|
1
|
<workflow-app name="lod_csv_export" xmlns="uri:oozie:workflow:0.4">
|
|
2
|
<!-- map reduce job that exports hbase data and prepares them for import to the lod_generation -->
|
|
3
|
|
|
4
|
<global>
|
|
5
|
<job-tracker>${jobTracker}</job-tracker>
|
|
6
|
<name-node>${nameNode}</name-node>
|
|
7
|
<configuration>
|
|
8
|
<property>
|
|
9
|
<name>mapred.job.queue.name</name>
|
|
10
|
<value>${queueName}</value>
|
|
11
|
</property>
|
|
12
|
<property>
|
|
13
|
<name>oozie.sqoop.log.level</name>
|
|
14
|
<value>DEBUG</value>
|
|
15
|
</property>
|
|
16
|
</configuration>
|
|
17
|
</global>
|
|
18
|
<start to='rdf_import'/>
|
|
19
|
<action name="csv_export">
|
|
20
|
<map-reduce>
|
|
21
|
|
|
22
|
<prepare>
|
|
23
|
<delete path="${nameNode}${lod_output}"/>
|
|
24
|
</prepare>
|
|
25
|
<configuration>
|
|
26
|
<property>
|
|
27
|
<name>hbase.mapreduce.scan</name>
|
|
28
|
<value>${wf:actionData('get-scanner')['scan']}</value>
|
|
29
|
</property>
|
|
30
|
<property>
|
|
31
|
<name>hbase.rootdir</name>
|
|
32
|
<value>$nameNode/hbase</value>
|
|
33
|
|
|
34
|
</property>
|
|
35
|
|
|
36
|
<property>
|
|
37
|
<name>hbase.security.authentication</name>
|
|
38
|
<value>simple</value>
|
|
39
|
</property>
|
|
40
|
<!-- ZOOKEEPER -->
|
|
41
|
|
|
42
|
<property>
|
|
43
|
<name>hbase.zookeeper.quorum</name>
|
|
44
|
<value>
|
|
45
|
namenode1.hadoop.dm.openaire.eu,namenode2.hadoop.dm.openaire.eu,jobtracker1.hadoop.dm.openaire.eu,jobtracker2.hadoop.dm.openaire.eu,hbase-master1.hadoop.dm.openaire.eu
|
|
46
|
</value>
|
|
47
|
<!-- <value>quorum1.t.hadoop.research-infrastructures.eu,quorum2.t.hadoop.research-infrastructures.eu,quorum3.t.hadoop.research-infrastructures.eu,quorum4.t.hadoop.research-infrastructures.eu,jobtracker.t.hadoop.research-infrastructures.eu
|
|
48
|
</value>-->
|
|
49
|
</property>
|
|
50
|
<property>
|
|
51
|
<name>zookeeper.znode.rootserver</name>
|
|
52
|
<value>root-region-server</value>
|
|
53
|
|
|
54
|
</property>
|
|
55
|
<property>
|
|
56
|
<name>hbase.zookeeper.property.clientPort</name>
|
|
57
|
<value>2181</value>
|
|
58
|
<!--<value>2182</value> -->
|
|
59
|
</property>
|
|
60
|
|
|
61
|
|
|
62
|
<!-- MR IO -->
|
|
63
|
|
|
64
|
<property>
|
|
65
|
<name>mapreduce.inputformat.class</name>
|
|
66
|
<value>org.apache.hadoop.hbase.mapreduce.TableInputFormat</value>
|
|
67
|
</property>
|
|
68
|
|
|
69
|
<property>
|
|
70
|
<name>mapred.mapoutput.key.class</name>
|
|
71
|
<value>org.apache.hadoop.io.Text</value>
|
|
72
|
</property>
|
|
73
|
<property>
|
|
74
|
<name>mapred.mapoutput.value.class</name>
|
|
75
|
<value>org.apache.hadoop.hbase.io.ImmutableBytesWritable</value>
|
|
76
|
</property>
|
|
77
|
<property>
|
|
78
|
<name>mapred.output.key.class</name>
|
|
79
|
<value>org.apache.hadoop.io.Text</value>
|
|
80
|
</property>
|
|
81
|
<property>
|
|
82
|
<name>mapred.output.value.class</name>
|
|
83
|
<value>org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat</value>
|
|
84
|
</property>
|
|
85
|
|
|
86
|
<!-- ## This is required for new MapReduce API usage -->
|
|
87
|
<property>
|
|
88
|
<name>mapred.mapper.new-api</name>
|
|
89
|
<value>true</value>
|
|
90
|
</property>
|
|
91
|
<property>
|
|
92
|
<name>mapred.reducer.new-api</name>
|
|
93
|
<value>true</value>
|
|
94
|
</property>
|
|
95
|
|
|
96
|
<!-- # Job-specific options -->
|
|
97
|
<property>
|
|
98
|
<name>dfs.blocksize</name>
|
|
99
|
<value>32M</value>
|
|
100
|
</property>
|
|
101
|
<property>
|
|
102
|
<name>mapred.output.compress</name>
|
|
103
|
<value>false</value>
|
|
104
|
</property>
|
|
105
|
<property>
|
|
106
|
<name>mapred.reduce.tasks.speculative.execution</name>
|
|
107
|
<value>false</value>
|
|
108
|
</property>
|
|
109
|
<property>
|
|
110
|
<name>mapred.reduce.tasks.speculative.execution</name>
|
|
111
|
<value>false</value>
|
|
112
|
</property>
|
|
113
|
<property>
|
|
114
|
<name>mapreduce.map.speculative</name>
|
|
115
|
<value>false</value>
|
|
116
|
</property>
|
|
117
|
|
|
118
|
<!-- I/O FORMAT -->
|
|
119
|
<!-- IMPORTANT: sets default delimeter used by text output writer. Required to fix
|
|
120
|
issue with traling tab added between id and value in multiple outputs -->
|
|
121
|
<property>
|
|
122
|
<name>mapred.textoutputformat.separator</name>
|
|
123
|
<value>${lod_delim}</value>
|
|
124
|
</property>
|
|
125
|
<!-- ## Names of all output ports -->
|
|
126
|
|
|
127
|
<property>
|
|
128
|
<name>mapreduce.multipleoutputs</name>
|
|
129
|
<value>
|
|
130
|
${out1} ${out2}
|
|
131
|
</value>
|
|
132
|
|
|
133
|
</property>
|
|
134
|
<property>
|
|
135
|
<name>mapreduce.multipleoutputs.namedOutput.${out1}.key</name>
|
|
136
|
<value>org.apache.hadoop.io.Text</value>
|
|
137
|
</property>
|
|
138
|
<property>
|
|
139
|
<name>mapreduce.multipleoutputs.namedOutput.${out1}.value</name>
|
|
140
|
<value>org.apache.hadoop.io.Text</value>
|
|
141
|
</property>
|
|
142
|
<property>
|
|
143
|
<name>mapreduce.multipleoutputs.namedOutput.${out1}.format</name>
|
|
144
|
<value>org.apache.hadoop.mapreduce.lib.output.TextOutputFormat</value>
|
|
145
|
</property>
|
|
146
|
<!-- datasourceLanguage -->
|
|
147
|
<property>
|
|
148
|
<name>mapreduce.multipleoutputs.namedOutput.${out2}.key</name>
|
|
149
|
<value>org.apache.hadoop.io.Text</value>
|
|
150
|
</property>
|
|
151
|
<property>
|
|
152
|
<name>mapreduce.multipleoutputs.namedOutput.${out2}.value</name>
|
|
153
|
<value>org.apache.hadoop.io.Text</value>
|
|
154
|
</property>
|
|
155
|
<property>
|
|
156
|
<name>mapreduce.multipleoutputs.namedOutput.${out2}.format</name>
|
|
157
|
<value>org.apache.hadoop.mapreduce.lib.output.TextOutputFormat</value>
|
|
158
|
</property>
|
|
159
|
|
|
160
|
|
|
161
|
<!-- ## Classes of mapper and reducer -->
|
|
162
|
|
|
163
|
<property>
|
|
164
|
<name>mapreduce.map.class</name>
|
|
165
|
<value>eu.dnetlib.data.mapreduce.hbase.lodExport.LodMapper</value>
|
|
166
|
</property>
|
|
167
|
<property>
|
|
168
|
<name>mapreduce.reduce.class</name>
|
|
169
|
<value>eu.dnetlib.data.mapreduce.hbase.lodExport.LodReducer</value>
|
|
170
|
</property>
|
|
171
|
<property>
|
|
172
|
<name>io.serializations</name>
|
|
173
|
<value>org.apache.hadoop.io.serializer.WritableSerialization</value>
|
|
174
|
</property>
|
|
175
|
|
|
176
|
<!-- ## Custom config -->
|
|
177
|
|
|
178
|
<!--delim character used to seperate fields in hdfs dump files <property> -->
|
|
179
|
<property>
|
|
180
|
<name>lod.delim</name>
|
|
181
|
<value>${lod_delim}</value>
|
|
182
|
</property>
|
|
183
|
<property>
|
|
184
|
<name>lod.enclosing</name>
|
|
185
|
<value>${lod_enclosing}</value>
|
|
186
|
</property>
|
|
187
|
|
|
188
|
<!--source hbase table -->
|
|
189
|
<property>
|
|
190
|
<name>hbase.mapreduce.inputtable</name>
|
|
191
|
<value>${lod_hbase_table}</value>
|
|
192
|
</property>
|
|
193
|
<property>
|
|
194
|
<name>hbase.mapred.inputtable</name>
|
|
195
|
<value>${lod_hbase_table}</value>
|
|
196
|
</property>
|
|
197
|
|
|
198
|
<!-- This directory does not correspond to a data store. In fact, this directory only
|
|
199
|
contains multiple data stores. It has to be set to the name of the workflow node.
|
|
200
|
-->
|
|
201
|
<property>
|
|
202
|
<name>mapred.output.dir</name>
|
|
203
|
<value>${lod_output}</value>
|
|
204
|
</property>
|
|
205
|
<property>
|
|
206
|
<name>index.conf</name>
|
|
207
|
<value>${lod_indexConf}</value>
|
|
208
|
</property>
|
|
209
|
<!-- ## Workflow node parameters -->
|
|
210
|
<property>
|
|
211
|
<name>mapred.reduce.tasks</name>
|
|
212
|
<value>${numReducers}</value>
|
|
213
|
</property>
|
|
214
|
|
|
215
|
</configuration>
|
|
216
|
</map-reduce>
|
|
217
|
<ok to="end"/>
|
|
218
|
|
|
219
|
<error to="fail"/>
|
|
220
|
</action>
|
|
221
|
<action name="rdf_import">
|
|
222
|
<map-reduce>
|
|
223
|
<prepare>
|
|
224
|
</prepare>
|
|
225
|
|
|
226
|
<configuration>
|
|
227
|
<property>
|
|
228
|
<name>hbase.security.authentication</name>
|
|
229
|
<value>simple</value>
|
|
230
|
</property>
|
|
231
|
|
|
232
|
<!-- ZOOKEEPER -->
|
|
233
|
<property>
|
|
234
|
<name>hbase.zookeeper.quorum</name>
|
|
235
|
<!--<value>
|
|
236
|
namenode1.hadoop.dm.openaire.eu,namenode2.hadoop.dm.openaire.eu,jobtracker1.hadoop.dm.openaire.eu,jobtracker2.hadoop.dm.openaire.eu,hbase-master1.hadoop.dm.openaire.eu
|
|
237
|
</value>-->
|
|
238
|
<value>
|
|
239
|
quorum1.t.hadoop.research-infrastructures.eu,quorum2.t.hadoop.research-infrastructures.eu,quorum3.t.hadoop.research-infrastructures.eu,quorum4.t.hadoop.research-infrastructures.eu,jobtracker.t.hadoop.research-infrastructures.eu
|
|
240
|
</value>
|
|
241
|
</property>
|
|
242
|
<property>
|
|
243
|
<name>zookeeper.znode.rootserver</name>
|
|
244
|
<value>root-region-server</value>
|
|
245
|
</property>
|
|
246
|
|
|
247
|
<!-- CSV PROPS GO HERE -->
|
|
248
|
<property>
|
|
249
|
<name>hbase.zookeeper.property.clientPort</name>
|
|
250
|
<value>2181</value>
|
|
251
|
<!--<value>2182</value> -->
|
|
252
|
</property>
|
|
253
|
|
|
254
|
<property>
|
|
255
|
<name>CSVLineRecordReader.FORMAT_DELIMITER</name>
|
|
256
|
<value>"</value>
|
|
257
|
</property>
|
|
258
|
<property>
|
|
259
|
<name>CSVLineRecordReader.FORMAT_SEPARATOR</name>
|
|
260
|
<value>,</value>
|
|
261
|
</property>
|
|
262
|
|
|
263
|
<property>
|
|
264
|
<name>CSVNLineInputFormat.LINES_PER_MAP</name>
|
|
265
|
<value>40000</value>
|
|
266
|
</property>
|
|
267
|
<property>
|
|
268
|
<name>CSVLineRecordReader.IS_ZIPFILE</name>
|
|
269
|
<value>false</value>
|
|
270
|
</property>
|
|
271
|
|
|
272
|
<!-- MR IO -->
|
|
273
|
<!-- TODO here: add csv input format to mapper
|
|
274
|
-->
|
|
275
|
|
|
276
|
<property>
|
|
277
|
<name>mapred.input.dir</name>
|
|
278
|
<value>${lod_EntitiesInputFile}</value>
|
|
279
|
</property>
|
|
280
|
|
|
281
|
<property>
|
|
282
|
<name>mapreduce.inputformat.class</name>
|
|
283
|
<value>org.apache.hadoop.mapreduce.lib.input.TextInputFormat</value>
|
|
284
|
</property>
|
|
285
|
|
|
286
|
<property>
|
|
287
|
<name>mapred.mapoutput.key.class</name>
|
|
288
|
<value>org.apache.hadoop.io.Text</value>
|
|
289
|
</property>
|
|
290
|
|
|
291
|
<property>
|
|
292
|
<name>mapred.mapoutput.value.class</name>
|
|
293
|
<value>java.util.List</value>
|
|
294
|
</property>
|
|
295
|
|
|
296
|
<!-- <property>
|
|
297
|
<name>mapred.output.key.class</name>
|
|
298
|
<value>org.apache.hadoop.io.Text</value>
|
|
299
|
</property>-->
|
|
300
|
<property>
|
|
301
|
<name>mapred.output.value.class</name>
|
|
302
|
<value>org.apache.hadoop.mapreduce.lib.output.NullOutputFormat</value>
|
|
303
|
</property>
|
|
304
|
|
|
305
|
|
|
306
|
<!-- ## This is required for new MapReduce API usage -->
|
|
307
|
<property>
|
|
308
|
<name>mapred.mapper.new-api</name>
|
|
309
|
<value>true</value>
|
|
310
|
</property>
|
|
311
|
<property>
|
|
312
|
<name>mapred.reducer.new-api</name>
|
|
313
|
<value>true</value>
|
|
314
|
</property>
|
|
315
|
|
|
316
|
<!-- # Job-specific options -->
|
|
317
|
<property>
|
|
318
|
<name>dfs.blocksize</name>
|
|
319
|
<value>32M</value>
|
|
320
|
</property>
|
|
321
|
<property>
|
|
322
|
<name>mapred.output.compress</name>
|
|
323
|
<value>false</value>
|
|
324
|
</property>
|
|
325
|
<property>
|
|
326
|
<name>mapred.reduce.tasks.speculative.execution</name>
|
|
327
|
<value>false</value>
|
|
328
|
</property>
|
|
329
|
<property>
|
|
330
|
<name>mapred.reduce.tasks.speculative.execution</name>
|
|
331
|
<value>false</value>
|
|
332
|
</property>
|
|
333
|
<property>
|
|
334
|
<name>mapreduce.map.speculative</name>
|
|
335
|
<value>false</value>
|
|
336
|
</property>
|
|
337
|
|
|
338
|
|
|
339
|
<property>
|
|
340
|
<name>map.output.key.field.separator</name>
|
|
341
|
<value>${lod_delim}</value>
|
|
342
|
</property>
|
|
343
|
|
|
344
|
|
|
345
|
<!-- ## Classes of mapper and reducer -->
|
|
346
|
|
|
347
|
<property>
|
|
348
|
<name>mapreduce.map.class</name>
|
|
349
|
<value>eu.dnetlib.data.mapreduce.hbase.lodImport.LodImportMapper</value>
|
|
350
|
</property>
|
|
351
|
<property>
|
|
352
|
<name>mapreduce.reduce.class</name>
|
|
353
|
<value>eu.dnetlib.data.mapreduce.hbase.lodImport.LodImportReducer</value>
|
|
354
|
</property>
|
|
355
|
<property>
|
|
356
|
<name>io.serializations</name>
|
|
357
|
<value>org.apache.hadoop.io.serializer.WritableSerialization</value>
|
|
358
|
</property>
|
|
359
|
|
|
360
|
<!-- ## Custom config -->
|
|
361
|
|
|
362
|
<!--delim character used to seperate fields in hdfs dump files <property> -->
|
|
363
|
<property>
|
|
364
|
<name>lod.delim</name>
|
|
365
|
<value>${lod_delim}</value>
|
|
366
|
</property>
|
|
367
|
<property>
|
|
368
|
<name>lod.enclosing</name>
|
|
369
|
<value>${lod_enclosing}</value>
|
|
370
|
</property>
|
|
371
|
|
|
372
|
<property>
|
|
373
|
<name>lod.lastExecutionDate</name>
|
|
374
|
<value>${lod_lastExecutionDate}</value>
|
|
375
|
</property>
|
|
376
|
|
|
377
|
<property>
|
|
378
|
<name>lod.conLine</name>
|
|
379
|
<value>${lod_conLine}</value>
|
|
380
|
</property>
|
|
381
|
|
|
382
|
|
|
383
|
<property>
|
|
384
|
<name>lod.username</name>
|
|
385
|
<value>${lod_username}</value>
|
|
386
|
</property>
|
|
387
|
|
|
388
|
|
|
389
|
<property>
|
|
390
|
<name>lod.password</name>
|
|
391
|
<value>${lod_password}</value>
|
|
392
|
</property>
|
|
393
|
|
|
394
|
<property>
|
|
395
|
<name>lod_minCpart</name>
|
|
396
|
<value>${lod_minCpart}</value>
|
|
397
|
</property>
|
|
398
|
|
|
399
|
|
|
400
|
<property>
|
|
401
|
<name>lod_maxCpart</name>
|
|
402
|
<value>${lod_maxCpart}</value>
|
|
403
|
</property>
|
|
404
|
|
|
405
|
<property>
|
|
406
|
<name>lod.part</name>
|
|
407
|
<value>${lod_part}</value>
|
|
408
|
</property>
|
|
409
|
|
|
410
|
<property>
|
|
411
|
<name>lod.jsonRels</name>
|
|
412
|
<value>${lod_jsonRels}</value>
|
|
413
|
</property>
|
|
414
|
|
|
415
|
<property>
|
|
416
|
<name>lod.jsonEntities</name>
|
|
417
|
<value>${lod_jsonEntities}</value>
|
|
418
|
</property>
|
|
419
|
|
|
420
|
<property>
|
|
421
|
<name>lod.defaultGraph</name>
|
|
422
|
<value>${lod_defaultGraph}</value>
|
|
423
|
</property>
|
|
424
|
|
|
425
|
<property>
|
|
426
|
<name>mapred.reduce.tasks</name>
|
|
427
|
<value>${numReducers}</value>
|
|
428
|
</property>
|
|
429
|
|
|
430
|
<property>
|
|
431
|
<name>index.conf</name>
|
|
432
|
<value>${lod_indexConf}</value>
|
|
433
|
</property>
|
|
434
|
|
|
435
|
<property>
|
|
436
|
<name>mapred.output.dir</name>
|
|
437
|
<value>${lod_output}test</value>
|
|
438
|
</property>
|
|
439
|
|
|
440
|
</configuration>
|
|
441
|
</map-reduce>
|
|
442
|
<ok to="end"/>
|
|
443
|
|
|
444
|
<error to="fail"/>
|
|
445
|
</action>
|
|
446
|
|
|
447
|
|
|
448
|
<action name="cleanUpHDFS">
|
|
449
|
<fs>
|
|
450
|
<delete path=">${lod_output}test" />
|
|
451
|
</fs>
|
|
452
|
<ok to="end" />
|
|
453
|
<error to="fail" />
|
|
454
|
</action>
|
|
455
|
|
|
456
|
<kill name="fail">
|
|
457
|
<message>
|
|
458
|
Unfortunately, the process failed -- error message: [${wf:errorMessage(wf:lastErrorNode())}]
|
|
459
|
</message>
|
|
460
|
</kill>
|
|
461
|
<end name="end"/>
|
|
462
|
</workflow-app>
|