1 |
26600
|
sandro.lab
|
package eu.dnetlib.data.hadoop.action;
|
2 |
|
|
|
3 |
|
|
import java.io.IOException;
|
4 |
|
|
import java.io.StringWriter;
|
5 |
|
|
import java.util.Map;
|
6 |
|
|
import java.util.Map.Entry;
|
7 |
|
|
|
8 |
|
|
import eu.dnetlib.data.hadoop.HadoopJob;
|
9 |
|
|
import eu.dnetlib.data.hadoop.config.ClusterName;
|
10 |
|
|
import eu.dnetlib.data.hadoop.mapred.MapreduceJobMonitor;
|
11 |
|
|
import eu.dnetlib.data.hadoop.rmi.HadoopServiceException;
|
12 |
|
|
import eu.dnetlib.data.hadoop.utils.JobProfile;
|
13 |
|
|
import eu.dnetlib.data.hadoop.utils.ScanFactory;
|
14 |
|
|
import eu.dnetlib.data.hadoop.utils.ScanProperties;
|
15 |
|
|
import eu.dnetlib.enabling.tools.blackboard.BlackboardJob;
|
16 |
|
|
import eu.dnetlib.miscutils.functional.xml.IndentXmlString;
|
17 |
49795
|
claudio.at
|
import org.apache.commons.logging.Log;
|
18 |
|
|
import org.apache.commons.logging.LogFactory;
|
19 |
|
|
import org.apache.hadoop.conf.Configuration;
|
20 |
|
|
import org.apache.hadoop.fs.Path;
|
21 |
|
|
import org.apache.hadoop.mapred.JobClient;
|
22 |
|
|
import org.apache.hadoop.mapred.JobConf;
|
23 |
|
|
import org.apache.hadoop.mapred.RunningJob;
|
24 |
26600
|
sandro.lab
|
|
25 |
|
|
public class SubmitMapreduceJobAction extends AbstractSubmitAction {
|
26 |
|
|
|
27 |
|
|
/**
|
28 |
|
|
* logger.
|
29 |
|
|
*/
|
30 |
|
|
private static final Log log = LogFactory.getLog(SubmitMapreduceJobAction.class); // NOPMD by marko on 11/24/08 5:02 PM
|
31 |
|
|
|
32 |
|
|
@Override
|
33 |
34703
|
claudio.at
|
public void submit(final JobCompletion callback, final BlackboardJob bbJob, final String jobName, final JobProfile jobProfile)
|
34 |
|
|
throws HadoopServiceException {
|
35 |
26600
|
sandro.lab
|
|
36 |
|
|
final ClusterName clusterName = ClusterName.valueOf(bbJob.getParameters().get("cluster"));
|
37 |
|
|
|
38 |
|
|
try {
|
39 |
34703
|
claudio.at
|
final JobConf jobConf = prepareJob(getConf(clusterName), jobName, jobProfile, bbJob.getParameters());
|
40 |
26600
|
sandro.lab
|
|
41 |
|
|
logJobDetails(jobConf);
|
42 |
|
|
|
43 |
49795
|
claudio.at
|
final JobClient jtClient = hadoopClientMap.getJtClient(clusterName);
|
44 |
|
|
final RunningJob runningJob = jtClient.submitJob(jobConf);
|
45 |
|
|
|
46 |
26600
|
sandro.lab
|
final String jobId = newJobId(clusterName, runningJob.getID().getId());
|
47 |
|
|
|
48 |
49795
|
claudio.at
|
jobRegistry.registerJob(
|
49 |
|
|
HadoopJob.newInstance(jobId, clusterName, jobProfile, new MapreduceJobMonitor(jtClient, runningJob, callback)));
|
50 |
26600
|
sandro.lab
|
|
51 |
34703
|
claudio.at
|
} catch (final IOException e) {
|
52 |
32818
|
claudio.at
|
throw new HadoopServiceException("error executing hadoop job: " + jobName, e);
|
53 |
26600
|
sandro.lab
|
}
|
54 |
|
|
}
|
55 |
|
|
|
56 |
31337
|
claudio.at
|
protected JobConf prepareJob(final Configuration configuration, final String jobName, final JobProfile jobProfile, final Map<String, String> parameters)
|
57 |
29516
|
claudio.at
|
throws IOException, HadoopServiceException {
|
58 |
26600
|
sandro.lab
|
|
59 |
|
|
log.info("creating job: " + jobName);
|
60 |
|
|
|
61 |
34703
|
claudio.at
|
final JobConf jobConf = new JobConf(configuration);
|
62 |
26600
|
sandro.lab
|
jobConf.setJobName(jobName);
|
63 |
|
|
jobConf.set("dnet.mapred.job.description", jobProfile.getDescription());
|
64 |
|
|
|
65 |
34703
|
claudio.at
|
final String jobLib = getJobLib(configuration, jobProfile);
|
66 |
26600
|
sandro.lab
|
jobConf.setJar(new Path(jobLib).toString());
|
67 |
|
|
|
68 |
|
|
set(jobConf, jobProfile.getJobDefinition());
|
69 |
|
|
set(jobConf, parameters);
|
70 |
|
|
|
71 |
|
|
// if we're reading from hbase table then we need also to set a scanner.
|
72 |
34703
|
claudio.at
|
final ScanProperties scanProperties = jobProfile.getScanProperties();
|
73 |
29516
|
claudio.at
|
if (jobProfile.getRequiredParams().contains("hbase.mapreduce.inputtable") && (scanProperties != null)) {
|
74 |
26600
|
sandro.lab
|
jobConf.set("hbase.mapreduce.scan", ScanFactory.getScan(scanProperties));
|
75 |
|
|
}
|
76 |
|
|
|
77 |
|
|
return jobConf;
|
78 |
|
|
}
|
79 |
|
|
|
80 |
31337
|
claudio.at
|
protected String getJobLib(final Configuration configuration, final JobProfile jobProfile) throws HadoopServiceException {
|
81 |
26600
|
sandro.lab
|
String jobLib = getDefaultLibPath(configuration.get("fs.defaultFS"));
|
82 |
|
|
|
83 |
|
|
if (jobProfile.getJobDefinition().containsKey("job.lib")) {
|
84 |
|
|
jobLib = jobProfile.getJobDefinition().get("job.lib");
|
85 |
|
|
}
|
86 |
|
|
|
87 |
29516
|
claudio.at
|
if ((jobLib == null) || jobLib.isEmpty()) throw new HadoopServiceException("job.lib must refer to an absolute or relative HDFS path");
|
88 |
26600
|
sandro.lab
|
if (!jobLib.startsWith("hdfs://")) {
|
89 |
|
|
jobLib = configuration.get("fs.defaultFS") + jobLib;
|
90 |
|
|
}
|
91 |
|
|
|
92 |
|
|
log.info("using job.lib: " + jobLib);
|
93 |
|
|
return jobLib;
|
94 |
|
|
}
|
95 |
|
|
|
96 |
32295
|
claudio.at
|
protected void set(final JobConf jobConf, final Map<String, String> properties) {
|
97 |
34703
|
claudio.at
|
for (final Entry<String, String> e : properties.entrySet()) {
|
98 |
31337
|
claudio.at
|
if (checkHdfsProperty(e)) {
|
99 |
34703
|
claudio.at
|
final String v = jobConf.get("fs.defaultFS") + e.getValue();
|
100 |
31337
|
claudio.at
|
e.setValue(v);
|
101 |
|
|
}
|
102 |
|
|
jobConf.set(e.getKey(), e.getValue());
|
103 |
|
|
}
|
104 |
|
|
}
|
105 |
|
|
|
106 |
|
|
protected void logJobDetails(final JobConf jobConf) {
|
107 |
34703
|
claudio.at
|
final StringWriter sw = new StringWriter();
|
108 |
26600
|
sandro.lab
|
try {
|
109 |
|
|
jobConf.writeXml(sw);
|
110 |
|
|
if (log.isDebugEnabled()) {
|
111 |
|
|
log.debug("\n" + IndentXmlString.apply(sw.toString()));
|
112 |
|
|
}
|
113 |
34703
|
claudio.at
|
} catch (final IOException e) {
|
114 |
26600
|
sandro.lab
|
log.warn("unable to log job details: " + jobConf.getJobName());
|
115 |
|
|
}
|
116 |
|
|
}
|
117 |
|
|
|
118 |
|
|
}
|