Github user satishd commented on a diff in the pull request:
https://github.com/apache/storm/pull/936#discussion_r47612027
--- Diff:
external/storm-hdfs/src/main/java/org/apache/storm/hdfs/spout/HdfsSpout.java ---
@@ -0,0 +1,654 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ * <p/>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p/>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.storm.hdfs.spout;
+
+import java.io.IOException;
+import java.lang.reflect.Constructor;
+import java.util.Collection;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Timer;
+import java.util.TimerTask;
+import java.util.concurrent.LinkedBlockingQueue;
+import java.util.concurrent.atomic.AtomicBoolean;
+
+import backtype.storm.Config;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.fs.FileSystem;
+import org.apache.hadoop.fs.Path;
+import org.apache.storm.hdfs.common.HdfsUtils;
+import org.apache.storm.hdfs.common.security.HdfsSecurityUtil;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import backtype.storm.spout.SpoutOutputCollector;
+import backtype.storm.task.TopologyContext;
+import backtype.storm.topology.OutputFieldsDeclarer;
+import backtype.storm.topology.base.BaseRichSpout;
+import backtype.storm.tuple.Fields;
+
+public class HdfsSpout extends BaseRichSpout {
+
+ private static final Logger LOG =
LoggerFactory.getLogger(HdfsSpout.class);
+
+ private Path sourceDirPath;
+ private Path archiveDirPath;
+ private Path badFilesDirPath;
+ private Path lockDirPath;
+
+ private int commitFrequencyCount = Configs.DEFAULT_COMMIT_FREQ_COUNT;
+ private int commitFrequencySec = Configs.DEFAULT_COMMIT_FREQ_SEC;
+ private int maxDuplicates = Configs.DEFAULT_MAX_DUPLICATES;
+ private int lockTimeoutSec = Configs.DEFAULT_LOCK_TIMEOUT;
+ private boolean clocksInSync = true;
+
+ private ProgressTracker tracker = new ProgressTracker();
+
+ private FileSystem hdfs;
+ private FileReader reader;
+
+ private SpoutOutputCollector collector;
+ HashMap<MessageId, List<Object> > inflight = new HashMap<>();
+ LinkedBlockingQueue<HdfsUtils.Pair<MessageId, List<Object>>> retryList =
new LinkedBlockingQueue<>();
+
+ private String inprogress_suffix = ".inprogress";
+ private String ignoreSuffix = ".ignore";
+
+ private Configuration hdfsConfig;
+ private String readerType;
+
+ private Map conf = null;
+ private FileLock lock;
+ private String spoutId = null;
+
+ HdfsUtils.Pair<Path,FileLock.LogEntry> lastExpiredLock = null;
+ private long lastExpiredLockTime = 0;
+
+ private long tupleCounter = 0;
+ private boolean ackEnabled = false;
+ private int acksSinceLastCommit = 0 ;
+ private final AtomicBoolean commitTimeElapsed = new AtomicBoolean(false);
+ private final Timer commitTimer = new Timer();
+ private boolean fileReadCompletely = false;
+
+ private String configKey = Configs.DEFAULT_HDFS_CONFIG_KEY; // key for
hdfs kerberos configs
+
+ public HdfsSpout() {
+ }
+
+ public Path getLockDirPath() {
+ return lockDirPath;
+ }
+
+ public SpoutOutputCollector getCollector() {
+ return collector;
+ }
+
+ public HdfsSpout withConfigKey(String configKey){
+ this.configKey = configKey;
+ return this;
+ }
+
+ public void nextTuple() {
+ LOG.debug("Next Tuple");
+ // 1) First re-emit any previously failed tuples (from retryList)
+ if (!retryList.isEmpty()) {
+ LOG.debug("Sending from retry list");
+ HdfsUtils.Pair<MessageId, List<Object>> pair = retryList.remove();
+ emitData(pair.getValue(), pair.getKey());
+ return;
+ }
+
+ if( ackEnabled && tracker.size()>=maxDuplicates ) {
+ LOG.warn("Waiting for more ACKs before generating new tuples. " +
+ "Progress tracker size has reached limit {}"
+ , maxDuplicates);
+ // Don't emit anything .. allow configured spout wait strategy to
kick in
+ return;
+ }
+
+ // 2) If no failed tuples, then send tuples from hdfs
+ while (true) {
+ try {
+ // 3) Select a new file if one is not open already
+ if (reader == null) {
+ reader = pickNextFile();
+ if (reader == null) {
+ LOG.info("Currently no new files to process under : " +
sourceDirPath);
+ return;
+ }
+ }
+
+ // 4) Read record from file, emit to collector and record progress
+ List<Object> tuple = reader.next();
+ if (tuple != null) {
+ fileReadCompletely= false;
+ ++tupleCounter;
+ MessageId msgId = new MessageId(tupleCounter,
reader.getFilePath(), reader.getFileOffset());
+ emitData(tuple, msgId);
+
+ if(!ackEnabled) {
+ ++acksSinceLastCommit; // assume message is immediately acked
in non-ack mode
+ commitProgress(reader.getFileOffset());
+ } else {
+ commitProgress(tracker.getCommitPosition());
+ }
+ return;
+ } else {
+ fileReadCompletely = true;
+ if(!ackEnabled) {
+ markFileAsDone(reader.getFilePath());
+ }
+ }
+ } catch (IOException e) {
+ LOG.error("I/O Error processing at file location " +
getFileProgress(reader), e);
+ // don't emit anything .. allow configured spout wait strategy to
kick in
+ return;
+ } catch (ParseException e) {
+ LOG.error("Parsing error when processing at file location " +
getFileProgress(reader) +
+ ". Skipping remainder of file.", e);
+ markFileAsBad(reader.getFilePath());
+ // note: Unfortunately not emitting anything here due to parse
error
+ // will trigger the configured spout wait strategy which is
unnecessary
+ }
+ }
+
+ }
+
+ // will commit progress into lock file if commit threshold is reached
+ private void commitProgress(FileOffset position) {
+ if ( lock!=null && canCommitNow() ) {
+ try {
+ lock.heartbeat(position.toString());
--- End diff --
position parameter can be null, it should be handled.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
