big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

YARN-3879 [Storage implementation] Create HDFS backing storage implementation for ATS reads. Contributed by Abhishek Modi.

Browse Source
This commit is contained in:
Vrushali C 2018-10-11 21:13:52 -07:00
parent 74db993a61
commit bca928d3c7
1 changed files with 50 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-timelineservice/src/main/java/org/apache/hadoop/yarn/server/timelineservice/storage/FileSystemTimelineReaderImpl.java
View File

@ -20,7 +20,6 @@
import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStreamReader;
@ -41,6 +40,11 @@
import org.apache.commons.csv.CSVParser;
import org.apache.commons.csv.CSVRecord;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.RemoteIterator;
import org.apache.hadoop.service.AbstractService;
import org.apache.hadoop.yarn.api.records.timelineservice.TimelineEntity;
import org.apache.hadoop.yarn.api.records.timelineservice.TimelineEvent;
@ -68,7 +72,9 @@ public class FileSystemTimelineReaderImpl extends AbstractService
private static final Logger LOG =
LoggerFactory.getLogger(FileSystemTimelineReaderImpl.class);
private String rootPath;
private FileSystem fs;
private Path rootPath;
private Path entitiesPath;
private static final String ENTITIES_DIR = "entities";
/** Default extension for output files. */
@ -94,7 +100,7 @@ public FileSystemTimelineReaderImpl() {
@VisibleForTesting
String getRootPath() {
return rootPath;
return rootPath.toString();
}
private static ObjectMapper mapper;
@ -162,12 +168,12 @@ private String getFlowRunPath(String userId, String clusterId,
if (clusterId == null || appId == null) {
throw new IOException("Unable to get flow info");
}
String appFlowMappingFile = rootPath + File.separator + ENTITIES_DIR +
File.separator + clusterId + File.separator + APP_FLOW_MAPPING_FILE;
Path clusterIdPath = new Path(entitiesPath, clusterId);
Path appFlowMappingFilePath = new Path(clusterIdPath,
APP_FLOW_MAPPING_FILE);
try (BufferedReader reader =
new BufferedReader(new InputStreamReader(
new FileInputStream(
appFlowMappingFile), Charset.forName("UTF-8")));
fs.open(appFlowMappingFilePath), Charset.forName("UTF-8")));
CSVParser parser = new CSVParser(reader, csvFormat)) {
for (CSVRecord record : parser.getRecords()) {
if (record.size() < 4) {
@ -266,7 +272,7 @@ private static TimelineEntity readEntityFromFile(BufferedReader reader)
return entity;
}
private Set<TimelineEntity> getEntities(File dir, String entityType,
private Set<TimelineEntity> getEntities(Path dir, String entityType,
TimelineEntityFilters filters, TimelineDataToRetrieve dataToRetrieve)
throws IOException {
// First sort the selected entities based on created/start time.
@ -280,15 +286,18 @@ public int compare(Long l1, Long l2) {
}
);
if (dir != null) {
File[] files = dir.listFiles();
if (files != null) {
for (File entityFile : files) {
RemoteIterator<LocatedFileStatus> fileStatuses = fs.listFiles(dir,
false);
if (fileStatuses != null) {
while (fileStatuses.hasNext()) {
LocatedFileStatus locatedFileStatus = fileStatuses.next();
Path entityFile = locatedFileStatus.getPath();
if (!entityFile.getName()
.contains(TIMELINE_SERVICE_STORAGE_EXTENSION)) {
continue;
}
try (BufferedReader reader = new BufferedReader(
new InputStreamReader(new FileInputStream(entityFile),
new InputStreamReader(fs.open(entityFile),
Charset.forName("UTF-8")))) {
TimelineEntity entity = readEntityFromFile(reader);
if (!entity.getType().equals(entityType)) {
@ -366,25 +375,30 @@ public int compare(Long l1, Long l2) {
@Override
public void serviceInit(Configuration conf) throws Exception {
rootPath = conf.get(TIMELINE_SERVICE_STORAGE_DIR_ROOT,
String outputRoot = conf.get(TIMELINE_SERVICE_STORAGE_DIR_ROOT,
conf.get("hadoop.tmp.dir") + File.separator + STORAGE_DIR_ROOT);
rootPath = new Path(outputRoot);
entitiesPath = new Path(rootPath, ENTITIES_DIR);
fs = rootPath.getFileSystem(conf);
super.serviceInit(conf);
}
@Override
public TimelineEntity getEntity(TimelineReaderContext context,
TimelineDataToRetrieve dataToRetrieve) throws IOException {
String flowRunPath = getFlowRunPath(context.getUserId(),
String flowRunPathStr = getFlowRunPath(context.getUserId(),
context.getClusterId(), context.getFlowName(), context.getFlowRunId(),
context.getAppId());
File dir = new File(new File(rootPath, ENTITIES_DIR),
context.getClusterId() + File.separator + flowRunPath + File.separator +
context.getAppId() + File.separator + context.getEntityType());
File entityFile = new File(
dir, context.getEntityId() + TIMELINE_SERVICE_STORAGE_EXTENSION);
Path clusterIdPath = new Path(entitiesPath, context.getClusterId());
Path flowRunPath = new Path(clusterIdPath, flowRunPathStr);
Path appIdPath = new Path(flowRunPath, context.getAppId());
Path entityTypePath = new Path(appIdPath, context.getEntityType());
Path entityFilePath = new Path(entityTypePath,
context.getEntityId() + TIMELINE_SERVICE_STORAGE_EXTENSION);
try (BufferedReader reader =
new BufferedReader(new InputStreamReader(
new FileInputStream(entityFile), Charset.forName("UTF-8")))) {
fs.open(entityFilePath), Charset.forName("UTF-8")))) {
TimelineEntity entity = readEntityFromFile(reader);
return createEntityToBeReturned(
entity, dataToRetrieve.getFieldsToRetrieve());
@ -399,32 +413,31 @@ public TimelineEntity getEntity(TimelineReaderContext context,
public Set<TimelineEntity> getEntities(TimelineReaderContext context,
TimelineEntityFilters filters, TimelineDataToRetrieve dataToRetrieve)
throws IOException {
String flowRunPath = getFlowRunPath(context.getUserId(),
String flowRunPathStr = getFlowRunPath(context.getUserId(),
context.getClusterId(), context.getFlowName(), context.getFlowRunId(),
context.getAppId());
File dir =
new File(new File(rootPath, ENTITIES_DIR),
context.getClusterId() + File.separator + flowRunPath +
File.separator + context.getAppId() + File.separator +
context.getEntityType());
return getEntities(dir, context.getEntityType(), filters, dataToRetrieve);
Path clusterIdPath = new Path(entitiesPath, context.getClusterId());
Path flowRunPath = new Path(clusterIdPath, flowRunPathStr);
Path appIdPath = new Path(flowRunPath, context.getAppId());
Path entityTypePath = new Path(appIdPath, context.getEntityType());
return getEntities(entityTypePath, context.getEntityType(), filters,
dataToRetrieve);
}
@Override public Set<String> getEntityTypes(TimelineReaderContext context)
throws IOException {
Set<String> result = new TreeSet<>();
String flowRunPath = getFlowRunPath(context.getUserId(),
String flowRunPathStr = getFlowRunPath(context.getUserId(),
context.getClusterId(), context.getFlowName(), context.getFlowRunId(),
context.getAppId());
File dir = new File(new File(rootPath, ENTITIES_DIR),
context.getClusterId() + File.separator + flowRunPath
+ File.separator + context.getAppId());
File[] fileList = dir.listFiles();
if (fileList != null) {
for (File f : fileList) {
if (f.isDirectory()) {
result.add(f.getName());
}
Path clusterIdPath = new Path(entitiesPath, context.getClusterId());
Path flowRunPath = new Path(clusterIdPath, flowRunPathStr);
Path appIdPath = new Path(flowRunPath, context.getAppId());
FileStatus[] fileStatuses = fs.listStatus(appIdPath);
for (FileStatus fileStatus : fileStatuses) {
if (fileStatus.isDirectory()) {
result.add(fileStatus.getPath().getName());
}
}
return result;