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

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

35
36
37
38
39
40
41
42
43
44
45
  public void parse(String line, OutputCollector<Text, ChukwaRecord> output,
      Reporter reporter)
  {
    log.info("record: [" + line + "] type[" + chunk.getDataType() + "]");
   
    ChukwaRecord record = new ChukwaRecord();
    buildGenericRecord(record,line, System.currentTimeMillis(),recordType);
    key.set("" + chunk.getSeqID());
    try
    {
      output.collect(key, record);
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
      log.info("JobLogHistoryProcessor Matches");

      try
      {
        Date d = sdf.parse(matcher.group(1).trim());
        ChukwaRecord record = new ChukwaRecord();

        logLevel = matcher.group(2);
        className = matcher.group(3);
        body = matcher.group(4);

        key.set("" + d.getTime());
       
        record.add(Record.logLevelField, logLevel);
        record.add(Record.classField, className);
       
        int firstSep = body.indexOf(" ");
        String logEntryType = body.substring(0,firstSep);
        record.add(jobEntryTypeField, logEntryType);
       
        internalMatcher.reset(body.substring(firstSep));
        String fieldName = null;
        String fieldValue = null;
        while (internalMatcher.matches())
        {
          fieldName = internalMatcher.group(1).trim();
          fieldValue = internalMatcher.group(2).trim();
          record.add(fieldName, fieldValue);
          log.info("JobLogHistoryProcessor Add field: [" + fieldName + "][" + fieldValue +"]" );
          internalMatcher.reset(internalMatcher.group(3));
        }

        buildGenericRecord(record, body, d.getTime(), JobLogHistoryProcessor.recordType);
View Full Code Here
155
156
157
158
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
192
193
      try
      {
        Date d = sdf.parse(matcher.group(1).trim());
        System.out.println(d);
        ChukwaRecord record = new ChukwaRecord();

        String logLevel = matcher.group(2);
        String className = matcher.group(3);
        String body = matcher.group(4);

        System.out.println(matcher.group(1));
        System.out.println(matcher.group(2));
        System.out.println(matcher.group(3));
        System.out.println(matcher.group(4));
       
        System.out.println(body);
       
        record.add(Record.logLevelField, logLevel);
        record.add(Record.classField, className);
       
        int firstSep = body.indexOf(" ");
        System.out.println(firstSep);
        String logEntryType = body.substring(0,firstSep);
        record.add(jobEntryTypeField, logEntryType);
       
        internalMatcher.reset(body.substring(firstSep));
        String fieldName = null;
        String fieldValue = null;
        while (internalMatcher.matches())
        {
          fieldName = internalMatcher.group(1).trim();
          fieldValue = internalMatcher.group(2).trim();
          record.add(fieldName, fieldValue);
          System.out.println("JobLogHistoryProcessor Add field: [" + fieldName + "][" + fieldValue +"]" );
          internalMatcher.reset(internalMatcher.group(3));
        }

       
View Full Code Here
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
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
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
    int totalDownNode = 0;
   
    String logLevel = null;
    String className = null;
    String body = null;
    ChukwaRecord record = null;
   
    matcher.reset(recordEntry);
    if (matcher.matches())
    {
      log.info("PbsNodeProcessor Matches");
     
      try
      {
        Date d = sdf.parse( matcher.group(0).trim());
       
        logLevel = matcher.group(2);
        className = matcher.group(3);
        body = matcher.group(4);
       
        // Raw PBS output
        /*
          record = new ChukwaRecord();
        buildGenericRecord(record,recordEntry,d.getTime(),rawPBSRecordType);
       
        //TODO create a more specific key structure
        // part of ChukwaArchiveKey + record index if needed
        key.set("" + d.getTime());
       
        record.add(Record.logLevelField, logLevel);
        record.add(Record.classField, className);
        record.add(Record.bodyField, body);
        //Output PbsNode record for all machines (raw information)
        output.collect(key, record);
        log.info("PbsNodeProcessor output 1 record");
        */
       
        // TODO if we're not saving the PBS' output, we don't need to parse and save it ...
        String[] lines = recordEntry.split("\n");
        while (i < lines.length)
        {
          while ((i < lines.length) && (lines[i].trim().length() > 0))
          {
            sb.append(lines[i].trim()).append("\n");
            i++;
          }

           if ( (i< lines.length) && (lines[i].trim().length() > 0) )
           {
           throw new PbsInvalidEntry(recordEntry);
           }

          // Empty line
          i++;

          if (sb.length() > 0)
          {
            body =  sb.toString();
            // Process all entries for a machine
            System.out.println("=========>>> Record [" + body+ "]");
           
            record = new ChukwaRecord();
            buildGenericRecord(record,body,d.getTime(),machinePBSRecordType);
           
            // TODO Change the key information
            key.set("" + d.getTime());
            record.add(Record.logLevelField, logLevel);
            record.add(Record.classField, className);
            record.add(Record.bodyField, body);
           
            parsePbsRecord(body, record);
           
            //Output PbsNode record for 1 machine
            output.collect(key, record);
            log.info("PbsNodeProcessor output 1 sub-record");
            //compute Node Activity information
            nodeActivityStatus = record.getValue("state");
            if (nodeActivityStatus != null)
            {
              if (nodeActivityStatus.equals("free"))
              {
                totalFreeNode ++;
View Full Code Here
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
  public void parse(String line, OutputCollector<Text, ChukwaRecord> output,
      Reporter reporter)
  {
    log.info("record: [" + line + "] type[" + chunk.getDataType() + "]");
   
    ChukwaRecord record = new ChukwaRecord();
   
    matcher.reset(line);
    if (matcher.matches())
    {
      try
      {
        Date d = sdf.parse( matcher.group(0).trim());
        if (this.chunk.getStreamName().indexOf("datanode") > 0)
        {
          buildGenericRecord(record,line,d.getTime(),dataNodeType);
        }
        else if (this.chunk.getStreamName().indexOf("namenode") > 0)
        {
          buildGenericRecord(record,line,d.getTime(),nameNodeType);
        }
        else
        {
          buildGenericRecord(record,line,d.getTime(),recordType);
        }
       
        key.set("" + d.getTime());
        record.add(Record.logLevelField, "" +matcher.group(2));
        record.add(Record.classField, "" +matcher.group(3));
        record.add(Record.bodyField, "" +matcher.group(4));
        output.collect(key, record);
      }
      catch (ParseException e)
      {
        e.printStackTrace();
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
  }

  public Tuple getNext() throws IOException
  {
    ChukwaRecord record = new ChukwaRecord();
    if (!reader.next(reader.createKey(), record))
      return null;

    Tuple ret = tf.newTuple(2);
    try
    {
      ret.set(0, new Long(record.getTime()));

      HashMap<Object, Object> pigMapFields = new HashMap<Object, Object>();
      TreeMap<String, Buffer> mapFields = record.getMapFields();

      if (mapFields != null)
      {
        for (String key : mapFields.keySet())
        {
          pigMapFields.put(key, new DataByteArray(record.getValue(key).getBytes()));
        }
      }
      ret.set(1, pigMapFields);

    } catch (ExecException e)
View Full Code Here
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
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
283
284
285
    {

      ChukwaRecordKey key = new ChukwaRecordKey();
    

      ChukwaRecord record = new ChukwaRecord();
      record.setTime(System.currentTimeMillis());
      int inputSize = f.size();     
      for(int i=0;i<inputSize;i++) {
        Object field = f.get(i);
       
        if (field == null) {
          continue;
        }
       
        if (i == this.pkFieldIndex) {
          pk = field.toString();
          continue;
        } else if ( i == this.sourceFieldIndex) {
          source = field.toString();
          continue;
         
        }else if ( i== this.recordTypeFieldIndex) {
          recordType = field.toString();
          continue;
         
        }else if ( i== this.applicationFieldIndex) {
          application = field.toString();
          continue;
         
        } else if ( i== this.clusterNameFieldIndex) {
          clusterName = field.toString();
          continue;
         
        }else if (i == this.timestampFieldIndex) {
         
          timestamp = Long.parseLong(field.toString());
          record.setTime(timestamp);

          synchronized (calendar)
          {
            calendar.setTimeInMillis(timestamp);
            calendar.set(Calendar.MINUTE, 0);
            calendar.set(Calendar.SECOND, 0);
            calendar.set(Calendar.MILLISECOND, 0);
            timePartition = calendar.getTimeInMillis();
          }
          record.setTime(Long.parseLong(field.toString()));
          continue;
         
        }  else if (field instanceof Map) {
          Map<Object, Object> m = (Map<Object, Object>)field;
          for(Object o: m.keySet()) {
            record.add(o.toString(),m.get(o).toString());
          }
          continue;
        } else {
          if (i <fields.length ) {
            record.add(fields[i],field.toString());
          } else {
            record.add("field-"+i,field.toString());
          }
         
          continue;
        }
      }

      record.add(Record.tagsField, " cluster=\"" + clusterName.trim() + "\" ");
      record.add(Record.sourceField, source);
      record.add(Record.applicationField, application);
      key.setKey("" + timePartition + "/" + pk + "/" + timestamp);
      key.setReduceType(recordType);
     
      writer.append(key, record);
    } catch (ExecException e)
View Full Code Here
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
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
      String jobID = "";
        matcher = jobPattern.matcher(capp);
        if(matcher.matches()) {
          jobID=matcher.group(2);
        }
        ChukwaRecord record = new ChukwaRecord();
        ChukwaRecord jobConfRecord = new ChukwaRecord();
      DocumentBuilderFactory docBuilderFactory
        = DocumentBuilderFactory.newInstance();
      //ignore all comments inside the xml file
      docBuilderFactory.setIgnoringComments(true);
      try {
          DocumentBuilder builder = docBuilderFactory.newDocumentBuilder();
          Document doc = null;
          String fileName = "test_"+randomNumber.nextInt();
          File tmp = new File(fileName);
          FileOutputStream out = new FileOutputStream(tmp);
          out.write(recordEntry.getBytes());
          out.close();
        doc = builder.parse(fileName);
        Element root = doc.getDocumentElement();
        if (!"configuration".equals(root.getTagName()))
            log.fatal("bad conf file: top-level element not <configuration>");
        NodeList props = root.getChildNodes();
            JSONObject json = new JSONObject();
            String queue = "default";
   
        for (int i = 0; i < props.getLength(); i++) {
            Node propNode = props.item(i);
            if (!(propNode instanceof Element))
                continue;
            Element prop = (Element)propNode;
            if (!"property".equals(prop.getTagName()))
                log.warn("bad conf file: element not <property>");
            NodeList fields = prop.getChildNodes();
            String attr = null;
            String value = null;
            boolean finalParameter = false;
            for (int j = 0; j < fields.getLength(); j++) {
                Node fieldNode = fields.item(j);
                if (!(fieldNode instanceof Element))
                    continue;
                Element field = (Element)fieldNode;
                if ("name".equals(field.getTagName()) && field.hasChildNodes())
                    attr = ((Text)field.getFirstChild()).getData().trim();
                if ("value".equals(field.getTagName()) && field.hasChildNodes())
                    value = ((Text)field.getFirstChild()).getData();
                if ("final".equals(field.getTagName()) && field.hasChildNodes())
                    finalParameter = "true".equals(((Text)field.getFirstChild()).getData());
            }
           
            // Ignore this parameter if it has already been marked as 'final'
            if (attr != null && value != null) {
                json.put(attr, value);
                if(attr.intern()=="mapred.job.queue.name".intern()) {
                    queue=value;
                }
                jobConfRecord.add("job_conf." + attr, value);
            }
        }
        record.add("JOBCONF-JSON", json.toString());
        record.add("mapred.job.queue.name", queue);
        record.add("JOBID", "job_" + jobID);
        buildGenericRecord(record, null, time, "JobData");
        calendar.setTimeInMillis(time);
        calendar.set(Calendar.MINUTE, 0);
        calendar.set(Calendar.SECOND, 0);
        calendar.set(Calendar.MILLISECOND, 0);
        key.setKey("" + calendar.getTimeInMillis() + "/job_" + jobID + "/" + time);
        output.collect(key, record);

        jobConfRecord.add("JOBID", "job_" + jobID);
        buildGenericRecord(jobConfRecord, null, time, "JobConfData");
        output.collect(key, jobConfRecord);
           
        tmp.delete();
      } catch(Exception e) {
View Full Code Here
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
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
      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
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
      if (stmt.execute(query)) {
        rs = stmt.getResultSet();
        ResultSetMetaData rmeta = rs.getMetaData();
        int col = rmeta.getColumnCount();
        while (rs.next()) {
          ChukwaRecord event = new ChukwaRecord();
          String cell = "";
          long timestamp = 0;

          for (int i = 1; i < col; i++) {
            String value = rs.getString(i);
            if (value != null) {
              cell = cell + " " + rmeta.getColumnName(i) + ":" + value;
            }
            if (rmeta.getColumnName(i).equals(timeField)) {
              timestamp = rs.getLong(i);
              event.setTime(timestamp);
            }
          }
          boolean isValid = false;
          if (filter == null || filter.equals("")) {
            isValid = true;
          } else if (cell.indexOf(filter) > 0) {
            isValid = true;
          }
          if (!isValid) {
            continue;
          }

          event.add(Record.bodyField, cell);
          event.add(Record.sourceField, cluster + "." + dataSource);
          if (records.containsKey(timestamp)) {
            records.get(timestamp).add(event);
          } else {
            List<Record> list = new LinkedList<Record>();
            list.add(event);
            records.put(event.getTime(), list);
          }
        }
      }
    } catch (SQLException e) {
      e.printStackTrace();
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.