Skip to content

Commit

Permalink
HBASE-27637 Zero length value would cause value compressor read nothi…
Browse files Browse the repository at this point in the history
…ng and not advance the position of the InputStream (apache#5025)

Signed-off-by: Xiaolin Ha <[email protected]>
Signed-off-by: Andrew Purtell <[email protected]>
  • Loading branch information
Apache9 authored Feb 15, 2023
1 parent 8ba56cc commit 2bbe036
Show file tree
Hide file tree
Showing 4 changed files with 46 additions and 36 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@
import java.lang.reflect.InvocationTargetException;
import java.util.EnumMap;
import java.util.Map;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hbase.HBaseInterfaceAudience;
import org.apache.hadoop.hbase.io.BoundedDelegatingInputStream;
Expand Down Expand Up @@ -105,7 +106,7 @@ public byte[] compress(byte[] valueArray, int valueOffset, int valueLength) thro
return compressed;
}

public int decompress(InputStream in, int inLength, byte[] outArray, int outOffset,
public void decompress(InputStream in, int inLength, byte[] outArray, int outOffset,
int outLength) throws IOException {

// Our input is a sequence of bounded byte ranges (call them segments), with
Expand All @@ -122,11 +123,16 @@ public int decompress(InputStream in, int inLength, byte[] outArray, int outOffs
} else {
lowerIn.setDelegate(in, inLength);
}

// Caller must handle short reads.
// With current Hadoop compression codecs all 'outLength' bytes are read in here, so not
// an issue for now.
return compressedIn.read(outArray, outOffset, outLength);
if (outLength == 0) {
// The BufferedInputStream will return earlier and skip reading anything if outLength == 0,
// but in fact for an empty value, the compressed output still contains some metadata so the
// compressed size is not 0, so here we need to manually skip inLength bytes otherwise the
// next read on this stream will start from an invalid position and cause critical problem,
// such as data loss when splitting wal or replicating wal.
IOUtils.skipFully(in, inLength);
} else {
IOUtils.readFully(compressedIn, outArray, outOffset, outLength);
}
}

public void clear() {
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -382,13 +382,9 @@ private static void checkLength(int len, int max) throws IOException {
private void readCompressedValue(InputStream in, byte[] outArray, int outOffset,
int expectedLength) throws IOException {
int compressedLen = StreamUtils.readRawVarint32(in);
int read = compression.getValueCompressor().decompress(in, compressedLen, outArray, outOffset,
compression.getValueCompressor().decompress(in, compressedLen, outArray, outOffset,
expectedLength);
if (read != expectedLength) {
throw new IOException("ValueCompressor state error: short read");
}
}

}

public static class EnsureKvEncoder extends BaseEncoder {
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,8 @@
*/
package org.apache.hadoop.hbase.wal;

import static org.hamcrest.MatcherAssert.assertThat;
import static org.hamcrest.Matchers.hasSize;
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertTrue;

Expand All @@ -26,8 +28,9 @@
import java.util.TreeMap;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hbase.Cell;
import org.apache.hadoop.hbase.CellBuilderFactory;
import org.apache.hadoop.hbase.CellBuilderType;
import org.apache.hadoop.hbase.HBaseTestingUtil;
import org.apache.hadoop.hbase.KeyValue;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.client.RegionInfo;
import org.apache.hadoop.hbase.client.RegionInfoBuilder;
Expand Down Expand Up @@ -82,33 +85,42 @@ public void doTest(TableName tableName) throws Exception {

for (int i = 0; i < total; i++) {
WALEdit kvs = new WALEdit();
kvs.add(new KeyValue(row, family, Bytes.toBytes(i), value));
kvs.add(CellBuilderFactory.create(CellBuilderType.SHALLOW_COPY).setType(Cell.Type.Put)
.setRow(row).setFamily(family).setQualifier(Bytes.toBytes(i)).setValue(value).build());
kvs.add(CellBuilderFactory.create(CellBuilderType.SHALLOW_COPY)
.setType(Cell.Type.DeleteFamily).setRow(row).setFamily(family).build());
wal.appendData(regionInfo, new WALKeyImpl(regionInfo.getEncodedNameAsBytes(), tableName,
System.currentTimeMillis(), mvcc, scopes), kvs);
wal.sync();
}
wal.sync();
final Path walPath = AbstractFSWALProvider.getCurrentFileName(wal);
wals.shutdown();

// Confirm the WAL can be read back
WAL.Reader reader = wals.createReader(TEST_UTIL.getTestFileSystem(), walPath);
int count = 0;
WAL.Entry entry = new WAL.Entry();
while (reader.next(entry) != null) {
count++;
List<Cell> cells = entry.getEdit().getCells();
assertTrue("Should be one KV per WALEdit", cells.size() == 1);
for (Cell cell : cells) {
assertTrue("Incorrect row", Bytes.equals(cell.getRowArray(), cell.getRowOffset(),
cell.getRowLength(), row, 0, row.length));
assertTrue("Incorrect family", Bytes.equals(cell.getFamilyArray(), cell.getFamilyOffset(),
cell.getFamilyLength(), family, 0, family.length));
assertTrue("Incorrect value", Bytes.equals(cell.getValueArray(), cell.getValueOffset(),
cell.getValueLength(), value, 0, value.length));
try (WAL.Reader reader = wals.createReader(TEST_UTIL.getTestFileSystem(), walPath)) {
int count = 0;
WAL.Entry entry = new WAL.Entry();
while (reader.next(entry) != null) {
count++;
List<Cell> cells = entry.getEdit().getCells();
assertThat("Should be two KVs per WALEdit", cells, hasSize(2));
Cell putCell = cells.get(0);
assertEquals(Cell.Type.Put, putCell.getType());
assertTrue("Incorrect row", Bytes.equals(putCell.getRowArray(), putCell.getRowOffset(),
putCell.getRowLength(), row, 0, row.length));
assertTrue("Incorrect family", Bytes.equals(putCell.getFamilyArray(),
putCell.getFamilyOffset(), putCell.getFamilyLength(), family, 0, family.length));
assertTrue("Incorrect value", Bytes.equals(putCell.getValueArray(),
putCell.getValueOffset(), putCell.getValueLength(), value, 0, value.length));

Cell deleteCell = cells.get(1);
assertEquals(Cell.Type.DeleteFamily, deleteCell.getType());
assertTrue("Incorrect row", Bytes.equals(deleteCell.getRowArray(),
deleteCell.getRowOffset(), deleteCell.getRowLength(), row, 0, row.length));
assertTrue("Incorrect family", Bytes.equals(deleteCell.getFamilyArray(),
deleteCell.getFamilyOffset(), deleteCell.getFamilyLength(), family, 0, family.length));
}
assertEquals("Should have read back as many KVs as written", total, count);
}
assertEquals("Should have read back as many KVs as written", total, count);
reader.close();
}

}
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@
import java.util.List;
import org.apache.hadoop.hbase.HBaseClassTestRule;
import org.apache.hadoop.hbase.HBaseCommonTestingUtil;
import org.apache.hadoop.hbase.HBaseTestingUtil;
import org.apache.hadoop.hbase.HConstants;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.io.compress.Compression;
Expand All @@ -46,9 +45,7 @@ public class TestCompressedWALValueCompression extends CompressedWALTestBase {
public static final HBaseClassTestRule CLASS_RULE =
HBaseClassTestRule.forClass(TestCompressedWALValueCompression.class);

private static final HBaseTestingUtil TEST_UTIL = new HBaseTestingUtil();

@Parameters
@Parameters(name = "{index}: compression={0}")
public static List<Object[]> params() {
return HBaseCommonTestingUtil.COMPRESSION_ALGORITHMS_PARAMETERIZED;
}
Expand Down Expand Up @@ -81,5 +78,4 @@ public void test() throws Exception {
TableName tableName = TableName.valueOf(name.getMethodName().replaceAll("[^a-zA-Z0-9]", "_"));
doTest(tableName);
}

}

0 comments on commit 2bbe036

Please sign in to comment.