if (LOG.isTraceEnabled()) {
205           LOG.trace("Considering the row."
206               + Bytes.toString(row.get(), row.getOffset(), row.getLength()));
207         }
208         if (filter == null || !filter.filterRowKey(row.get(), row.getOffset(), row.getLength())) {
209           for (Cell kv : value.rawCells()) {
210             kv = filterKv(filter, kv);
211             // skip if we filtered it out
212             if (kv == null) continue;
213             // TODO get rid of ensureKeyValue
214             KeyValue ret = KeyValueUtil.ensureKeyValueTypeForMR(convertKv(kv, cfRenameMap));
215             context.write(new KeyValueWritableComparable(ret.createKeyOnly(false)), ret); 
216           }
217         }
218       } catch (InterruptedException e) {
219         e.printStackTrace();
220       }
221     }
222     
223     @Override
224     public void setup(Context context) throws IOException { 
225       cfRenameMap = createCfRenameMap(context.getConfiguration());
226       filter = instantiateFilter(context.getConfiguration());
227       int reduceNum = context.getNumReduceTasks();
228       Configuration conf = context.getConfiguration();
229       TableName tableName = TableName.valueOf(context.getConfiguration().get(TABLE_NAME));
230       try (Connection conn = ConnectionFactory.createConnection(conf); 
231           RegionLocator regionLocator = conn.getRegionLocator(tableName)) {
232         byte[][] startKeys = regionLocator.getStartKeys();
233         if (startKeys.length != reduceNum) {
234           throw new IOException("Region split after job initialization");
235         }
236         KeyValueWritableComparable[] startKeyWraps = 
237             new KeyValueWritableComparable[startKeys.length - 1];
238         for (int i = 1; i < startKeys.length; ++i) {
239           startKeyWraps[i - 1] = 
240               new KeyValueWritableComparable(KeyValueUtil.createFirstOnRow(startKeys[i]));
241         }
242         KeyValueWritableComparablePartitioner.START_KEYS = startKeyWraps;
243       }
244     }
245   }
246 
247   
248   /**
249    * A mapper that just writes out KeyValues.
250    */
251   @edu.umd.cs.findbugs.annotations.SuppressWarnings(value="EQ_COMPARETO_USE_OBJECT_EQUALS",
252       justification="Writables are going away and this has been this way forever")
253   public static class KeyValueImporter extends TableMapper<ImmutableBytesWritable, KeyValue> {
254     private Map<byte[], byte[]> cfRenameMap;
255     private Filter filter;
256     private static final Log LOG = LogFactory.getLog(KeyValueImporter.class);
257 
258     /**
259      * @param row  The current table row key.
260      * @param value  The columns.
261      * @param context  The current context.
262      * @throws IOException When something is broken with the data.
263      */
264     @Override
265     public void map(ImmutableBytesWritable row, Result value,
266       Context context)
267     throws IOException {
268       try {
269         if (LOG.isTraceEnabled()) {
270           LOG.trace("Considering the row."
271               + Bytes.toString(row.get(), row.getOffset(), row.getLength()));
272         }
273         if (filter == null || !filter.filterRowKey(row.get(), row.getOffset(), row.getLength())) {
274           for (Cell kv : value.rawCells()) {
275             kv = filterKv(filter, kv);
276             // skip if we filtered it out
277             if (kv == null) continue;
278             // TODO get rid of ensureKeyValue
279             context.write(row, KeyValueUtil.ensureKeyValueTypeForMR(convertKv(kv, cfRenameMap)));
280           }
281         }
282       } catch (InterruptedException e) {
283         e.printStackTrace();
284       }
285     }
286 
287     @Override
288     public void setup(Context context) {
289       cfRenameMap = createCfRenameMap(context.getConfiguration());
290       filter = instantiateFilter(context.getConfiguration());
291     }
292   }
293 
294   /**
295    * Write table content out to files in hdfs.
296    */
297   public static class Importer extends TableMapper<ImmutableBytesWritable, Mutation> {
298     private Map<byte[], byte[]> cfRenameMap;
299     private List<UUID> clusterIds;
300     private Filter filter;
301     private Durability durability;
302 
303     /**
304      * @param row  The current table row key.
305      * @param value  The columns.
306      * @param context  The current context.
307      * @throws IOException When something is broken with the data.
308      */
309     @Override
310     public void map(ImmutableBytesWritable row, Result value,
311       Context context)
312     throws IOException {
313       try {
314         writeResult(row, value, context);
315       } catch (InterruptedException e) {
316         e.printStackTrace();
317       }
318     }
319 
320     private void writeResult(ImmutableBytesWritable key, Result result, Context context)
321     throws IOException, InterruptedException {
322       Put put = null;
323       Delete delete = null;
324       if (LOG.isTraceEnabled()) {
325         LOG.trace("Considering the row."
326             + Bytes.toString(key.get(), key.getOffset(), key.getLength()));
327       }
328       if (filter == null || !filter.filterRowKey(key.get(), key.getOffset(), key.getLength())) {
329         processKV(key, result, context, put, delete);
330       }
331     }
332 
333     protected void processKV(ImmutableBytesWritable key, Result result, Context context, Put put,
334         Delete delete) throws IOException, InterruptedException {
335       for (Cell kv : result.rawCells()) {
336         kv = filterKv(filter, kv);
337         // skip if we filter it out
338         if (kv == null) continue;
339 
340         kv = convertKv(kv, cfRenameMap);
341         // Deletes and Puts are gathered and written when finished
342         /*
343          * If there are sequence of mutations and tombstones in an Export, and after Import the same
344          * sequence should be restored as it is. If we combine all Delete tombstones into single
345          * request then there is chance of ignoring few DeleteFamily tombstones, because if we
346          * submit multiple DeleteFamily tombstones in single Delete request then we are maintaining
347          * only newest in hbase table and ignoring other. Check - HBASE-12065
348          */
349         if (CellUtil.isDeleteFamily(kv)) {
350           Delete deleteFamily = new Delete(key.get());
351           deleteFamily.addDeleteMarker(kv);
352           if (durability != null) {
353             deleteFamily.setDurability(durability);
354           }
355           deleteFamily.setClusterIds(clusterIds);
356           context.write(key, deleteFamily);
357         } else if (CellUtil.isDelete(kv)) {
358           if (delete == null) {
359             delete = new Delete(key.get());
360           }
361           delete.addDeleteMarker(kv);
362         } else {
363           if (put == null) {
364             put = new Put(key.get());
365           }
366           addPutToKv(put, kv);
367         }
368       }
369       if (put != null) {
370         if (durability != null) {
371           put.setDurability(durability);
372         }
373         put.setClusterIds(clusterIds);
374         context.write(key, put);
375       }
376       if (delete != null) {
377         if (durability != null) {
378           delete.setDurability(durability);
379         }
380         delete.setClusterIds(clusterIds);
381         context.write(key, delete);
382       }
383     }
384 
385     protected void addPutToKv(Put put, Cell kv) throws IOException {
386       put.add(kv);
387     }
388 
389     @Override
390     public void setup(Context context) {
391       Configuration conf = context.getConfiguration();
392       cfRenameMap = createCfRenameMap(conf);
393       filter = instantiateFilter(conf);
394       String durabilityStr = conf.get(WAL_DURABILITY);
395       if(durabilityStr != null){
396         durability = Durability.valueOf(durabilityStr.toUpperCase(Locale.ROOT));
397       }
398       // TODO: This is kind of ugly doing setup of ZKW just to read the clusterid.
399       ZooKeeperWatcher zkw = null;
400       Exception ex = null;
401       try {
402         zkw = new ZooKeeperWatcher(conf, context.getTaskAttemptID().toString(), null);
403         clusterIds = Collections.singletonList(ZKClusterId.getUUIDForCluster(zkw));
404       } catch (ZooKeeperConnectionException e) {
405         ex = e;
406         LOG.error("Problem connecting to ZooKeper during task setup", e);
407       } catch (KeeperException e) {
408         ex = e;
409         LOG.error("Problem reading ZooKeeper data during task setup", e);
410       } catch (IOException e) {
411         ex = e;
412         LOG.error("Problem setting up task", e);
413       } finally {
414         if (zkw != null) zkw.close();
415       }
416       if (clusterIds == null) {
417         // exit early if setup fails
418         throw new RuntimeException(ex);
419       }
420     }
421   }
422 
423   /**
424    * Create a {@link Filter} to apply to all incoming keys ({@link KeyValue KeyValues}) to
425    * optionally not include in the job output
426    * @param conf {@link Configuration} from which to load the filter
427    * @return the filter to use for the task, or <tt>null</tt> if no filter to should be used
428    * @throws IllegalArgumentException if the filter is misconfigured
429    */
430   public static Filter instantiateFilter(Configuration conf) {
431     // get the filter, if it was configured    
432     Class<? extends Filter> filterClass = conf.getClass(FILTER_CLASS_CONF_KEY, null, Filter.class);
433     if (filterClass == null) {
434       LOG.debug("No configured filter class, accepting all keyvalues.");
435       return null;
436     }
437     LOG.debug("Attempting to create filter:" + filterClass);
438     String[] filterArgs = conf.getStrings(FILTER_ARGS_CONF_KEY);
439     ArrayList<byte[]> quotedArgs = toQuotedByteArrays(filterArgs);
440     try {
441       Method m = filterClass.getMethod("createFilterFromArguments", ArrayList.class);
442       return (Filter) m.invoke(null, quotedArgs);
443     } catch (IllegalAccessException e) {
444       LOG.error("Couldn't instantiate filter!", e);
445       throw new RuntimeException(e);
446     } catch (SecurityException e) {
447       LOG.error("Couldn't instantiate filter!", e);
448       throw new RuntimeException(e);
449     } catch (NoSuchMethodException e) {
450       LOG.error("Couldn't instantiate filter!", e);
451       throw new RuntimeException(e);
452     } catch (IllegalArgumentException e) {
453       LOG.error("Couldn't instantiate filter!", e);
454       throw new RuntimeException(e);
455     } catch (InvocationTargetException e) {
456       LOG.error("Couldn't instantiate filter!", e);
457       throw new RuntimeException(e);
458     }
459   }
460 
461   private static ArrayList<byte[]> toQuotedByteArrays(String... stringArgs) {
462     ArrayList<byte[]> quotedArgs = new ArrayList<byte[]>();
463     for (String stringArg : stringArgs) {
464       // all the filters' instantiation methods expected quoted args since they are coming from
465       // the shell, so add them here, though it shouldn't really be needed :-/
466       quotedArgs.add(Bytes.toBytes("'" + stringArg + "'"));
467     }
468     return quotedArgs;
469   }
470 
471   /**
472    * Attempt to filter out the keyvalue
473    * @param kv {@link KeyValue} on which to apply the filter
474    * @return <tt>null</tt> if the key should not be written, otherwise returns the original
475    *         {@link KeyValue}
476    */
477   public static Cell filterKv(Filter filter, Cell kv) throws IOException {
478     // apply the filter and skip this kv if the filter doesn't apply
479     if (filter != null) {
480       Filter.ReturnCode code = filter.filterKeyValue(kv);
481       if (LOG.isTraceEnabled()) {
482         LOG.trace("Filter returned:" + code + " for the key value:" + kv);
483       }
484       // if its not an accept type, then skip this kv
485       if (!(code.equals(Filter.ReturnCode.INCLUDE) || code
486           .equals(Filter.ReturnCode.INCLUDE_AND_NEXT_COL))) {
487         return null;
488       }
489     }
490     return kv;
491   }
492 
493   // helper: create a new KeyValue based on CF rename map
494   private static Cell convertKv(Cell kv, Map<byte[], byte[]> cfRenameMap) {
495     if(cfRenameMap != null) {
496       // If there's a rename mapping for this CF, create a new KeyValue
497       byte[] newCfName = cfRenameMap.get(CellUtil.cloneFamily(kv));
498       if(newCfName != null) {
499         List<Tag> tags = null;
500         if (kv.getTagsLength() > 0) {
501           tags = Tag.asList(kv.getTagsArray(), kv.getTagsOffset(), kv.getTagsLength());
502         }
503           kv = new KeyValue(kv.getRowArray(), // row buffer 
504                   kv.getRowOffset(),        // row offset
505                   kv.getRowLength(),        // row length
506                   newCfName,                // CF buffer
507                   0,                        // CF offset 
508                   newCfName.length,         // CF length 
509                   kv.getQualifierArray(),   // qualifier buffer
510                   kv.getQualifierOffset(),  // qualifier offset
511                   kv.getQualifierLength(),  // qualifier length
512                   kv.getTimestamp(),        // timestamp
513                   KeyValue.Type.codeToType(kv.getTypeByte()), // KV Type
514                   kv.getValueArray(),       // value buffer 
515                   kv.getValueOffset(),      // value offset
516                   kv.getValueLength(),      // value length
517                   tags);
518       }
519     }
520     return kv;
521   }
522 
523   // helper: make a map from sourceCfName to destCfName by parsing a config key
524   private static Map<byte[], byte[]> createCfRenameMap(Configuration conf) {
525     Map<byte[], byte[]> cfRenameMap = null;
526     String allMappingsPropVal = conf.get(CF_RENAME_PROP);
527     if(allMappingsPropVal != null) {
528       // The conf value format should be sourceCf1:destCf1,sourceCf2:destCf2,...
529       String[] allMappings = allMappingsPropVal.split(",");
530       for (String mapping: allMappings) {
531         if(cfRenameMap == null) {
532             cfRenameMap = new TreeMap<byte[],byte[]>(Bytes.BYTES_COMPARATOR);
533         }
534         String [] srcAndDest = mapping.split(":");
535         if(srcAndDest.length != 2) {
536             continue;
537         }
538         cfRenameMap.put(srcAndDest[0].getBytes(), srcAndDest[1].getBytes());
539       }
540     }
541     return cfRenameMap;
542   }
543 
544   /**
545    * <p>Sets a configuration property with key {@link #CF_RENAME_PROP} in conf that tells
546    * the mapper how to rename column families.
547    * 
548    * <p>Alternately, instead of calling this function, you could set the configuration key 
549    * {@link #CF_RENAME_PROP} yourself. The value should look like 
550    * <pre>srcCf1:destCf1,srcCf2:destCf2,....</pre>. This would have the same effect on
551    * the mapper behavior.
552    * 
553    * @param conf the Configuration in which the {@link #CF_RENAME_PROP} key will be
554    *  set
555    * @param renameMap a mapping from source CF names to destination CF names
556    */
557   static public void configureCfRenaming(Configuration conf, 
558           Map<String, String> renameMap) {
559     StringBuilder sb = new StringBuilder();
560     for(Map.Entry<String,String> entry: renameMap.entrySet()) {
561       String sourceCf = entry.getKey();
562       String destCf = entry.getValue();
563 
564       if(sourceCf.contains(":") || sourceCf.contains(",") || 
565               destCf.contains(":") || destCf.contains(",")) {
566         throw new IllegalArgumentException("Illegal character in CF names: " 
567               + sourceCf + ", " + destCf);
568       }
569 
570       if(sb.length() != 0) {
571         sb.append(",");
572       }
573       sb.append(sourceCf + ":" + destCf);
574     }
575     conf.set(CF_RENAME_PROP, sb.toString());
576   }
577 
578   /**
579    * Add a Filter to be instantiated on import
580    * @param conf Configuration to update (will be passed to the job)
581    * @param clazz {@link Filter} subclass to instantiate on the server.
582    * @param filterArgs List of arguments to pass to the filter on instantiation
583    */
584   public static void addFilterAndArguments(Configuration conf, Class<? extends Filter> clazz,
585       List<String> filterArgs) throws IOException {
586     conf.set(Import.FILTER_CLASS_CONF_KEY, clazz.getName());
587     conf.setStrings(Import.FILTER_ARGS_CONF_KEY, filterArgs.toArray(new String[filterArgs.size()]));
588   }
589 
590   /**
591    * Sets up the actual job.
592    * @param conf The current configuration.
593    * @param args The command line parameters.
594    * @return The newly created job.
595    * @throws IOException When setting up the job fails.
596    */
597   public static Job createSubmittableJob(Configuration conf, String[] args)
598   throws IOException {
599     TableName tableName = TableName.valueOf(args[0]);
600     conf.set(TABLE_NAME, tableName.getNameAsString());
601     Path inputDir = new Path(args[1]);
602     Job job = new Job(conf, NAME + "_" + tableName);
603     job.setJarByClass(Importer.class);
604     FileInputFormat.setInputPaths(job, inputDir);
605     job.setInputFormatClass(SequenceFileInputFormat.class);
606     String hfileOutPath = conf.get(BULK_OUTPUT_CONF_KEY);
607 
608     // make sure we get the filter in the jars
609     try {
610       Class<? extends Filter> filter = conf.getClass(FILTER_CLASS_CONF_KEY, null, Filter.class);
611       if (filter != null) {
612         TableMapReduceUtil.addDependencyJarsForClasses(conf, filter);
613       }
614     } catch (Exception e) {
615       throw new IOException(e);
616     }
617 
618     if (hfileOutPath != null && conf.getBoolean(HAS_LARGE_RESULT, false)) {
619       LOG.info("Use Large Result!!");
620       try (Connection conn = ConnectionFactory.createConnection(conf); 
621           Table table = conn.getTable(tableName);
622           RegionLocator regionLocator = conn.getRegionLocator(tableName)) {
623         HFileOutputFormat2.configureIncrementalLoad(job, table.getTableDescriptor(), regionLocator);
624         job.setMapperClass(KeyValueSortImporter.class);
625         job.setReducerClass(KeyValueReducer.class);
626         Path outputDir = new Path(hfileOutPath);
627         FileOutputFormat.setOutputPath(job, outputDir);
628         job.setMapOutputKeyClass(KeyValueWritableComparable.class);
629         job.setMapOutputValueClass(KeyValue.class);
630         job.getConfiguration().setClass("mapreduce.job.output.key.comparator.class", 
631             KeyValueWritableComparable.KeyValueWritableComparator.class,
632             RawComparator.class);
633         Path partitionsPath = 
634             new Path(TotalOrderPartitioner.getPartitionFile(job.getConfiguration()));
635         FileSystem fs = FileSystem.get(job.getConfiguration());
636         fs.deleteOnExit(partitionsPath);
637         job.setPartitionerClass(KeyValueWritableComparablePartitioner.class);
638         job.setNumReduceTasks(regionLocator.getStartKeys().length);
639         TableMapReduceUtil.addDependencyJarsForClasses(job.getConfiguration(),
640             com.google.common.base.Preconditions.class);
641       }
642     } else if (hfileOutPath != null) {
643       job.setMapperClass(KeyValueImporter.class);
644       try (Connection conn = ConnectionFactory.createConnection(conf); 
645           Table table = conn.getTable(tableName);
646           RegionLocator regionLocator = conn.getRegionLocator(tableName)){
647         job.setReducerClass(KeyValueSortReducer.class);
648         Path outputDir = new Path(hfileOutPath);
649         FileOutputFormat.setOutputPath(job, outputDir);
650         job.setMapOutputKeyClass(ImmutableBytesWritable.class);
651         job.setMapOutputValueClass(KeyValue.class);
652         HFileOutputFormat2.configureIncrementalLoad(job, table.getTableDescriptor(), regionLocator);
653         TableMapReduceUtil.addDependencyJarsForClasses(job.getConfiguration(),
654             com.google.common.base.Preconditions.class);
655       }
656     } else {
657       // No reducers.  Just write straight to table.  Call initTableReducerJob
658       // because it sets up the TableOutputFormat.
659       job.setMapperClass(Importer.class);
660       TableMapReduceUtil.initTableReducerJob(tableName.getNameAsString(), null, job);
661       job.setNumReduceTasks(0);
662     }
663     return job;
664   }
665 
666   /*
667    * @param errorMsg Error message.  Can be null.
668    */
669   private static void usage(final String errorMsg) {
670     if (errorMsg != null && errorMsg.length() > 0) {
671       System.err.println("ERROR: " + errorMsg);
672     }
673     System.err.println("Usage: Import [options] <tablename> <inputdir>");
674     System.err.println("By default Import will load data directly into HBase. To instead generate");
675     System.err.println("HFiles of data to prepare for a bulk data load, pass the option:");
676     System.err.println("  -D" + BULK_OUTPUT_CONF_KEY + "=/path/for/output");
677     System.err.println("If there is a large result that includes too much KeyValue "
678         + "whitch can occur OOME caused by the memery sort in reducer, pass the option:");
679     System.err.println("  -D" + HAS_LARGE_RESULT + "=true");
680     System.err
681         .println(" To apply a generic org.apache.hadoop.hbase.filter.Filter to the input, use");
682     System.err.println("  -D" + FILTER_CLASS_CONF_KEY + "=<name of filter class>");
683     System.err.println("  -D" + FILTER_ARGS_CONF_KEY + "=<comma separated list of args for filter");
684     System.err.println(" NOTE: The filter will be applied BEFORE doing key renames via the "
685         + CF_RENAME_PROP + " property. Futher, filters will only use the"
686         + " Filter#filterRowKey(byte[] buffer, int offset, int length) method to identify "
687         + " whether the current row needs to be ignored completely for processing and "
688         + " Filter#filterKeyValue(KeyValue) method to determine if the KeyValue should be added;"
689         + " Filter.ReturnCode#INCLUDE and #INCLUDE_AND_NEXT_COL will be considered as including"
690         + " the KeyValue.");
691     System.err.println("To import data exported from HBase 0.94, use");
692     System.err.println("  -Dhbase.import.version=0.94");
693     System.err.println("For performance consider the following options:\n"
694         + "  -Dmapreduce.map.speculative=false\n"
695         + "  -Dmapreduce.reduce.speculative=false\n"
696         + "  -D" + WAL_DURABILITY + "=<Used while writing data to hbase."
697             +" Allowed values are the supported durability values"
698             +" like SKIP_WAL/ASYNC_WAL/SYNC_WAL/...>");
699   }
700 
701   /**
702    * If the durability is set to {@link Durability#SKIP_WAL} and the data is imported to hbase, we
703    * need to flush all the regions of the table as the data is held in memory and is also not
704    * present in the Write Ahead Log to replay in scenarios of a crash. This method flushes all the
705    * regions of the table in the scenarios of import data to hbase with {@link Durability#SKIP_WAL}
706    */
707   public static void flushRegionsIfNecessary(Configuration conf) throws IOException,
708       InterruptedException {
709     String tableName = conf.get(TABLE_NAME);
710     HBaseAdmin hAdmin = null;
711     String durability = conf.get(WAL_DURABILITY);
712     // Need to flush if the data is written to hbase and skip wal is enabled.
713     if (conf.get(BULK_OUTPUT_CONF_KEY) == null && durability != null
714         && Durability.SKIP_WAL.name().equalsIgnoreCase(durability)) {
715       try {
716         hAdmin = new HBaseAdmin(conf);
717         hAdmin.flush(tableName);
718       } finally {
719         if (hAdmin != null) {
720           hAdmin.close();
721         }
722       }
723     }
724   }
725 
726   /**
727    * Main entry point.
728    *
729    * @param args  The command line parameters.
730    * @throws Exception When running the job fails.
731    */
732   public static void main(String[] args) throws Exception {
733     Configuration conf = HBaseConfiguration.create();
734     String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs();
735     if (otherArgs.length < 2) {
736       usage("Wrong number of arguments: " + otherArgs.length);
737       System.exit(-1);
738     }
739     String inputVersionString = System.getProperty(ResultSerialization.IMPORT_FORMAT_VER);
740     if (inputVersionString != null) {
741       conf.set(ResultSerialization.IMPORT_FORMAT_VER, inputVersionString);
742     }
743     Job job = createSubmittableJob(conf, otherArgs);
744     boolean isJobSuccessful = job.waitForCompletion(true);
745     if(isJobSuccessful){
746       // Flush all the regions of the table
747       flushRegionsIfNecessary(conf);
748     }
749     long inputRecords = job.getCounters().findCounter(TaskCounter.MAP_INPUT_RECORDS).getValue();
750     long outputRecords = job.getCounters().findCounter(TaskCounter.MAP_OUTPUT_RECORDS).getValue();
751     if (outputRecords < inputRecords) {
752       System.err.println("Warning, not all records were imported (maybe filtered out).");
753       if (outputRecords == 0) {
754         System.err.println("If the data was exported from HBase 0.94 "+
755             "consider using -Dhbase.import.version=0.94.");
756       }
757     }
758 
759     System.exit(isJobSuccessful ? 0 : 1);
760   }
761 }