Explicitly release resources in LogFileReader and TestHoodieClientBase
This commit is contained in:
committed by
vinoth chandar
parent
2728f96505
commit
5cb28e7b1f
@@ -30,6 +30,7 @@ import com.uber.hoodie.common.table.log.block.HoodieDeleteBlock;
|
||||
import com.uber.hoodie.common.table.log.block.HoodieLogBlock;
|
||||
import com.uber.hoodie.common.util.SpillableMapUtils;
|
||||
import com.uber.hoodie.exception.HoodieIOException;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayDeque;
|
||||
import java.util.Arrays;
|
||||
import java.util.Deque;
|
||||
@@ -115,9 +116,10 @@ public abstract class AbstractHoodieLogRecordScanner {
|
||||
* Scan Log files
|
||||
*/
|
||||
public void scan() {
|
||||
HoodieLogFormatReader logFormatReaderWrapper = null;
|
||||
try {
|
||||
// iterate over the paths
|
||||
HoodieLogFormatReader logFormatReaderWrapper =
|
||||
logFormatReaderWrapper =
|
||||
new HoodieLogFormatReader(fs,
|
||||
logFilePaths.stream().map(logFile -> new HoodieLogFile(new Path(logFile)))
|
||||
.collect(Collectors.toList()), readerSchema, readBlocksLazily, reverseReader, bufferSize);
|
||||
@@ -239,6 +241,15 @@ public abstract class AbstractHoodieLogRecordScanner {
|
||||
} catch (Exception e) {
|
||||
log.error("Got exception when reading log file", e);
|
||||
throw new HoodieIOException("IOException when reading log file ");
|
||||
} finally {
|
||||
try {
|
||||
if (null != logFormatReaderWrapper) {
|
||||
logFormatReaderWrapper.close();
|
||||
}
|
||||
} catch (IOException ioe) {
|
||||
// Eat exception as we do not want to mask the original exception that can happen
|
||||
log.error("Unable to close log format reader", ioe);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -62,6 +62,7 @@ class HoodieLogFileReader implements HoodieLogFormat.Reader {
|
||||
private long reverseLogFilePosition;
|
||||
private long lastReverseLogFilePosition;
|
||||
private boolean reverseReader;
|
||||
private boolean closed = false;
|
||||
|
||||
HoodieLogFileReader(FileSystem fs, HoodieLogFile logFile, Schema readerSchema, int bufferSize,
|
||||
boolean readBlockLazily, boolean reverseReader) throws IOException {
|
||||
@@ -95,13 +96,13 @@ class HoodieLogFileReader implements HoodieLogFormat.Reader {
|
||||
}
|
||||
|
||||
/**
|
||||
* Close the inputstream when the JVM exits
|
||||
* Close the inputstream if not closed when the JVM exits
|
||||
*/
|
||||
private void addShutDownHook() {
|
||||
Runtime.getRuntime().addShutdownHook(new Thread() {
|
||||
public void run() {
|
||||
try {
|
||||
inputStream.close();
|
||||
close();
|
||||
} catch (Exception e) {
|
||||
log.warn("unable to close input stream for log file " + logFile, e);
|
||||
// fail silently for any sort of exception
|
||||
@@ -277,7 +278,10 @@ class HoodieLogFileReader implements HoodieLogFormat.Reader {
|
||||
|
||||
@Override
|
||||
public void close() throws IOException {
|
||||
this.inputStream.close();
|
||||
if (!closed) {
|
||||
this.inputStream.close();
|
||||
closed = true;
|
||||
}
|
||||
}
|
||||
|
||||
@Override
|
||||
|
||||
@@ -20,6 +20,7 @@ import com.uber.hoodie.common.model.HoodieLogFile;
|
||||
import com.uber.hoodie.common.table.log.block.HoodieLogBlock;
|
||||
import com.uber.hoodie.exception.HoodieIOException;
|
||||
import java.io.IOException;
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
import org.apache.avro.Schema;
|
||||
import org.apache.hadoop.fs.FileSystem;
|
||||
@@ -29,6 +30,8 @@ import org.apache.log4j.Logger;
|
||||
public class HoodieLogFormatReader implements HoodieLogFormat.Reader {
|
||||
|
||||
private final List<HoodieLogFile> logFiles;
|
||||
// Readers for previously scanned log-files that are still open
|
||||
private final List<HoodieLogFileReader> prevReadersInOpenState;
|
||||
private HoodieLogFileReader currentReader;
|
||||
private final FileSystem fs;
|
||||
private final Schema readerSchema;
|
||||
@@ -46,6 +49,7 @@ public class HoodieLogFormatReader implements HoodieLogFormat.Reader {
|
||||
this.readBlocksLazily = readBlocksLazily;
|
||||
this.reverseLogReader = reverseLogReader;
|
||||
this.bufferSize = bufferSize;
|
||||
this.prevReadersInOpenState = new ArrayList<>();
|
||||
if (logFiles.size() > 0) {
|
||||
HoodieLogFile nextLogFile = logFiles.remove(0);
|
||||
this.currentReader = new HoodieLogFileReader(fs, nextLogFile, readerSchema, bufferSize, readBlocksLazily, false);
|
||||
@@ -53,7 +57,20 @@ public class HoodieLogFormatReader implements HoodieLogFormat.Reader {
|
||||
}
|
||||
|
||||
@Override
|
||||
/**
|
||||
* Note : In lazy mode, clients must ensure close() should be called only after processing
|
||||
* all log-blocks as the underlying inputstream will be closed.
|
||||
* TODO: We can introduce invalidate() API at HoodieLogBlock and this object can call invalidate on
|
||||
* all returned log-blocks so that we check this scenario specifically in HoodieLogBlock
|
||||
*/
|
||||
public void close() throws IOException {
|
||||
|
||||
for (HoodieLogFileReader reader : prevReadersInOpenState) {
|
||||
reader.close();
|
||||
}
|
||||
|
||||
prevReadersInOpenState.clear();
|
||||
|
||||
if (currentReader != null) {
|
||||
currentReader.close();
|
||||
}
|
||||
@@ -69,6 +86,12 @@ public class HoodieLogFormatReader implements HoodieLogFormat.Reader {
|
||||
} else if (logFiles.size() > 0) {
|
||||
try {
|
||||
HoodieLogFile nextLogFile = logFiles.remove(0);
|
||||
// First close previous reader only if readBlockLazily is true
|
||||
if (!readBlocksLazily) {
|
||||
this.currentReader.close();
|
||||
} else {
|
||||
this.prevReadersInOpenState.add(currentReader);
|
||||
}
|
||||
this.currentReader = new HoodieLogFileReader(fs, nextLogFile, readerSchema, bufferSize, readBlocksLazily,
|
||||
false);
|
||||
} catch (IOException io) {
|
||||
|
||||
@@ -304,6 +304,7 @@ public class HoodieLogFormatTest {
|
||||
dataBlockRead.getRecords().size());
|
||||
assertEquals("Both records lists should be the same. (ordering guaranteed)", copyOfRecords,
|
||||
dataBlockRead.getRecords());
|
||||
reader.close();
|
||||
}
|
||||
|
||||
@SuppressWarnings("unchecked")
|
||||
@@ -370,6 +371,7 @@ public class HoodieLogFormatTest {
|
||||
dataBlockRead.getRecords().size());
|
||||
assertEquals("Both records lists should be the same. (ordering guaranteed)", copyOfRecords3,
|
||||
dataBlockRead.getRecords());
|
||||
reader.close();
|
||||
}
|
||||
|
||||
@SuppressWarnings("unchecked")
|
||||
@@ -454,6 +456,8 @@ public class HoodieLogFormatTest {
|
||||
//assertEquals("", "something-random", new String(corruptBlock.getCorruptedBytes()));
|
||||
assertFalse("There should be no more block left", reader.hasNext());
|
||||
|
||||
reader.close();
|
||||
|
||||
// Simulate another failure back to back
|
||||
outputStream = fs.append(writer.getLogFile().getPath());
|
||||
// create a block with
|
||||
@@ -493,6 +497,7 @@ public class HoodieLogFormatTest {
|
||||
assertTrue("We should get the last block next", reader.hasNext());
|
||||
reader.next();
|
||||
assertFalse("We should have no more blocks left", reader.hasNext());
|
||||
reader.close();
|
||||
}
|
||||
|
||||
|
||||
@@ -1097,7 +1102,7 @@ public class HoodieLogFormatTest {
|
||||
assertEquals(block.getBlockType(), HoodieLogBlockType.AVRO_DATA_BLOCK);
|
||||
dBlock = (HoodieAvroDataBlock) block;
|
||||
assertEquals(dBlock.getRecords().size(), 100);
|
||||
|
||||
reader.close();
|
||||
}
|
||||
|
||||
@SuppressWarnings("unchecked")
|
||||
@@ -1167,6 +1172,7 @@ public class HoodieLogFormatTest {
|
||||
dataBlockRead.getRecords());
|
||||
|
||||
assertFalse(reader.hasPrev());
|
||||
reader.close();
|
||||
}
|
||||
|
||||
@Test
|
||||
@@ -1224,6 +1230,7 @@ public class HoodieLogFormatTest {
|
||||
e.printStackTrace();
|
||||
// We should have corrupted block
|
||||
}
|
||||
reader.close();
|
||||
}
|
||||
|
||||
@SuppressWarnings("unchecked")
|
||||
@@ -1283,5 +1290,6 @@ public class HoodieLogFormatTest {
|
||||
dataBlockRead.getRecords());
|
||||
|
||||
assertFalse(reader.hasPrev());
|
||||
reader.close();
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user