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.FileNotFoundException;
22  import java.io.IOException;
23  import java.nio.ByteBuffer;
24  import java.util.ArrayList;
25  import java.util.Arrays;
26  import java.util.Collection;
27  import java.util.Deque;
28  import java.util.HashSet;
29  import java.util.LinkedList;
30  import java.util.List;
31  import java.util.Map;
32  import java.util.Map.Entry;
33  import java.util.Set;
34  import java.util.TreeMap;
35  import java.util.concurrent.Callable;
36  import java.util.concurrent.ExecutionException;
37  import java.util.concurrent.ExecutorService;
38  import java.util.concurrent.Future;
39  import java.util.concurrent.LinkedBlockingQueue;
40  import java.util.concurrent.ThreadPoolExecutor;
41  import java.util.concurrent.TimeUnit;
42  import java.util.concurrent.atomic.AtomicLong;
43  
44  import org.apache.commons.logging.Log;
45  import org.apache.commons.logging.LogFactory;
46  import org.apache.hadoop.classification.InterfaceAudience;
47  import org.apache.hadoop.classification.InterfaceStability;
48  import org.apache.hadoop.conf.Configuration;
49  import org.apache.hadoop.conf.Configured;
50  import org.apache.hadoop.fs.FileStatus;
51  import org.apache.hadoop.fs.FileSystem;
52  import org.apache.hadoop.fs.FileUtil;
53  import org.apache.hadoop.fs.Path;
54  import org.apache.hadoop.hbase.HBaseConfiguration;
55  import org.apache.hadoop.hbase.HColumnDescriptor;
56  import org.apache.hadoop.hbase.HConstants;
57  import org.apache.hadoop.hbase.HTableDescriptor;
58  import org.apache.hadoop.hbase.KeyValue;
59  import org.apache.hadoop.hbase.TableName;
60  import org.apache.hadoop.hbase.TableNotFoundException;
61  import org.apache.hadoop.hbase.client.HBaseAdmin;
62  import org.apache.hadoop.hbase.client.HConnection;
63  import org.apache.hadoop.hbase.client.HTable;
64  import org.apache.hadoop.hbase.client.RegionServerCallable;
65  import org.apache.hadoop.hbase.client.RpcRetryingCallerFactory;
66  import org.apache.hadoop.hbase.client.coprocessor.SecureBulkLoadClient;
67  import org.apache.hadoop.hbase.io.HalfStoreFileReader;
68  import org.apache.hadoop.hbase.io.Reference;
69  import org.apache.hadoop.hbase.io.compress.Compression.Algorithm;
70  import org.apache.hadoop.hbase.io.hfile.CacheConfig;
71  import org.apache.hadoop.hbase.io.hfile.HFile;
72  import org.apache.hadoop.hbase.io.hfile.HFileDataBlockEncoder;
73  import org.apache.hadoop.hbase.io.hfile.HFileDataBlockEncoderImpl;
74  import org.apache.hadoop.hbase.io.hfile.HFileScanner;
75  import org.apache.hadoop.hbase.protobuf.ProtobufUtil;
76  import org.apache.hadoop.hbase.regionserver.BloomType;
77  import org.apache.hadoop.hbase.regionserver.HStore;
78  import org.apache.hadoop.hbase.regionserver.StoreFile;
79  import org.apache.hadoop.hbase.security.UserProvider;
80  import org.apache.hadoop.hbase.util.Bytes;
81  import org.apache.hadoop.hbase.util.Pair;
82  import org.apache.hadoop.security.token.Token;
83  import org.apache.hadoop.util.Tool;
84  import org.apache.hadoop.util.ToolRunner;
85  
86  import com.google.common.collect.HashMultimap;
87  import com.google.common.collect.Multimap;
88  import com.google.common.collect.Multimaps;
89  import com.google.common.util.concurrent.ThreadFactoryBuilder;
90  
91  /**
92   * Tool to load the output of HFileOutputFormat into an existing table.
93   * @see #usage()
94   */
95  @InterfaceAudience.Public
96  @InterfaceStability.Stable
97  public class LoadIncrementalHFiles extends Configured implements Tool {
98    private static final Log LOG = LogFactory.getLog(LoadIncrementalHFiles.class);
99    static final AtomicLong regionCount = new AtomicLong(0);
100   private HBaseAdmin hbAdmin;
101   private Configuration cfg;
102 
103   public static final String NAME = "completebulkload";
104   private static final String ASSIGN_SEQ_IDS = "hbase.mapreduce.bulkload.assign.sequenceNumbers";
105   private boolean assignSeqIds;
106 
107   private boolean hasForwardedToken;
108   private Token<?> userToken;
109   private String bulkToken;
110   private UserProvider userProvider;
111 
112   public LoadIncrementalHFiles(Configuration conf) throws Exception {
113     super(conf);
114     this.cfg = conf;
115     this.hbAdmin = new HBaseAdmin(conf);
116     this.userProvider = UserProvider.instantiate(conf);
117     assignSeqIds = conf.getBoolean(ASSIGN_SEQ_IDS, true);
118   }
119 
120   private void usage() {
121     System.err.println("usage: " + NAME +
122         " /path/to/hfileoutputformat-output " +
123         "tablename");
124   }
125 
126   /**
127    * Represents an HFile waiting to be loaded. An queue is used
128    * in this class in order to support the case where a region has
129    * split during the process of the load. When this happens,
130    * the HFile is split into two physical parts across the new
131    * region boundary, and each part is added back into the queue.
132    * The import process finishes when the queue is empty.
133    */
134   static class LoadQueueItem {
135     final byte[] family;
136     final Path hfilePath;
137 
138     public LoadQueueItem(byte[] family, Path hfilePath) {
139       this.family = family;
140       this.hfilePath = hfilePath;
141     }
142 
143     public String toString() {
144       return "family:"+ Bytes.toString(family) + " path:" + hfilePath.toString();
145     }
146   }
147 
148   /**
149    * Walk the given directory for all HFiles, and return a Queue
150    * containing all such files.
151    */
152   private void discoverLoadQueue(Deque<LoadQueueItem> ret, Path hfofDir)
153   throws IOException {
154     FileSystem fs = hfofDir.getFileSystem(getConf());
155 
156     if (!fs.exists(hfofDir)) {
157       throw new FileNotFoundException("HFileOutputFormat dir " +
158           hfofDir + " not found");
159     }
160 
161     FileStatus[] familyDirStatuses = fs.listStatus(hfofDir);
162     if (familyDirStatuses == null) {
163       throw new FileNotFoundException("No families found in " + hfofDir);
164     }
165 
166     for (FileStatus stat : familyDirStatuses) {
167       if (!stat.isDir()) {
168         LOG.warn("Skipping non-directory " + stat.getPath());
169         continue;
170       }
171       Path familyDir = stat.getPath();
172       // Skip _logs, etc
173       if (familyDir.getName().startsWith("_")) continue;
174       byte[] family = familyDir.getName().getBytes();
175       Path[] hfiles = FileUtil.stat2Paths(fs.listStatus(familyDir));
176       for (Path hfile : hfiles) {
177         if (hfile.getName().startsWith("_")) continue;
178         ret.add(new LoadQueueItem(family, hfile));
179       }
180     }
181   }
182 
183   /**
184    * Perform a bulk load of the given directory into the given
185    * pre-existing table.  This method is not threadsafe.
186    *
187    * @param hfofDir the directory that was provided as the output path
188    * of a job using HFileOutputFormat
189    * @param table the table to load into
190    * @throws TableNotFoundException if table does not yet exist
191    */
192   @SuppressWarnings("deprecation")
193   public void doBulkLoad(Path hfofDir, final HTable table)
194     throws TableNotFoundException, IOException
195   {
196     final HConnection conn = table.getConnection();
197 
198     if (!conn.isTableAvailable(table.getName())) {
199       throw new TableNotFoundException("Table " +
200           Bytes.toStringBinary(table.getTableName()) +
201           "is not currently available.");
202     }
203 
204     // initialize thread pools
205     int nrThreads = cfg.getInt("hbase.loadincremental.threads.max",
206         Runtime.getRuntime().availableProcessors());
207     ThreadFactoryBuilder builder = new ThreadFactoryBuilder();
208     builder.setNameFormat("LoadIncrementalHFiles-%1$d");
209     ExecutorService pool = new ThreadPoolExecutor(nrThreads, nrThreads,
210         60, TimeUnit.SECONDS,
211         new LinkedBlockingQueue<Runnable>(),
212         builder.build());
213     ((ThreadPoolExecutor)pool).allowCoreThreadTimeOut(true);
214 
215     // LQI queue does not need to be threadsafe -- all operations on this queue
216     // happen in this thread
217     Deque<LoadQueueItem> queue = new LinkedList<LoadQueueItem>();
218     try {
219       discoverLoadQueue(queue, hfofDir);
220       // check whether there is invalid family name in HFiles to be bulkloaded
221       Collection<HColumnDescriptor> families = table.getTableDescriptor().getFamilies();
222       ArrayList<String> familyNames = new ArrayList<String>();
223       for (HColumnDescriptor family : families) {
224         familyNames.add(family.getNameAsString());
225       }
226       ArrayList<String> unmatchedFamilies = new ArrayList<String>();
227       for (LoadQueueItem lqi : queue) {
228         String familyNameInHFile = Bytes.toString(lqi.family);
229         if (!familyNames.contains(familyNameInHFile)) {
230           unmatchedFamilies.add(familyNameInHFile);
231         }
232       }
233       if (unmatchedFamilies.size() > 0) {
234         String msg =
235             "Unmatched family names found: unmatched family names in HFiles to be bulkloaded: "
236                 + unmatchedFamilies + "; valid family names of table "
237                 + Bytes.toString(table.getTableName()) + " are: " + familyNames;
238         LOG.error(msg);
239         throw new IOException(msg);
240       }
241       int count = 0;
242 
243       if (queue.isEmpty()) {
244         LOG.warn("Bulk load operation did not find any files to load in " +
245             "directory " + hfofDir.toUri() + ".  Does it contain files in " +
246             "subdirectories that correspond to column family names?");
247         return;
248       }
249 
250       //If using secure bulk load
251       //prepare staging directory and token
252       if (userProvider.isHBaseSecurityEnabled()) {
253         FileSystem fs = FileSystem.get(cfg);
254         //This condition is here for unit testing
255         //Since delegation token doesn't work in mini cluster
256         if (userProvider.isHadoopSecurityEnabled()) {
257           userToken = userProvider.getCurrent().getToken("HDFS_DELEGATION_TOKEN",
258                                                          fs.getCanonicalServiceName());
259           if (userToken == null) {
260             hasForwardedToken = false;
261             userToken = fs.getDelegationToken("renewer");
262           } else {
263             hasForwardedToken = true;
264             LOG.info("Use the existing token: " + userToken);
265           }
266         }
267         bulkToken = new SecureBulkLoadClient(table).prepareBulkLoad(table.getName());
268       }
269 
270       // Assumes that region splits can happen while this occurs.
271       while (!queue.isEmpty()) {
272         // need to reload split keys each iteration.
273         final Pair<byte[][], byte[][]> startEndKeys = table.getStartEndKeys();
274         if (count != 0) {
275           LOG.info("Split occured while grouping HFiles, retry attempt " +
276               + count + " with " + queue.size() + " files remaining to group or split");
277         }
278 
279         int maxRetries = cfg.getInt("hbase.bulkload.retries.number", 0);
280         if (maxRetries != 0 && count >= maxRetries) {
281           LOG.error("Retry attempted " + count +  " times without completing, bailing out");
282           return;
283         }
284         count++;
285 
286         // Using ByteBuffer for byte[] equality semantics
287         Multimap<ByteBuffer, LoadQueueItem> regionGroups = groupOrSplitPhase(table,
288             pool, queue, startEndKeys);
289 
290         bulkLoadPhase(table, conn, pool, queue, regionGroups);
291 
292         // NOTE: The next iteration's split / group could happen in parallel to
293         // atomic bulkloads assuming that there are splits and no merges, and
294         // that we can atomically pull out the groups we want to retry.
295       }
296 
297     } finally {
298       if (userProvider.isHBaseSecurityEnabled()) {
299         if (userToken != null && !hasForwardedToken) {
300           try {
301             userToken.cancel(cfg);
302           } catch (Exception e) {
303             LOG.warn("Failed to cancel HDFS delegation token.", e);
304           }
305         }
306         if(bulkToken != null) {
307           new SecureBulkLoadClient(table).cleanupBulkLoad(bulkToken);
308         }
309       }
310       pool.shutdown();
311       if (queue != null && !queue.isEmpty()) {
312         StringBuilder err = new StringBuilder();
313         err.append("-------------------------------------------------\n");
314         err.append("Bulk load aborted with some files not yet loaded:\n");
315         err.append("-------------------------------------------------\n");
316         for (LoadQueueItem q : queue) {
317           err.append("  ").append(q.hfilePath).append('\n');
318         }
319         LOG.error(err);
320       }
321     }
322     
323     if (queue != null && !queue.isEmpty()) {
324         throw new RuntimeException("Bulk load aborted with some files not yet loaded."
325           + "Please check log for more details.");
326     }
327   }
328 
329   /**
330    * This takes the LQI's grouped by likely regions and attempts to bulk load
331    * them.  Any failures are re-queued for another pass with the
332    * groupOrSplitPhase.
333    */
334   protected void bulkLoadPhase(final HTable table, final HConnection conn,
335       ExecutorService pool, Deque<LoadQueueItem> queue,
336       final Multimap<ByteBuffer, LoadQueueItem> regionGroups) throws IOException {
337     // atomically bulk load the groups.
338     Set<Future<List<LoadQueueItem>>> loadingFutures = new HashSet<Future<List<LoadQueueItem>>>();
339     for (Entry<ByteBuffer, ? extends Collection<LoadQueueItem>> e: regionGroups.asMap().entrySet()) {
340       final byte[] first = e.getKey().array();
341       final Collection<LoadQueueItem> lqis =  e.getValue();
342 
343       final Callable<List<LoadQueueItem>> call = new Callable<List<LoadQueueItem>>() {
344         public List<LoadQueueItem> call() throws Exception {
345           List<LoadQueueItem> toRetry =
346               tryAtomicRegionLoad(conn, table.getName(), first, lqis);
347           return toRetry;
348         }
349       };
350       loadingFutures.add(pool.submit(call));
351     }
352 
353     // get all the results.
354     for (Future<List<LoadQueueItem>> future : loadingFutures) {
355       try {
356         List<LoadQueueItem> toRetry = future.get();
357 
358         // LQIs that are requeued to be regrouped.
359         queue.addAll(toRetry);
360 
361       } catch (ExecutionException e1) {
362         Throwable t = e1.getCause();
363         if (t instanceof IOException) {
364           // At this point something unrecoverable has happened.
365           // TODO Implement bulk load recovery
366           throw new IOException("BulkLoad encountered an unrecoverable problem", t);
367         }
368         LOG.error("Unexpected execution exception during bulk load", e1);
369         throw new IllegalStateException(t);
370       } catch (InterruptedException e1) {
371         LOG.error("Unexpected interrupted exception during bulk load", e1);
372         throw new IllegalStateException(e1);
373       }
374     }
375   }
376 
377   /**
378    * @return A Multimap<startkey, LoadQueueItem> that groups LQI by likely
379    * bulk load region targets.
380    */
381   private Multimap<ByteBuffer, LoadQueueItem> groupOrSplitPhase(final HTable table,
382       ExecutorService pool, Deque<LoadQueueItem> queue,
383       final Pair<byte[][], byte[][]> startEndKeys) throws IOException {
384     // <region start key, LQI> need synchronized only within this scope of this
385     // phase because of the puts that happen in futures.
386     Multimap<ByteBuffer, LoadQueueItem> rgs = HashMultimap.create();
387     final Multimap<ByteBuffer, LoadQueueItem> regionGroups = Multimaps.synchronizedMultimap(rgs);
388 
389     // drain LQIs and figure out bulk load groups
390     Set<Future<List<LoadQueueItem>>> splittingFutures = new HashSet<Future<List<LoadQueueItem>>>();
391     while (!queue.isEmpty()) {
392       final LoadQueueItem item = queue.remove();
393 
394       final Callable<List<LoadQueueItem>> call = new Callable<List<LoadQueueItem>>() {
395         public List<LoadQueueItem> call() throws Exception {
396           List<LoadQueueItem> splits = groupOrSplit(regionGroups, item, table, startEndKeys);
397           return splits;
398         }
399       };
400       splittingFutures.add(pool.submit(call));
401     }
402     // get all the results.  All grouping and splitting must finish before
403     // we can attempt the atomic loads.
404     for (Future<List<LoadQueueItem>> lqis : splittingFutures) {
405       try {
406         List<LoadQueueItem> splits = lqis.get();
407         if (splits != null) {
408           queue.addAll(splits);
409         }
410       } catch (ExecutionException e1) {
411         Throwable t = e1.getCause();
412         if (t instanceof IOException) {
413           LOG.error("IOException during splitting", e1);
414           throw (IOException)t; // would have been thrown if not parallelized,
415         }
416         LOG.error("Unexpected execution exception during splitting", e1);
417         throw new IllegalStateException(t);
418       } catch (InterruptedException e1) {
419         LOG.error("Unexpected interrupted exception during splitting", e1);
420         throw new IllegalStateException(e1);
421       }
422     }
423     return regionGroups;
424   }
425 
426   // unique file name for the table
427   String getUniqueName(TableName tableName) {
428     String name = tableName + "," + regionCount.incrementAndGet();
429     return name;
430   }
431 
432   protected List<LoadQueueItem> splitStoreFile(final LoadQueueItem item,
433       final HTable table, byte[] startKey,
434       byte[] splitKey) throws IOException {
435     final Path hfilePath = item.hfilePath;
436 
437     // We use a '_' prefix which is ignored when walking directory trees
438     // above.
439     final Path tmpDir = new Path(item.hfilePath.getParent(), "_tmp");
440 
441     LOG.info("HFile at " + hfilePath + " no longer fits inside a single " +
442     "region. Splitting...");
443 
444     String uniqueName = getUniqueName(table.getName());
445     HColumnDescriptor familyDesc = table.getTableDescriptor().getFamily(item.family);
446     Path botOut = new Path(tmpDir, uniqueName + ".bottom");
447     Path topOut = new Path(tmpDir, uniqueName + ".top");
448     splitStoreFile(getConf(), hfilePath, familyDesc, splitKey,
449         botOut, topOut);
450 
451     // Add these back at the *front* of the queue, so there's a lower
452     // chance that the region will just split again before we get there.
453     List<LoadQueueItem> lqis = new ArrayList<LoadQueueItem>(2);
454     lqis.add(new LoadQueueItem(item.family, botOut));
455     lqis.add(new LoadQueueItem(item.family, topOut));
456 
457     LOG.info("Successfully split into new HFiles " + botOut + " and " + topOut);
458     return lqis;
459   }
460 
461   /**
462    * Attempt to assign the given load queue item into its target region group.
463    * If the hfile boundary no longer fits into a region, physically splits
464    * the hfile such that the new bottom half will fit and returns the list of
465    * LQI's corresponding to the resultant hfiles.
466    *
467    * protected for testing
468    */
469   protected List<LoadQueueItem> groupOrSplit(Multimap<ByteBuffer, LoadQueueItem> regionGroups,
470       final LoadQueueItem item, final HTable table,
471       final Pair<byte[][], byte[][]> startEndKeys)
472       throws IOException {
473     final Path hfilePath = item.hfilePath;
474     final FileSystem fs = hfilePath.getFileSystem(getConf());
475     HFile.Reader hfr = HFile.createReader(fs, hfilePath,
476         new CacheConfig(getConf()));
477     final byte[] first, last;
478     try {
479       hfr.loadFileInfo();
480       first = hfr.getFirstRowKey();
481       last = hfr.getLastRowKey();
482     }  finally {
483       hfr.close();
484     }
485 
486     LOG.info("Trying to load hfile=" + hfilePath +
487         " first=" + Bytes.toStringBinary(first) +
488         " last="  + Bytes.toStringBinary(last));
489     if (first == null || last == null) {
490       assert first == null && last == null;
491       // TODO what if this is due to a bad HFile?
492       LOG.info("hfile " + hfilePath + " has no entries, skipping");
493       return null;
494     }
495     if (Bytes.compareTo(first, last) > 0) {
496       throw new IllegalArgumentException(
497       "Invalid range: " + Bytes.toStringBinary(first) +
498       " > " + Bytes.toStringBinary(last));
499     }
500     int idx = Arrays.binarySearch(startEndKeys.getFirst(), first,
501         Bytes.BYTES_COMPARATOR);
502     if (idx < 0) {
503       // not on boundary, returns -(insertion index).  Calculate region it
504       // would be in.
505       idx = -(idx + 1) - 1;
506     }
507     final int indexForCallable = idx;
508     boolean lastKeyInRange =
509       Bytes.compareTo(last, startEndKeys.getSecond()[idx]) < 0 ||
510       Bytes.equals(startEndKeys.getSecond()[idx], HConstants.EMPTY_BYTE_ARRAY);
511     if (!lastKeyInRange) {
512       List<LoadQueueItem> lqis = splitStoreFile(item, table,
513           startEndKeys.getFirst()[indexForCallable],
514           startEndKeys.getSecond()[indexForCallable]);
515       return lqis;
516     }
517 
518     // group regions.
519     regionGroups.put(ByteBuffer.wrap(startEndKeys.getFirst()[idx]), item);
520     return null;
521   }
522 
523   /**
524    * @deprecated Use {@link #tryAtomicRegionLoad(HConnection, TableName, byte[], Collection)}
525    */
526   @Deprecated
527   protected List<LoadQueueItem> tryAtomicRegionLoad(final HConnection conn,
528       final byte [] tableName, final byte[] first, Collection<LoadQueueItem> lqis)
529   throws IOException {
530     return tryAtomicRegionLoad(conn, TableName.valueOf(tableName), first, lqis);
531   }
532 
533   /**
534    * Attempts to do an atomic load of many hfiles into a region.  If it fails,
535    * it returns a list of hfiles that need to be retried.  If it is successful
536    * it will return an empty list.
537    *
538    * NOTE: To maintain row atomicity guarantees, region server callable should
539    * succeed atomically and fails atomically.
540    *
541    * Protected for testing.
542    *
543    * @return empty list if success, list of items to retry on recoverable
544    * failure
545    */
546   protected List<LoadQueueItem> tryAtomicRegionLoad(final HConnection conn,
547       final TableName tableName, final byte[] first, Collection<LoadQueueItem> lqis)
548   throws IOException {
549     final List<Pair<byte[], String>> famPaths =
550       new ArrayList<Pair<byte[], String>>(lqis.size());
551     for (LoadQueueItem lqi : lqis) {
552       famPaths.add(Pair.newPair(lqi.family, lqi.hfilePath.toString()));
553     }
554 
555     final RegionServerCallable<Boolean> svrCallable =
556         new RegionServerCallable<Boolean>(conn, tableName, first) {
557       @Override
558       public Boolean call() throws Exception {
559         SecureBulkLoadClient secureClient = null;
560         boolean success = false;
561 
562         try {
563           LOG.debug("Going to connect to server " + getLocation() + " for row "
564               + Bytes.toStringBinary(getRow()) + " with hfile group " + famPaths);
565           byte[] regionName = getLocation().getRegionInfo().getRegionName();
566           if(!userProvider.isHBaseSecurityEnabled()) {
567             success = ProtobufUtil.bulkLoadHFile(getStub(), famPaths, regionName, assignSeqIds);
568           } else {
569             HTable table = new HTable(conn.getConfiguration(), getTableName());
570             secureClient = new SecureBulkLoadClient(table);
571             success = secureClient.bulkLoadHFiles(famPaths, userToken, bulkToken,
572               getLocation().getRegionInfo().getStartKey());
573           }
574           return success;
575         } finally {
576           //Best effort copying of files that might not have been imported
577           //from the staging directory back to original location
578           //in user directory
579           if(secureClient != null && !success) {
580             FileSystem fs = FileSystem.get(cfg);
581             for(Pair<byte[], String> el : famPaths) {
582               Path hfileStagingPath = null;
583               Path hfileOrigPath = new Path(el.getSecond());
584               try {
585                 hfileStagingPath= new Path(secureClient.getStagingPath(bulkToken, el.getFirst()),
586                     hfileOrigPath.getName());
587                 if(fs.rename(hfileStagingPath, hfileOrigPath)) {
588                   LOG.debug("Moved back file " + hfileOrigPath + " from " +
589                       hfileStagingPath);
590                 } else if(fs.exists(hfileStagingPath)){
591                   LOG.debug("Unable to move back file " + hfileOrigPath + " from " +
592                       hfileStagingPath);
593                 }
594               } catch(Exception ex) {
595                 LOG.debug("Unable to move back file " + hfileOrigPath + " from " +
596                     hfileStagingPath, ex);
597               }
598             }
599           }
600         }
601       }
602     };
603 
604     try {
605       List<LoadQueueItem> toRetry = new ArrayList<LoadQueueItem>();
606       Configuration conf = getConf();
607       boolean success = RpcRetryingCallerFactory.instantiate(conf).<Boolean> newCaller()
608           .callWithRetries(svrCallable);
609       if (!success) {
610         LOG.warn("Attempt to bulk load region containing "
611             + Bytes.toStringBinary(first) + " into table "
612             + tableName  + " with files " + lqis
613             + " failed.  This is recoverable and they will be retried.");
614         toRetry.addAll(lqis); // return lqi's to retry
615       }
616       // success
617       return toRetry;
618     } catch (IOException e) {
619       LOG.error("Encountered unrecoverable error from region server", e);
620       throw e;
621     }
622   }
623 
624   /**
625    * Split a storefile into a top and bottom half, maintaining
626    * the metadata, recreating bloom filters, etc.
627    */
628   static void splitStoreFile(
629       Configuration conf, Path inFile,
630       HColumnDescriptor familyDesc, byte[] splitKey,
631       Path bottomOut, Path topOut) throws IOException
632   {
633     // Open reader with no block cache, and not in-memory
634     Reference topReference = Reference.createTopReference(splitKey);
635     Reference bottomReference = Reference.createBottomReference(splitKey);
636 
637     copyHFileHalf(conf, inFile, topOut, topReference, familyDesc);
638     copyHFileHalf(conf, inFile, bottomOut, bottomReference, familyDesc);
639   }
640 
641   /**
642    * Copy half of an HFile into a new HFile.
643    */
644   private static void copyHFileHalf(
645       Configuration conf, Path inFile, Path outFile, Reference reference,
646       HColumnDescriptor familyDescriptor)
647   throws IOException {
648     FileSystem fs = inFile.getFileSystem(conf);
649     CacheConfig cacheConf = new CacheConfig(conf);
650     HalfStoreFileReader halfReader = null;
651     StoreFile.Writer halfWriter = null;
652     HFileDataBlockEncoder dataBlockEncoder = new HFileDataBlockEncoderImpl(
653       familyDescriptor.getDataBlockEncoding());
654     try {
655       halfReader = new HalfStoreFileReader(fs, inFile, cacheConf, reference);
656       Map<byte[], byte[]> fileInfo = halfReader.loadFileInfo();
657 
658       int blocksize = familyDescriptor.getBlocksize();
659       Algorithm compression = familyDescriptor.getCompression();
660       BloomType bloomFilterType = familyDescriptor.getBloomFilterType();
661 
662       halfWriter = new StoreFile.WriterBuilder(conf, cacheConf,
663           fs, blocksize)
664               .withFilePath(outFile)
665               .withCompression(compression)
666               .withDataBlockEncoder(dataBlockEncoder)
667               .withBloomType(bloomFilterType)
668               .withChecksumType(HStore.getChecksumType(conf))
669               .withBytesPerChecksum(HStore.getBytesPerChecksum(conf))
670               .build();
671       HFileScanner scanner = halfReader.getScanner(false, false, false);
672       scanner.seekTo();
673       do {
674         KeyValue kv = scanner.getKeyValue();
675         halfWriter.append(kv);
676       } while (scanner.next());
677 
678       for (Map.Entry<byte[],byte[]> entry : fileInfo.entrySet()) {
679         if (shouldCopyHFileMetaKey(entry.getKey())) {
680           halfWriter.appendFileInfo(entry.getKey(), entry.getValue());
681         }
682       }
683     } finally {
684       if (halfWriter != null) halfWriter.close();
685       if (halfReader != null) halfReader.close(cacheConf.shouldEvictOnClose());
686     }
687   }
688 
689   private static boolean shouldCopyHFileMetaKey(byte[] key) {
690     return !HFile.isReservedFileInfoKey(key);
691   }
692 
693   private boolean doesTableExist(TableName tableName) throws Exception {
694     return hbAdmin.tableExists(tableName);
695   }
696 
697   /*
698    * Infers region boundaries for a new table.
699    * Parameter:
700    *   bdryMap is a map between keys to an integer belonging to {+1, -1}
701    *     If a key is a start key of a file, then it maps to +1
702    *     If a key is an end key of a file, then it maps to -1
703    * Algo:
704    * 1) Poll on the keys in order:
705    *    a) Keep adding the mapped values to these keys (runningSum)
706    *    b) Each time runningSum reaches 0, add the start Key from when the runningSum had started to a boundary list.
707    * 2) Return the boundary list.
708    */
709   public static byte[][] inferBoundaries(TreeMap<byte[], Integer> bdryMap) {
710     ArrayList<byte[]> keysArray = new ArrayList<byte[]>();
711     int runningValue = 0;
712     byte[] currStartKey = null;
713     boolean firstBoundary = true;
714 
715     for (Map.Entry<byte[], Integer> item: bdryMap.entrySet()) {
716       if (runningValue == 0) currStartKey = item.getKey();
717       runningValue += item.getValue();
718       if (runningValue == 0) {
719         if (!firstBoundary) keysArray.add(currStartKey);
720         firstBoundary = false;
721       }
722     }
723 
724     return keysArray.toArray(new byte[0][0]);
725   }
726 
727   /*
728    * If the table is created for the first time, then "completebulkload" reads the files twice.
729    * More modifications necessary if we want to avoid doing it.
730    */
731   private void createTable(TableName tableName, String dirPath) throws Exception {
732     Path hfofDir = new Path(dirPath);
733     FileSystem fs = hfofDir.getFileSystem(getConf());
734 
735     if (!fs.exists(hfofDir)) {
736       throw new FileNotFoundException("HFileOutputFormat dir " +
737           hfofDir + " not found");
738     }
739 
740     FileStatus[] familyDirStatuses = fs.listStatus(hfofDir);
741     if (familyDirStatuses == null) {
742       throw new FileNotFoundException("No families found in " + hfofDir);
743     }
744 
745     HTableDescriptor htd = new HTableDescriptor(tableName);
746     HColumnDescriptor hcd;
747 
748     // Add column families
749     // Build a set of keys
750     byte[][] keys;
751     TreeMap<byte[], Integer> map = new TreeMap<byte[], Integer>(Bytes.BYTES_COMPARATOR);
752 
753     for (FileStatus stat : familyDirStatuses) {
754       if (!stat.isDir()) {
755         LOG.warn("Skipping non-directory " + stat.getPath());
756         continue;
757       }
758       Path familyDir = stat.getPath();
759       // Skip _logs, etc
760       if (familyDir.getName().startsWith("_")) continue;
761       byte[] family = familyDir.getName().getBytes();
762 
763       hcd = new HColumnDescriptor(family);
764       htd.addFamily(hcd);
765 
766       Path[] hfiles = FileUtil.stat2Paths(fs.listStatus(familyDir));
767       for (Path hfile : hfiles) {
768         if (hfile.getName().startsWith("_")) continue;
769         HFile.Reader reader = HFile.createReader(fs, hfile,
770             new CacheConfig(getConf()));
771         final byte[] first, last;
772         try {
773           if (hcd.getCompressionType() != reader.getCompressionAlgorithm()) {
774             hcd.setCompressionType(reader.getCompressionAlgorithm());
775             LOG.info("Setting compression " + hcd.getCompressionType().name() +
776                      " for family " + hcd.toString());
777           }
778           reader.loadFileInfo();
779           first = reader.getFirstRowKey();
780           last =  reader.getLastRowKey();
781 
782           LOG.info("Trying to figure out region boundaries hfile=" + hfile +
783             " first=" + Bytes.toStringBinary(first) +
784             " last="  + Bytes.toStringBinary(last));
785 
786           // To eventually infer start key-end key boundaries
787           Integer value = map.containsKey(first)? map.get(first):0;
788           map.put(first, value+1);
789 
790           value = map.containsKey(last)? map.get(last):0;
791           map.put(last, value-1);
792         }  finally {
793           reader.close();
794         }
795       }
796     }
797 
798     keys = LoadIncrementalHFiles.inferBoundaries(map);
799     this.hbAdmin.createTable(htd,keys);
800 
801     LOG.info("Table "+ tableName +" is available!!");
802   }
803 
804   @Override
805   public int run(String[] args) throws Exception {
806     if (args.length != 2) {
807       usage();
808       return -1;
809     }
810 
811     String dirPath   = args[0];
812     TableName tableName = TableName.valueOf(args[1]);
813 
814     boolean tableExists   = this.doesTableExist(tableName);
815     if (!tableExists) this.createTable(tableName,dirPath);
816 
817     Path hfofDir = new Path(dirPath);
818     HTable table = new HTable(this.cfg, tableName);
819 
820     doBulkLoad(hfofDir, table);
821     return 0;
822   }
823 
824   public static void main(String[] args) throws Exception {
825     int ret = ToolRunner.run(new LoadIncrementalHFiles(HBaseConfiguration.create()), args);
826     System.exit(ret);
827   }
828 
829 }