Package org.apache.hadoop.chukwa.extraction.engine

Examples of org.apache.hadoop.chukwa.extraction.engine.ChukwaRecord

40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
  public void process(ChukwaRecordKey key, Iterator<ChukwaRecord> values,
      OutputCollector<ChukwaRecordKey, ChukwaRecord> output, Reporter reporter) {
    try {
      HashMap<String, String> data = new HashMap<String, String>();

      ChukwaRecord record = null;
      String[] fields = null;
      while (values.hasNext()) {
        record = values.next();
        fields = record.getFields();
        for (String field : fields) {
          data.put(field, record.getValue(field));
        }
      }

      // Extract initial time: SUBMIT_TIME
      long initTime = Long.parseLong(data.get("SUBMIT_TIME"));

      // Extract HodId
      // maybe use a regex to extract this and load it from configuration
      // JOBCONF=
      // "/user/xxx/mapredsystem/563976.xxx.yyy.com/job_200809062051_0001/job.xml"
      String jobConf = data.get("JOBCONF");
      int idx = jobConf.indexOf("mapredsystem/");
      idx += 13;
      int idx2 = jobConf.indexOf(".", idx);
      data.put("HodId", jobConf.substring(idx, idx2));

      ChukwaRecordKey newKey = new ChukwaRecordKey();
      newKey.setKey("" + initTime);
      newKey.setReduceType("MRJob");

      ChukwaRecord newRecord = new ChukwaRecord();
      newRecord.add(Record.tagsField, record.getValue(Record.tagsField));
      newRecord.setTime(initTime);
      newRecord.add(Record.tagsField, record.getValue(Record.tagsField));
      Iterator<String> it = data.keySet().iterator();
      while (it.hasNext()) {
        String field = it.next();
        newRecord.add(field, data.get(field));
      }

      output.collect(newKey, newRecord);
    } catch (IOException e) {
      log.warn("Unable to collect output in JobLogHistoryReduceProcessor ["
View Full Code Here
26
27
28
29
30
31
32
33
34
  protected void parse(String recordEntry,
                       OutputCollector<ChukwaRecordKey, ChukwaRecord> output,
                       Reporter reporter) throws Throwable {
    ChukwaRecordKey key = new ChukwaRecordKey("someReduceType", recordEntry);
    ChukwaRecord record = new ChukwaRecord();

    output.collect(key, record);
  }
View Full Code Here
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
    try {

      Pattern task_id_pat = Pattern.compile("attempt_[0-9]*_[0-9]*_[mr]_([0-9]*)_[0-9]*");

      ChukwaRecordKey key = new ChukwaRecordKey();
      ChukwaRecord record = new ChukwaRecord();

      // initialize data structures for checking FSM
      // should see 10 maps, 8 reduces
      boolean mapSeen[] = new boolean[10];
      boolean reduceSeen[] = new boolean[8];
      boolean reduceShuffleSeen[] = new boolean[8];
      boolean reduceSortSeen[] = new boolean[8];
      boolean reduceReducerSeen[] = new boolean[8];
      for (int i = 0; i < 10; i++) mapSeen[i] = false;
      for (int i = 0; i < 8; i++) {
        reduceSeen[i] = false;
        reduceShuffleSeen[i] = false;
        reduceSortSeen[i] = false;
        reduceReducerSeen[i] = false;
      }

      Path fsm_outputs = new Path(FSM_OUTPUT_PATH.toString()+File.separator+
        "/*/MAPREDUCE_FSM/MAPREDUCE_FSM*.evt");
      FileStatus [] files;
      files = fileSys.globStatus(fsm_outputs);
      int count = 0;

      for (int i = 0; i < files.length; i++) {
        SequenceFile.Reader r = new SequenceFile.Reader(fileSys, files[i].getPath(), conf);
        System.out.println("Processing files " + files[i].getPath().toString());
        while (r.next(key, record)) {
          String state_name = record.getValue("STATE_NAME");
          String task_id = record.getValue("TASK_ID");
         
          Matcher m = task_id_pat.matcher(task_id);
          if (!m.matches()) {
            continue;
          }
View Full Code Here
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
      FileStatus [] files;
      files = fileSys.globStatus(fsm_outputs);
      int count = 0;
      int numHDFSRead = 0, numHDFSWrite = 0, numShuffles = 0;
      ChukwaRecordKey key = new ChukwaRecordKey();
      ChukwaRecord record = new ChukwaRecord();

      for (int i = 0; i < files.length; i++) {
        SequenceFile.Reader r = new SequenceFile.Reader(fileSys, files[i].getPath(), conf);
        System.out.println("Processing files " + files[i].getPath().toString());
        while (r.next(key, record)) {
          String state_name = record.getValue("STATE_NAME");
 
          if (state_name.equals("READ_LOCAL") || state_name.equals("READ_REMOTE"))
          {
            numHDFSRead++;
          } else if (state_name.equals("WRITE_LOCAL") || state_name.equals("WRITE_REMOTE")
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
      String body = recordEntry.substring(idx + 1);
      body = body.replaceAll("\n", "");
      // log.info("record [" + recordEntry + "] body [" + body +"]");
      Date d = sdf.parse(dStr);

      ChukwaRecord record = new ChukwaRecord();

      String[] list = body.split(",");
      for(String pair : list) {
        String[] kv = pair.split("=");
        record.add(kv[0], kv[1]);
      }
      record.add("cluster", chunk.getTag("cluster"));
      buildGenericRecord(record, d.getTime(), "summary");
      output.collect(key, record);
    } catch (ParseException e) {
      log.warn("Wrong format in JobSummary [" + recordEntry + "]",
          e);
View Full Code Here
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
      }
      kvMatcher.reset(recordEntry);
      if (!kvMatcher.find()) {
        throw new IOException("Failed to find record");
      }
      ChukwaRecord rec = new ChukwaRecord();
      do {
        rec.add(kvMatcher.group(1), kvMatcher.group(2));
      } while (kvMatcher.find());
      Locality loc = getLocality(rec.getValue("src"), rec.getValue("dest"));
      rec.add("locality", loc.getLabel());

      calendar.setTimeInMillis(ms);
      calendar.set(Calendar.SECOND, 0);
      calendar.set(Calendar.MILLISECOND, 0);
      ms = calendar.getTimeInMillis();
      calendar.set(Calendar.MINUTE, 0);
      key.setKey(calendar.getTimeInMillis() + "/" + loc.getLabel() + "/" +
                 rec.getValue("op").toLowerCase() + "/" + ms);
      key.setReduceType("ClientTrace");
      rec.setTime(ms);

      rec.add(Record.tagsField, chunk.getTags());
      rec.add(Record.sourceField, chunk.getSource());
      rec.add(Record.applicationField, chunk.getStreamName());
      rec.add("actual_time",(new Long(ms_fullresolution)).toString());
      output.collect(key, rec);

    } catch (ParseException e) {
      log.warn("Unable to parse the date in DefaultProcessor ["
          + recordEntry + "]", e);
View Full Code Here
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
          long ts = Long.parseLong(jsonTs);

          String jsonValue = null;
          Iterator<String> it = jsonData.keys();

          ChukwaRecord record = null;

          while (it.hasNext()) {
            jsonTs = it.next();
            jsonValue = jsonData.getString(jsonTs);

            record = new ChukwaRecord();
            key = new ChukwaRecordKey();
            this.buildGenericRecord(record, null, ts, "Ywatch");
            record.add("poller", poller);
            record.add("host", host);
            record.add("metricName", metricName);
            record.add("value", jsonValue);
            output.collect(key, record);
            log.info("YWatchProcessor output 1 metric");
          }

        } catch (IOException e) {
View Full Code Here
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
      log.debug("Parser Open [" + fileName + "]");

      long timestamp = 0;
      int listSize = 0;
      ChukwaRecordKey key = new ChukwaRecordKey();
      ChukwaRecord record = new ChukwaRecord();

      r = new SequenceFile.Reader(fs, new Path(fileName), conf);

      log.debug("readData Open2 [" + fileName + "]");
      if ((fileName.equals(res.fileName)) && (res.position != -1)) {
        r.seek(res.position);
      }
      res.fileName = fileName;

      while (r.next(key, record)) {
        if (record != null) {
          res.position = r.getPosition();

          timestamp = record.getTime();
          res.currentTs = timestamp;
          log.debug("\nSearch for startDate: " + new Date(t0) + " is :"
              + new Date(timestamp));

          if (timestamp < t0) {
            // log.debug("Line not in range. Skipping: " +record);
            continue;
          } else if (timestamp < t1) {
            log.debug("In Range: " + record.toString());
            boolean valid = false;

            if ((filter == null || filter.equals(""))) {
              valid = true;
            } else if (isValid(record, filter)) {
              valid = true;
            }

            if (valid) {
              records.add(record);
              record = new ChukwaRecord();
              listSize = records.size();
              if (listSize >= maxRows) {
                // maxRow so stop here
                // Update token
                token.key = key.getKey();
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
      OutputCollector<ChukwaRecordKey, ChukwaRecord> output, Reporter reporter)
      throws Throwable {
    try {
      String dStr = recordEntry.substring(0, 23);
      Date d = sdf.parse(dStr);
      ChukwaRecord record = new ChukwaRecord();

      if (this.chunk.getStreamName().indexOf("datanode") > 0) {
        buildGenericRecord(record, recordEntry, d.getTime(), dataNodeType);
      } else if (this.chunk.getStreamName().indexOf("namenode") > 0) {
        buildGenericRecord(record, recordEntry, d.getTime(), nameNodeType);
View Full Code Here
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
              valid = true;
            }

            if (valid) {
              // System.out.println("In Range In Filter: " + line);
              ChukwaRecord record = new ChukwaRecord();
              record.setTime(timestamp);
              record.add("offset", "" + offset);
              record.add(Record.bodyField, data[1]);
              record.add(Record.sourceField, dataSource);

              records.add(record);
              listSize = records.size();
              if (listSize > maxRows) {
                records.remove(0);
View Full Code Here
TOP

Related Classes of org.apache.hadoop.chukwa.extraction.engine.ChukwaRecord

  • edu.berkeley.chukwa_xtrace.XtrExtract$MapClass
  • org.apache.hadoop.chukwa.analysis.salsa.fsm.TestFSMBuilder
  • org.apache.hadoop.chukwa.ChukwaStorage
  • org.apache.hadoop.chukwa.datacollection.writer.test.demux.TextParser
  • org.apache.hadoop.chukwa.dataloader.MetricDataLoader
  • org.apache.hadoop.chukwa.extraction.database.DatabaseHelper
  • org.apache.hadoop.chukwa.extraction.database.DatabaseLoader
  • org.apache.hadoop.chukwa.extraction.database.MetricDataLoader
  • org.apache.hadoop.chukwa.extraction.database.MRJobCounters
  • org.apache.hadoop.chukwa.extraction.demux.processor.mapper.ChunkSaver
    • 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.