Revision 48800
Added by Claudio Atzori over 6 years ago
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/utils/HadoopUtils.java | ||
|---|---|---|
| 1 | 1 |
package eu.dnetlib.data.hadoop.utils; |
| 2 | 2 |
|
| 3 |
import java.util.Map.Entry; |
|
| 4 |
import java.util.Set; |
|
| 5 |
|
|
| 6 |
import org.apache.hadoop.hbase.HColumnDescriptor; |
|
| 7 |
import org.apache.hadoop.hbase.HTableDescriptor; |
|
| 8 |
|
|
| 9 |
import com.google.common.base.Function; |
|
| 10 |
import com.google.common.base.Predicate; |
|
| 11 |
|
|
| 12 | 3 |
import eu.dnetlib.data.hadoop.HadoopJob; |
| 13 |
import eu.dnetlib.data.hadoop.config.ClusterName; |
|
| 14 | 4 |
import eu.dnetlib.data.hadoop.rmi.HadoopJobDescriptor; |
| 15 | 5 |
import eu.dnetlib.data.hadoop.rmi.HadoopServiceException; |
| 6 |
import org.apache.commons.logging.Log; |
|
| 7 |
import org.apache.commons.logging.LogFactory; |
|
| 16 | 8 |
|
| 17 | 9 |
public class HadoopUtils {
|
| 18 | 10 |
|
| 19 |
public static Function<HTableDescriptor, String> tableName() {
|
|
| 20 |
return new Function<HTableDescriptor, String>() {
|
|
| 11 |
private static final Log log = LogFactory.getLog(HadoopUtils.class); |
|
| 21 | 12 |
|
| 22 |
@Override |
|
| 23 |
public String apply(final HTableDescriptor d) {
|
|
| 24 |
return d.getNameAsString(); |
|
| 13 |
public static java.util.function.Function<HadoopJob, HadoopJobDescriptor> asDescriptor() {
|
|
| 14 |
return d -> {
|
|
| 15 |
try {
|
|
| 16 |
return d.asDescriptor(); |
|
| 17 |
} catch (HadoopServiceException e) {
|
|
| 18 |
log.warn(e); |
|
| 19 |
return null; |
|
| 25 | 20 |
} |
| 26 | 21 |
}; |
| 27 | 22 |
} |
| 28 | 23 |
|
| 29 |
public static Function<HColumnDescriptor, String> columnName() {
|
|
| 30 |
return new Function<HColumnDescriptor, String>() {
|
|
| 31 |
|
|
| 32 |
@Override |
|
| 33 |
public String apply(final HColumnDescriptor d) {
|
|
| 34 |
return d.getNameAsString(); |
|
| 35 |
} |
|
| 36 |
}; |
|
| 37 |
} |
|
| 38 |
|
|
| 39 |
public static Predicate<String> columnPredicate(final Set<String> cols) {
|
|
| 40 |
return new HadoopUtils().getSetPredicate(cols); |
|
| 41 |
} |
|
| 42 |
|
|
| 43 |
public SetPredicate getSetPredicate(final Set<String> set) {
|
|
| 44 |
return new SetPredicate(set); |
|
| 45 |
} |
|
| 46 |
|
|
| 47 |
class SetPredicate implements Predicate<String> {
|
|
| 48 |
|
|
| 49 |
private final Set<String> set; |
|
| 50 |
|
|
| 51 |
public SetPredicate(final Set<String> set) {
|
|
| 52 |
this.set = set; |
|
| 53 |
} |
|
| 54 |
|
|
| 55 |
@Override |
|
| 56 |
public boolean apply(final String s) {
|
|
| 57 |
return !set.contains(s); |
|
| 58 |
} |
|
| 59 |
} |
|
| 60 |
|
|
| 61 |
public static Function<Entry<String, HadoopJob>, HadoopJobDescriptor> hadoopJobDescriptor() {
|
|
| 62 |
return new Function<Entry<String, HadoopJob>, HadoopJobDescriptor>() {
|
|
| 63 |
|
|
| 64 |
@Override |
|
| 65 |
public HadoopJobDescriptor apply(final Entry<String, HadoopJob> e) {
|
|
| 66 |
try {
|
|
| 67 |
return e.getValue().asDescriptor(); |
|
| 68 |
} catch (HadoopServiceException e1) {
|
|
| 69 |
return null; |
|
| 70 |
} |
|
| 71 |
} |
|
| 72 |
}; |
|
| 73 |
} |
|
| 74 |
|
|
| 75 |
public static Predicate<HadoopJob> filterByCluster(final ClusterName clusterName) {
|
|
| 76 |
return new Predicate<HadoopJob>() {
|
|
| 77 |
|
|
| 78 |
@Override |
|
| 79 |
public boolean apply(final HadoopJob job) {
|
|
| 80 |
return job.getClusterName().equals(clusterName); |
|
| 81 |
} |
|
| 82 |
|
|
| 83 |
}; |
|
| 84 |
} |
|
| 85 |
|
|
| 86 | 24 |
} |
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/JobRegistry.java | ||
|---|---|---|
| 2 | 2 |
|
| 3 | 3 |
import java.util.Date; |
| 4 | 4 |
import java.util.List; |
| 5 |
import java.util.Map; |
|
| 6 | 5 |
import java.util.Map.Entry; |
| 6 |
import java.util.Objects; |
|
| 7 |
import java.util.stream.Collectors; |
|
| 7 | 8 |
|
| 8 |
import org.apache.commons.logging.Log; |
|
| 9 |
import org.apache.commons.logging.LogFactory; |
|
| 10 |
import org.springframework.beans.factory.annotation.Required; |
|
| 11 |
|
|
| 12 | 9 |
import com.google.common.collect.BiMap; |
| 13 | 10 |
import com.google.common.collect.HashBiMap; |
| 14 |
import com.google.common.collect.Iterables; |
|
| 15 |
import com.google.common.collect.Lists; |
|
| 16 |
import com.google.common.collect.Maps; |
|
| 17 |
|
|
| 18 | 11 |
import eu.dnetlib.data.hadoop.HadoopJob.Status; |
| 19 | 12 |
import eu.dnetlib.data.hadoop.config.ClusterName; |
| 20 | 13 |
import eu.dnetlib.data.hadoop.rmi.HadoopJobDescriptor; |
| 21 | 14 |
import eu.dnetlib.data.hadoop.rmi.HadoopServiceException; |
| 22 | 15 |
import eu.dnetlib.data.hadoop.utils.HadoopUtils; |
| 16 |
import org.apache.commons.logging.Log; |
|
| 17 |
import org.apache.commons.logging.LogFactory; |
|
| 18 |
import org.springframework.beans.factory.annotation.Required; |
|
| 23 | 19 |
|
| 24 | 20 |
public class JobRegistry {
|
| 25 | 21 |
|
| ... | ... | |
| 84 | 80 |
} |
| 85 | 81 |
|
| 86 | 82 |
public List<HadoopJobDescriptor> listJobs(ClusterName clusterName) {
|
| 87 |
Map<String, HadoopJob> filtered = Maps.filterValues(jobs, HadoopUtils.filterByCluster(clusterName)); |
|
| 88 |
return Lists.newArrayList(Iterables.transform(filtered.entrySet(), HadoopUtils.hadoopJobDescriptor())); |
|
| 83 |
return jobs.values().stream() |
|
| 84 |
.filter(j -> clusterName.equals(j.getClusterName())) |
|
| 85 |
.map(HadoopUtils.asDescriptor()) |
|
| 86 |
.filter(Objects::nonNull) |
|
| 87 |
.collect(Collectors.toList()); |
|
| 89 | 88 |
} |
| 90 | 89 |
|
| 91 | 90 |
@Required |
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/HadoopServiceCore.java | ||
|---|---|---|
| 2 | 2 |
|
| 3 | 3 |
import java.io.IOException; |
| 4 | 4 |
import java.net.URI; |
| 5 |
import java.util.List; |
|
| 6 |
import java.util.Map; |
|
| 5 |
import java.util.*; |
|
| 7 | 6 |
import java.util.Map.Entry; |
| 8 |
import java.util.NavigableMap; |
|
| 9 |
import java.util.Set; |
|
| 7 |
import java.util.stream.Collectors; |
|
| 10 | 8 |
|
| 11 |
import com.google.common.base.Function; |
|
| 12 |
import com.google.common.collect.Iterables; |
|
| 13 | 9 |
import com.google.common.collect.Lists; |
| 14 | 10 |
import com.google.common.collect.Maps; |
| 15 | 11 |
import com.google.common.collect.Sets; |
| ... | ... | |
| 20 | 16 |
import eu.dnetlib.data.hadoop.rmi.hbase.HBaseRowDescriptor; |
| 21 | 17 |
import eu.dnetlib.data.hadoop.rmi.hbase.schema.HBaseTableDescriptor; |
| 22 | 18 |
import eu.dnetlib.data.hadoop.rmi.hbase.schema.HBaseTableRegionInfo; |
| 23 |
import eu.dnetlib.data.hadoop.utils.HadoopUtils; |
|
| 24 | 19 |
import org.apache.commons.lang.StringUtils; |
| 25 | 20 |
import org.apache.commons.logging.Log; |
| 26 | 21 |
import org.apache.commons.logging.LogFactory; |
| ... | ... | |
| 46 | 41 |
|
| 47 | 42 |
public List<String> listTables(final ClusterName clusterName) throws IOException, HadoopServiceException {
|
| 48 | 43 |
final HBaseAdmin admin = getHBaseAdmin(clusterName); |
| 49 |
return Lists.newArrayList(Iterables.transform(Lists.newArrayList(admin.listTables()), HadoopUtils.tableName())); |
|
| 44 |
return Arrays.asList(admin.listTables()) |
|
| 45 |
.stream() |
|
| 46 |
.map(HTableDescriptor::getNameAsString) |
|
| 47 |
.collect(Collectors.toList()); |
|
| 50 | 48 |
} |
| 51 | 49 |
|
| 52 | 50 |
private HBaseAdmin getHBaseAdmin(final ClusterName clusterName) throws HadoopServiceException {
|
| ... | ... | |
| 70 | 68 |
|
| 71 | 69 |
final Set<String> columns = Sets.newHashSet(); |
| 72 | 70 |
|
| 73 |
for (HColumnDescriptor hColDesc : Lists.newArrayList(desc.getColumnFamilies())) {
|
|
| 71 |
for (HColumnDescriptor hColDesc : Arrays.asList(desc.getColumnFamilies())) {
|
|
| 74 | 72 |
columns.add(hColDesc.getNameAsString()); |
| 75 | 73 |
} |
| 76 | 74 |
|
| ... | ... | |
| 94 | 92 |
public List<String> describeTable(final ClusterName clusterName, final String table) throws IOException, HadoopServiceException {
|
| 95 | 93 |
final HBaseAdmin admin = getHBaseAdmin(clusterName); |
| 96 | 94 |
final HTableDescriptor desc = admin.getTableDescriptor(table.getBytes()); |
| 97 |
|
|
| 98 |
return Lists.newArrayList(Iterables.transform(desc.getFamilies(), new Function<HColumnDescriptor, String>() {
|
|
| 99 |
|
|
| 100 |
@Override |
|
| 101 |
public String apply(final HColumnDescriptor desc) {
|
|
| 102 |
return desc.getNameAsString(); |
|
| 103 |
} |
|
| 104 |
})); |
|
| 95 |
return desc.getFamilies().stream() |
|
| 96 |
.map(d -> d.getNameAsString()) |
|
| 97 |
.collect(Collectors.toList()); |
|
| 105 | 98 |
} |
| 106 | 99 |
|
| 107 | 100 |
public void truncateTable(final ClusterName clusterName, final String table) throws IOException, HadoopServiceException {
|
| ... | ... | |
| 201 | 194 |
createTable(clusterName, table, columns); |
| 202 | 195 |
} else {
|
| 203 | 196 |
final HTableDescriptor desc = admin.getTableDescriptor(Bytes.toBytes(table)); |
| 204 |
final Set<String> foundColumns = Sets.newHashSet(Iterables.transform(Lists.newArrayList(desc.getColumnFamilies()), HadoopUtils.columnName())); |
|
| 205 | 197 |
|
| 198 |
final Set<String> foundColumns = desc.getFamilies().stream() |
|
| 199 |
.map(d -> d.getNameAsString()) |
|
| 200 |
.collect(Collectors.toCollection(HashSet::new)); |
|
| 201 |
|
|
| 206 | 202 |
log.info("ensuring columns on table " + table + ": " + columns);
|
| 207 |
final List<String> missingColumns = Lists.newArrayList(Iterables.filter(columns, HadoopUtils.columnPredicate(foundColumns)));
|
|
| 203 |
final Collection<String> missingColumns = Sets.difference(columns, foundColumns);
|
|
| 208 | 204 |
if (!missingColumns.isEmpty()) {
|
| 209 | 205 |
|
| 210 | 206 |
if (admin.isTableEnabled(table)) {
|
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/oozie/OozieJobMonitor.java | ||
|---|---|---|
| 64 | 64 |
if (status == Status.SUCCEEDED) {
|
| 65 | 65 |
// TODO set some content to return to the blackboard msg. |
| 66 | 66 |
|
| 67 |
log.info(String.format("start looking for oozie job(%s) output values: %s", getHadoopId(), workflowActions));
|
|
| 67 | 68 |
final Properties report = getReport(getOozieClient(), getHadoopId(), workflowActions); |
| 68 | 69 |
if (report != null) {
|
| 69 | 70 |
final Map<String, String> map = Maps.newHashMap(); |
| 70 | 71 |
report.forEach((k, v) -> map.put(k.toString(), v.toString())); |
| 72 |
log.info("found oozie job report, size: " + map.size());
|
|
| 71 | 73 |
getCallback().done(map); |
| 72 | 74 |
} else {
|
| 75 |
log.warn("cannot find oozie job report!");
|
|
| 73 | 76 |
getCallback().done(new HashMap<>()); |
| 74 | 77 |
} |
| 75 | 78 |
} else {
|
| ... | ... | |
| 89 | 92 |
private static Properties getReport(final OozieClient oozieClient, final String oozieJobId, final Set<String> workflowActions) throws OozieClientException, IOException {
|
| 90 | 93 |
WorkflowJob oozieJob = oozieClient.getJobInfo(oozieJobId); |
| 91 | 94 |
for (WorkflowAction currentAction : oozieJob.getActions()) {
|
| 92 |
log.debug(String.format("looking for workflo actions to report, current: '%s'", currentAction.getName()));
|
|
| 95 |
log.info(String.format("looking for workflow actions to report, current: '%s'", currentAction.getName()));
|
|
| 93 | 96 |
if (workflowActions.contains(currentAction.getName())) {
|
| 97 |
log.info(String.format("found workflow action %s", currentAction.getName()));
|
|
| 94 | 98 |
if (ACTION_TYPE_SUBWORKFLOW.equals(currentAction.getType())) {
|
| 99 |
log.info(String.format("looking for sub-workflow actions external id: %s", currentAction.getExternalId()));
|
|
| 95 | 100 |
Properties subworkflowProperties = getReport(oozieClient, currentAction.getExternalId(), workflowActions); |
| 96 | 101 |
if (subworkflowProperties != null) {
|
| 97 | 102 |
return subworkflowProperties; |
| ... | ... | |
| 99 | 104 |
} else if (StringUtils.isNotBlank(currentAction.getData())) {
|
| 100 | 105 |
Properties properties = new Properties(); |
| 101 | 106 |
properties.load(IOUtils.toInputStream(currentAction.getData())); |
| 107 |
log.info(String.format("found workflow action(%s) properties size %s", currentAction.getName(), properties.values().size()));
|
|
| 102 | 108 |
return properties; |
| 103 | 109 |
} |
| 110 |
} else {
|
|
| 111 |
log.info(String.format("cannot find workflow action(%s) properties", currentAction.getName()));
|
|
| 104 | 112 |
} |
| 105 | 113 |
} |
| 106 |
|
|
| 107 | 114 |
return null; |
| 108 | 115 |
} |
| 109 | 116 |
|
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/action/AbstractSubmitAction.java | ||
|---|---|---|
| 137 | 137 |
@Override |
| 138 | 138 |
public void done(final Map<String, String> properties) {
|
| 139 | 139 |
bbJob.getParameters().putAll(properties); |
| 140 |
log.info(jobName + " completed successfully");
|
|
| 140 |
log.info(String.format("%s completed successfully, returning %s output params", jobName, properties.size()));
|
|
| 141 | 141 |
handler.done(bbJob); |
| 142 | 142 |
decrementRunningJobs(jobName); |
| 143 | 143 |
} |
| modules/dnet-hadoop-service/trunk/src/main/java/eu/dnetlib/data/hadoop/action/AbstractHadoopAction.java | ||
|---|---|---|
| 57 | 57 |
|
| 58 | 58 |
@Override |
| 59 | 59 |
public void execute(final BlackboardServerHandler handler, final BlackboardJob job) {
|
| 60 |
executor.execute(new Runnable() {
|
|
| 61 |
|
|
| 62 |
@Override |
|
| 63 |
public void run() {
|
|
| 64 |
try {
|
|
| 65 |
handler.ongoing(job); |
|
| 66 |
executeAsync(handler, job); |
|
| 67 |
} catch (final Throwable e) {
|
|
| 68 |
log.error(ExceptionUtils.getFullStackTrace(e)); |
|
| 69 |
handler.failed(job, e); |
|
| 70 |
} |
|
| 60 |
executor.execute(() -> {
|
|
| 61 |
try {
|
|
| 62 |
handler.ongoing(job); |
|
| 63 |
executeAsync(handler, job); |
|
| 64 |
} catch (final Throwable e) {
|
|
| 65 |
log.error(ExceptionUtils.getFullStackTrace(e)); |
|
| 66 |
handler.failed(job, e); |
|
| 71 | 67 |
} |
| 72 | 68 |
}); |
| 73 | 69 |
} |
Also available in: Unified diff
more logging in the oozie job monitor procedure that tries to fetch the job report. More java8 refactorings