Package org.apache.hadoop.mapreduce.filecache

Examples of org.apache.hadoop.mapreduce.filecache.TaskDistributedCacheManager

159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
    os.close();

    // ****** Imitate TaskRunner code.
    TrackerDistributedCacheManager manager =
      new TrackerDistributedCacheManager(conf, taskController);
    TaskDistributedCacheManager handle =
      manager.newTaskDistributedCacheManager(subConf);
    assertNull(null, DistributedCache.getLocalCacheFiles(subConf));
    File workDir = new File(new Path(TEST_ROOT_DIR, "workdir").toString());
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName),
        TaskTracker.getPublicDistributedCacheDir());
    // ****** End of imitating TaskRunner code

    Path[] localCacheFiles = DistributedCache.getLocalCacheFiles(subConf);
    assertNotNull(null, localCacheFiles);
    assertEquals(2, localCacheFiles.length);
    Path cachedFirstFile = localCacheFiles[0];
    Path cachedSecondFile = localCacheFiles[1];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile));

    assertEquals(1, handle.getClassPaths().size());
    assertEquals(cachedSecondFile.toString(), handle.getClassPaths().get(0));

    checkFilePermissions(localCacheFiles);

    // Cleanup
    handle.release();
    manager.purgeCache();
    assertFalse(pathToFile(cachedFirstFile).exists());
  }
View Full Code Here
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
    Configuration conf1 = job1.getConfiguration();
    TrackerDistributedCacheManager.determineTimestamps(conf1);
    TrackerDistributedCacheManager.determineCacheVisibilities(conf1);

    // Task localizing for first job
    TaskDistributedCacheManager handle = manager
        .newTaskDistributedCacheManager(conf1);
    handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    handle.release();
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      assertEquals(0, manager.getReferenceCount(c.uri, conf1, c.timestamp));
    }
   
    Path thirdCacheFile = new Path(TEST_ROOT_DIR, "thirdcachefile");
    createPrivateTempFile(thirdCacheFile);
   
    // Configures another job with three regular files.
    Job job2 = Job.getInstance(cluster, conf);
    job2.setUser(userName);
    // add a file that would get failed to localize
    job2.addCacheFile(firstCacheFile.toUri());
    // add a file that is already localized by different job
    job2.addCacheFile(secondCacheFile.toUri());
    // add a file that is never localized
    job2.addCacheFile(thirdCacheFile.toUri());
    Configuration conf2 = job2.getConfiguration();
    TrackerDistributedCacheManager.determineTimestamps(conf2);
    TrackerDistributedCacheManager.determineCacheVisibilities(conf2);

    // Task localizing for second job
    // localization for the "firstCacheFile" will fail.
    handle = manager.newTaskDistributedCacheManager(conf2);
    Throwable th = null;
    try {
      handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException e) {
      th = e;
      Log.info("Exception during setup", e);
    }
    assertNotNull(th);
    assertTrue(th.getMessage().contains("fake fail"));
    handle.release();
    th = null;
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      try {
        assertEquals(0, manager.getReferenceCount(c.uri, conf2, c.timestamp));
      } catch (IOException ie) {
        th = ie;
        Log.info("Exception getting reference count for " + c.uri, ie);
View Full Code Here
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
    Configuration conf1 = job1.getConfiguration();
    TrackerDistributedCacheManager.determineTimestamps(conf1);
    TrackerDistributedCacheManager.determineCacheVisibilities(conf1);

    // Task localizing for job
    TaskDistributedCacheManager handle = manager
        .newTaskDistributedCacheManager(conf1);
    handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    TaskDistributedCacheManager.CacheFile c = handle.getCacheFiles().get(0);
    String distCacheDir;
    if ("true".equals(visibility)) {
      distCacheDir = TaskTracker.getPublicDistributedCacheDir();
    } else {
      distCacheDir = TaskTracker.getPrivateDistributedCacheDir(userName);
View Full Code Here
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
    TrackerDistributedCacheManager.determineTimestamps(subConf);
    TrackerDistributedCacheManager.determineCacheVisibilities(subConf);
    // ****** End of imitating JobClient code

    // ****** Imitate TaskRunner code.
    TaskDistributedCacheManager handle =
      manager.newTaskDistributedCacheManager(subConf);
    assertNull(null, DistributedCache.getLocalCacheFiles(subConf));
    File workDir = new File(new Path(TEST_ROOT_DIR, "workdir").toString());
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName),
        TaskTracker.getPublicDistributedCacheDir());
    // ****** End of imitating TaskRunner code

    Path[] localCacheFiles = DistributedCache.getLocalCacheFiles(subConf);
    assertNotNull(null, localCacheFiles);
    assertEquals(1, localCacheFiles.length);
    Path cachedFirstFile = localCacheFiles[0];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile));
    // release
    handle.release();
   
    // change the file timestamp
    FileSystem fs = FileSystem.get(myConf);
    ((FakeFileSystem)fs).advanceClock(1);

    // running a task of the same job
    Throwable th = null;
    try {
      handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName), TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException ie) {
      th = ie;
    }
    assertNotNull("Throwable is null", th);
    assertTrue("Exception message does not match",
        th.getMessage().contains("has changed on HDFS since job started"));
    // release
    handle.release();
   
    // submit another job
    Configuration subConf2 = new Configuration(myConf);
    subConf2.set(MRJobConfig.USER_NAME, userName);
    DistributedCache.addCacheFile(firstCacheFile.toUri(), subConf2);
    TrackerDistributedCacheManager.determineTimestamps(subConf2);
    TrackerDistributedCacheManager.determineCacheVisibilities(subConf2);
   
    handle =
      manager.newTaskDistributedCacheManager(subConf2);
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName), TaskTracker.getPublicDistributedCacheDir());
    Path[] localCacheFiles2 = DistributedCache.getLocalCacheFiles(subConf2);
    assertNotNull(null, localCacheFiles2);
    assertEquals(1, localCacheFiles2.length);
    Path cachedFirstFile2 = localCacheFiles2[0];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile2);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile2));
   
    // assert that two localizations point to different paths
    assertFalse("two jobs with different timestamps did not localize" +
        " in different paths", cachedFirstFile.equals(cachedFirstFile2));
    // release
    handle.release();
  }
View Full Code Here
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
    os.close();

    // ****** Imitate TaskRunner code.
    TrackerDistributedCacheManager manager =
      new TrackerDistributedCacheManager(conf, taskController);
    TaskDistributedCacheManager handle =
      manager.newTaskDistributedCacheManager(subConf);
    assertNull(null, DistributedCache.getLocalCacheFiles(subConf));
    File workDir = new File(new Path(TEST_ROOT_DIR, "workdir").toString());
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName),
        TaskTracker.getPublicDistributedCacheDir());
    // ****** End of imitating TaskRunner code

    Path[] localCacheFiles = DistributedCache.getLocalCacheFiles(subConf);
    assertNotNull(null, localCacheFiles);
    assertEquals(2, localCacheFiles.length);
    Path cachedFirstFile = localCacheFiles[0];
    Path cachedSecondFile = localCacheFiles[1];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile));

    assertEquals(1, handle.getClassPaths().size());
    assertEquals(cachedSecondFile.toString(), handle.getClassPaths().get(0));

    checkFilePermissions(localCacheFiles);

    // Cleanup
    handle.release();
    manager.purgeCache();
    assertFalse(pathToFile(cachedFirstFile).exists());
  }
View Full Code Here
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
    Configuration conf1 = job1.getConfiguration();
    ClientDistributedCacheManager.determineTimestamps(conf1);
    ClientDistributedCacheManager.determineCacheVisibilities(conf1);

    // Task localizing for first job
    TaskDistributedCacheManager handle = manager
        .newTaskDistributedCacheManager(conf1);
    handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    handle.release();
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      assertEquals(0, manager.getReferenceCount(c.uri, conf1, c.timestamp,
          c.owner, false));
    }
   
    Path thirdCacheFile = new Path(TEST_ROOT_DIR, "thirdcachefile");
    createPrivateTempFile(thirdCacheFile);
   
    // Configures another job with three regular files.
    Job job2 = Job.getInstance(conf);
    job2.setUser(userName);
    // add a file that would get failed to localize
    job2.addCacheFile(firstCacheFile.toUri());
    // add a file that is already localized by different job
    job2.addCacheFile(secondCacheFile.toUri());
    // add a file that is never localized
    job2.addCacheFile(thirdCacheFile.toUri());
    Configuration conf2 = job2.getConfiguration();
    ClientDistributedCacheManager.determineTimestamps(conf2);
    ClientDistributedCacheManager.determineCacheVisibilities(conf2);

    // Task localizing for second job
    // localization for the "firstCacheFile" will fail.
    handle = manager.newTaskDistributedCacheManager(conf2);
    Throwable th = null;
    try {
      handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException e) {
      th = e;
      Log.info("Exception during setup", e);
    }
    assertNotNull(th);
    assertTrue(th.getMessage().contains("fake fail"));
    handle.release();
    th = null;
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      try {
        assertEquals(0, manager.getReferenceCount(c.uri, conf2, c.timestamp,
            c.owner, false));
      } catch (IOException ie) {
        th = ie;
View Full Code Here
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
    Configuration conf1 = job1.getConfiguration();
    ClientDistributedCacheManager.determineTimestamps(conf1);
    ClientDistributedCacheManager.determineCacheVisibilities(conf1);

    // Task localizing for job
    TaskDistributedCacheManager handle = manager
        .newTaskDistributedCacheManager(conf1);
    handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    TaskDistributedCacheManager.CacheFile c = handle.getCacheFiles().get(0);
    String distCacheDir;
    if (visibility) {
      distCacheDir = TaskTracker.getPublicDistributedCacheDir();
    } else {
      distCacheDir = TaskTracker.getPrivateDistributedCacheDir(userName);
View Full Code Here
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
    ClientDistributedCacheManager.determineTimestamps(subConf);
    ClientDistributedCacheManager.determineCacheVisibilities(subConf);
    // ****** End of imitating JobClient code

    // ****** Imitate TaskRunner code.
    TaskDistributedCacheManager handle =
      manager.newTaskDistributedCacheManager(subConf);
    assertNull(null, DistributedCache.getLocalCacheFiles(subConf));
    File workDir = new File(new Path(TEST_ROOT_DIR, "workdir").toString());
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName),
        TaskTracker.getPublicDistributedCacheDir());
    // ****** End of imitating TaskRunner code

    Path[] localCacheFiles = DistributedCache.getLocalCacheFiles(subConf);
    assertNotNull(null, localCacheFiles);
    assertEquals(1, localCacheFiles.length);
    Path cachedFirstFile = localCacheFiles[0];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile));
    // release
    handle.release();
   
    // change the file timestamp
    FileSystem fs = FileSystem.get(myConf);
    ((FakeFileSystem)fs).advanceClock(1);

    // running a task of the same job
    Throwable th = null;
    try {
      handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName), TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException ie) {
      th = ie;
    }
    assertNotNull("Throwable is null", th);
    assertTrue("Exception message does not match",
        th.getMessage().contains("has changed on HDFS since job started"));
    // release
    handle.release();
   
    // running a task of the same job on another TaskTracker which has never
    // initialized the cache
    TrackerDistributedCacheManager manager2 =
      new TrackerDistributedCacheManager(myConf, taskController);
    TaskDistributedCacheManager handle2 =
      manager2.newTaskDistributedCacheManager(subConf);
    File workDir2 = new File(new Path(TEST_ROOT_DIR, "workdir2").toString());
    th = null;
    try {
      handle2.setup(localDirAllocator, workDir2, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException ie) {
      th = ie;
    }
View Full Code Here
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
    os.close();

    // ****** Imitate TaskRunner code.
    TrackerDistributedCacheManager manager =
      new TrackerDistributedCacheManager(conf, taskController);
    TaskDistributedCacheManager handle =
      manager.newTaskDistributedCacheManager(subConf);
    assertNull(null, DistributedCache.getLocalCacheFiles(subConf));
    File workDir = new File(new Path(TEST_ROOT_DIR, "workdir").toString());
    handle.setup(localDirAllocator, workDir, TaskTracker
        .getPrivateDistributedCacheDir(userName),
        TaskTracker.getPublicDistributedCacheDir());
    // ****** End of imitating TaskRunner code

    Path[] localCacheFiles = DistributedCache.getLocalCacheFiles(subConf);
    assertNotNull(null, localCacheFiles);
    assertEquals(2, localCacheFiles.length);
    Path cachedFirstFile = localCacheFiles[0];
    Path cachedSecondFile = localCacheFiles[1];
    assertFileLengthEquals(firstCacheFile, cachedFirstFile);
    assertFalse("Paths should be different.",
        firstCacheFile.equals(cachedFirstFile));

    assertEquals(1, handle.getClassPaths().size());
    assertEquals(cachedSecondFile.toString(), handle.getClassPaths().get(0));

    checkFilePermissions(localCacheFiles);

    // Cleanup
    handle.release();
    manager.purgeCache();
    assertFalse(pathToFile(cachedFirstFile).exists());
  }
View Full Code Here
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
    Configuration conf1 = job1.getConfiguration();
    TrackerDistributedCacheManager.determineTimestamps(conf1);
    TrackerDistributedCacheManager.determineCacheVisibilities(conf1);

    // Task localizing for first job
    TaskDistributedCacheManager handle = manager
        .newTaskDistributedCacheManager(conf1);
    handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    handle.release();
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      assertEquals(0, manager.getReferenceCount(c.uri, conf1, c.timestamp,
          c.owner));
    }
   
    Path thirdCacheFile = new Path(TEST_ROOT_DIR, "thirdcachefile");
    createPrivateTempFile(thirdCacheFile);
   
    // Configures another job with three regular files.
    Job job2 = Job.getInstance(cluster, conf);
    job2.setUser(userName);
    // add a file that would get failed to localize
    job2.addCacheFile(firstCacheFile.toUri());
    // add a file that is already localized by different job
    job2.addCacheFile(secondCacheFile.toUri());
    // add a file that is never localized
    job2.addCacheFile(thirdCacheFile.toUri());
    Configuration conf2 = job2.getConfiguration();
    TrackerDistributedCacheManager.determineTimestamps(conf2);
    TrackerDistributedCacheManager.determineCacheVisibilities(conf2);

    // Task localizing for second job
    // localization for the "firstCacheFile" will fail.
    handle = manager.newTaskDistributedCacheManager(conf2);
    Throwable th = null;
    try {
      handle.setup(localDirAllocator, workDir, TaskTracker
          .getPrivateDistributedCacheDir(userName),
          TaskTracker.getPublicDistributedCacheDir());
    } catch (IOException e) {
      th = e;
      Log.info("Exception during setup", e);
    }
    assertNotNull(th);
    assertTrue(th.getMessage().contains("fake fail"));
    handle.release();
    th = null;
    for (TaskDistributedCacheManager.CacheFile c : handle.getCacheFiles()) {
      try {
        assertEquals(0, manager.getReferenceCount(c.uri, conf2, c.timestamp,
            c.owner));
      } catch (IOException ie) {
        th = ie;
View Full Code Here
TOP

Related Classes of org.apache.hadoop.mapreduce.filecache.TaskDistributedCacheManager

  • org.apache.hadoop.fs.FileStatus
  • org.apache.hadoop.fs.FileSystem
  • org.apache.hadoop.fs.Path
  • org.apache.hadoop.mapreduce.filecache.TestTrackerDistributedCacheManager
  • java.io.File
  • java.net.URLClassLoader
  • java.net.URI
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.