|
|
@@ -19,30 +19,16 @@ package org.apache.hadoop.hdfs.server.datanode;
|
|
|
|
|
|
import com.google.common.base.Preconditions;
|
|
|
import org.apache.hadoop.hdfs.DFSUtilClient;
|
|
|
-import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
|
|
|
-import org.apache.hadoop.hdfs.protocol.ErasureCodingPolicy;
|
|
|
import org.apache.hadoop.hdfs.protocol.ExtendedBlock;
|
|
|
-import org.apache.hadoop.hdfs.protocol.StripedBlockInfo;
|
|
|
-import org.apache.hadoop.hdfs.protocol.datatransfer.DataTransferProtoUtil;
|
|
|
-import org.apache.hadoop.hdfs.protocol.datatransfer.IOStreamPair;
|
|
|
-import org.apache.hadoop.hdfs.protocol.datatransfer.Op;
|
|
|
-import org.apache.hadoop.hdfs.protocol.datatransfer.Sender;
|
|
|
-import org.apache.hadoop.hdfs.protocol.proto.DataTransferProtos;
|
|
|
-import org.apache.hadoop.hdfs.protocolPB.PBHelperClient;
|
|
|
-import org.apache.hadoop.hdfs.security.token.block.BlockTokenIdentifier;
|
|
|
import org.apache.hadoop.hdfs.server.datanode.fsdataset.LengthInputStream;
|
|
|
-import org.apache.hadoop.hdfs.util.StripedBlockUtil;
|
|
|
-import org.apache.hadoop.io.DataOutputBuffer;
|
|
|
import org.apache.hadoop.io.IOUtils;
|
|
|
import org.apache.hadoop.io.MD5Hash;
|
|
|
-import org.apache.hadoop.security.token.Token;
|
|
|
import org.apache.hadoop.util.DataChecksum;
|
|
|
import org.slf4j.Logger;
|
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
|
|
import java.io.BufferedInputStream;
|
|
|
import java.io.DataInputStream;
|
|
|
-import java.io.DataOutputStream;
|
|
|
import java.io.IOException;
|
|
|
import java.io.InputStream;
|
|
|
import java.security.MessageDigest;
|
|
|
@@ -55,87 +41,13 @@ final class BlockChecksumHelper {
|
|
|
|
|
|
static final Logger LOG = LoggerFactory.getLogger(BlockChecksumHelper.class);
|
|
|
|
|
|
- private BlockChecksumHelper() {
|
|
|
- }
|
|
|
+ private BlockChecksumHelper() {}
|
|
|
|
|
|
/**
|
|
|
* The abstract base block checksum computer.
|
|
|
*/
|
|
|
- static abstract class AbstractBlockChecksumComputer {
|
|
|
+ static abstract class BlockChecksumComputer {
|
|
|
private final DataNode datanode;
|
|
|
-
|
|
|
- private byte[] outBytes;
|
|
|
- private int bytesPerCRC = -1;
|
|
|
- private DataChecksum.Type crcType = null;
|
|
|
- private long crcPerBlock = -1;
|
|
|
- private int checksumSize = -1;
|
|
|
-
|
|
|
- AbstractBlockChecksumComputer(DataNode datanode) throws IOException {
|
|
|
- this.datanode = datanode;
|
|
|
- }
|
|
|
-
|
|
|
- abstract void compute() throws IOException;
|
|
|
-
|
|
|
- Sender createSender(IOStreamPair pair) {
|
|
|
- DataOutputStream out = (DataOutputStream) pair.out;
|
|
|
- return new Sender(out);
|
|
|
- }
|
|
|
-
|
|
|
- DataNode getDatanode() {
|
|
|
- return datanode;
|
|
|
- }
|
|
|
-
|
|
|
- InputStream getBlockInputStream(ExtendedBlock block, long seekOffset)
|
|
|
- throws IOException {
|
|
|
- return datanode.data.getBlockInputStream(block, seekOffset);
|
|
|
- }
|
|
|
-
|
|
|
- void setOutBytes(byte[] bytes) {
|
|
|
- this.outBytes = bytes;
|
|
|
- }
|
|
|
-
|
|
|
- byte[] getOutBytes() {
|
|
|
- return outBytes;
|
|
|
- }
|
|
|
-
|
|
|
- int getBytesPerCRC() {
|
|
|
- return bytesPerCRC;
|
|
|
- }
|
|
|
-
|
|
|
- public void setBytesPerCRC(int bytesPerCRC) {
|
|
|
- this.bytesPerCRC = bytesPerCRC;
|
|
|
- }
|
|
|
-
|
|
|
- public void setCrcType(DataChecksum.Type crcType) {
|
|
|
- this.crcType = crcType;
|
|
|
- }
|
|
|
-
|
|
|
- public void setCrcPerBlock(long crcPerBlock) {
|
|
|
- this.crcPerBlock = crcPerBlock;
|
|
|
- }
|
|
|
-
|
|
|
- public void setChecksumSize(int checksumSize) {
|
|
|
- this.checksumSize = checksumSize;
|
|
|
- }
|
|
|
-
|
|
|
- DataChecksum.Type getCrcType() {
|
|
|
- return crcType;
|
|
|
- }
|
|
|
-
|
|
|
- long getCrcPerBlock() {
|
|
|
- return crcPerBlock;
|
|
|
- }
|
|
|
-
|
|
|
- int getChecksumSize() {
|
|
|
- return checksumSize;
|
|
|
- }
|
|
|
- }
|
|
|
-
|
|
|
- /**
|
|
|
- * The abstract base block checksum computer.
|
|
|
- */
|
|
|
- static abstract class BlockChecksumComputer
|
|
|
- extends AbstractBlockChecksumComputer {
|
|
|
private final ExtendedBlock block;
|
|
|
// client side now can specify a range of the block for checksum
|
|
|
private final long requestLength;
|
|
|
@@ -144,12 +56,17 @@ final class BlockChecksumHelper {
|
|
|
private final long visibleLength;
|
|
|
private final boolean partialBlk;
|
|
|
|
|
|
+ private byte[] outBytes;
|
|
|
+ private int bytesPerCRC = -1;
|
|
|
+ private DataChecksum.Type crcType = null;
|
|
|
+ private long crcPerBlock = -1;
|
|
|
+ private int checksumSize = -1;
|
|
|
private BlockMetadataHeader header;
|
|
|
private DataChecksum checksum;
|
|
|
|
|
|
BlockChecksumComputer(DataNode datanode,
|
|
|
ExtendedBlock block) throws IOException {
|
|
|
- super(datanode);
|
|
|
+ this.datanode = datanode;
|
|
|
this.block = block;
|
|
|
this.requestLength = block.getNumBytes();
|
|
|
Preconditions.checkArgument(requestLength >= 0);
|
|
|
@@ -164,80 +81,98 @@ final class BlockChecksumHelper {
|
|
|
new BufferedInputStream(metadataIn, ioFileBufferSize));
|
|
|
}
|
|
|
|
|
|
- Sender createSender(IOStreamPair pair) {
|
|
|
- DataOutputStream out = (DataOutputStream) pair.out;
|
|
|
- return new Sender(out);
|
|
|
+ protected DataNode getDatanode() {
|
|
|
+ return datanode;
|
|
|
}
|
|
|
|
|
|
-
|
|
|
- ExtendedBlock getBlock() {
|
|
|
+ protected ExtendedBlock getBlock() {
|
|
|
return block;
|
|
|
}
|
|
|
|
|
|
- long getRequestLength() {
|
|
|
+ protected long getRequestLength() {
|
|
|
return requestLength;
|
|
|
}
|
|
|
|
|
|
- LengthInputStream getMetadataIn() {
|
|
|
+ protected LengthInputStream getMetadataIn() {
|
|
|
return metadataIn;
|
|
|
}
|
|
|
|
|
|
- DataInputStream getChecksumIn() {
|
|
|
+ protected DataInputStream getChecksumIn() {
|
|
|
return checksumIn;
|
|
|
}
|
|
|
|
|
|
- long getVisibleLength() {
|
|
|
+ protected long getVisibleLength() {
|
|
|
return visibleLength;
|
|
|
}
|
|
|
|
|
|
- boolean isPartialBlk() {
|
|
|
+ protected boolean isPartialBlk() {
|
|
|
return partialBlk;
|
|
|
}
|
|
|
|
|
|
- BlockMetadataHeader getHeader() {
|
|
|
+ protected void setOutBytes(byte[] bytes) {
|
|
|
+ this.outBytes = bytes;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected byte[] getOutBytes() {
|
|
|
+ return outBytes;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected int getBytesPerCRC() {
|
|
|
+ return bytesPerCRC;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected DataChecksum.Type getCrcType() {
|
|
|
+ return crcType;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected long getCrcPerBlock() {
|
|
|
+ return crcPerBlock;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected int getChecksumSize() {
|
|
|
+ return checksumSize;
|
|
|
+ }
|
|
|
+
|
|
|
+ protected BlockMetadataHeader getHeader() {
|
|
|
return header;
|
|
|
}
|
|
|
|
|
|
- DataChecksum getChecksum() {
|
|
|
+ protected DataChecksum getChecksum() {
|
|
|
return checksum;
|
|
|
}
|
|
|
|
|
|
/**
|
|
|
* Perform the block checksum computing.
|
|
|
- *
|
|
|
* @throws IOException
|
|
|
*/
|
|
|
abstract void compute() throws IOException;
|
|
|
|
|
|
/**
|
|
|
* Read block metadata header.
|
|
|
- *
|
|
|
* @throws IOException
|
|
|
*/
|
|
|
- void readHeader() throws IOException {
|
|
|
+ protected void readHeader() throws IOException {
|
|
|
//read metadata file
|
|
|
header = BlockMetadataHeader.readHeader(checksumIn);
|
|
|
checksum = header.getChecksum();
|
|
|
- setChecksumSize(checksum.getChecksumSize());
|
|
|
- setBytesPerCRC(checksum.getBytesPerChecksum());
|
|
|
- long crcPerBlock = checksum.getChecksumSize() <= 0 ? 0 :
|
|
|
+ checksumSize = checksum.getChecksumSize();
|
|
|
+ bytesPerCRC = checksum.getBytesPerChecksum();
|
|
|
+ crcPerBlock = checksumSize <= 0 ? 0 :
|
|
|
(metadataIn.getLength() -
|
|
|
- BlockMetadataHeader.getHeaderSize()) / checksum.getChecksumSize();
|
|
|
- setCrcPerBlock(crcPerBlock);
|
|
|
- setCrcType(checksum.getChecksumType());
|
|
|
+ BlockMetadataHeader.getHeaderSize()) / checksumSize;
|
|
|
+ crcType = checksum.getChecksumType();
|
|
|
}
|
|
|
|
|
|
/**
|
|
|
* Calculate partial block checksum.
|
|
|
- *
|
|
|
* @return
|
|
|
* @throws IOException
|
|
|
*/
|
|
|
- byte[] crcPartialBlock() throws IOException {
|
|
|
- int partialLength = (int) (requestLength % getBytesPerCRC());
|
|
|
+ protected byte[] crcPartialBlock() throws IOException {
|
|
|
+ int partialLength = (int) (requestLength % bytesPerCRC);
|
|
|
if (partialLength > 0) {
|
|
|
byte[] buf = new byte[partialLength];
|
|
|
- final InputStream blockIn = getBlockInputStream(block,
|
|
|
+ final InputStream blockIn = datanode.data.getBlockInputStream(block,
|
|
|
requestLength - partialLength);
|
|
|
try {
|
|
|
// Get the CRC of the partialLength.
|
|
|
@@ -246,7 +181,7 @@ final class BlockChecksumHelper {
|
|
|
IOUtils.closeStream(blockIn);
|
|
|
}
|
|
|
checksum.update(buf, 0, partialLength);
|
|
|
- byte[] partialCrc = new byte[getChecksumSize()];
|
|
|
+ byte[] partialCrc = new byte[checksumSize];
|
|
|
checksum.writeValue(partialCrc, 0, true);
|
|
|
return partialCrc;
|
|
|
}
|
|
|
@@ -294,7 +229,7 @@ final class BlockChecksumHelper {
|
|
|
}
|
|
|
|
|
|
private MD5Hash checksumPartialBlock() throws IOException {
|
|
|
- byte[] buffer = new byte[4 * 1024];
|
|
|
+ byte[] buffer = new byte[4*1024];
|
|
|
MessageDigest digester = MD5Hash.getDigester();
|
|
|
|
|
|
long remaining = (getRequestLength() / getBytesPerCRC())
|
|
|
@@ -316,115 +251,4 @@ final class BlockChecksumHelper {
|
|
|
return new MD5Hash(digester.digest());
|
|
|
}
|
|
|
}
|
|
|
-
|
|
|
- /**
|
|
|
- * Non-striped block group checksum computer for striped blocks.
|
|
|
- */
|
|
|
- static class BlockGroupNonStripedChecksumComputer
|
|
|
- extends AbstractBlockChecksumComputer {
|
|
|
-
|
|
|
- private final ExtendedBlock blockGroup;
|
|
|
- private final ErasureCodingPolicy ecPolicy;
|
|
|
- private final DatanodeInfo[] datanodes;
|
|
|
- private final Token<BlockTokenIdentifier>[] blockTokens;
|
|
|
-
|
|
|
- private final DataOutputBuffer md5writer = new DataOutputBuffer();
|
|
|
-
|
|
|
- BlockGroupNonStripedChecksumComputer(DataNode datanode,
|
|
|
- StripedBlockInfo stripedBlockInfo)
|
|
|
- throws IOException {
|
|
|
- super(datanode);
|
|
|
- this.blockGroup = stripedBlockInfo.getBlock();
|
|
|
- this.ecPolicy = stripedBlockInfo.getErasureCodingPolicy();
|
|
|
- this.datanodes = stripedBlockInfo.getDatanodes();
|
|
|
- this.blockTokens = stripedBlockInfo.getBlockTokens();
|
|
|
- }
|
|
|
-
|
|
|
- @Override
|
|
|
- void compute() throws IOException {
|
|
|
- for (int idx = 0; idx < ecPolicy.getNumDataUnits(); idx++) {
|
|
|
- ExtendedBlock block =
|
|
|
- StripedBlockUtil.constructInternalBlock(blockGroup,
|
|
|
- ecPolicy.getCellSize(), ecPolicy.getNumDataUnits(), idx);
|
|
|
- DatanodeInfo targetDatanode = datanodes[idx];
|
|
|
- Token<BlockTokenIdentifier> blockToken = blockTokens[idx];
|
|
|
- checksumBlock(block, idx, blockToken, targetDatanode);
|
|
|
- }
|
|
|
-
|
|
|
- MD5Hash md5out = MD5Hash.digest(md5writer.getData());
|
|
|
- setOutBytes(md5out.getDigest());
|
|
|
- }
|
|
|
-
|
|
|
- private void checksumBlock(ExtendedBlock block, int blockIdx,
|
|
|
- Token<BlockTokenIdentifier> blockToken,
|
|
|
- DatanodeInfo targetDatanode) throws IOException {
|
|
|
- int timeout = 3000;
|
|
|
- try (IOStreamPair pair = getDatanode().connectToDN(targetDatanode,
|
|
|
- timeout, block, blockToken)) {
|
|
|
-
|
|
|
- LOG.debug("write to {}: {}, block={}",
|
|
|
- getDatanode(), Op.BLOCK_CHECKSUM, block);
|
|
|
-
|
|
|
- // get block MD5
|
|
|
- createSender(pair).blockChecksum(block, blockToken);
|
|
|
-
|
|
|
- final DataTransferProtos.BlockOpResponseProto reply =
|
|
|
- DataTransferProtos.BlockOpResponseProto.parseFrom(
|
|
|
- PBHelperClient.vintPrefixed(pair.in));
|
|
|
-
|
|
|
- String logInfo = "for block " + block
|
|
|
- + " from datanode " + targetDatanode;
|
|
|
- DataTransferProtoUtil.checkBlockOpStatus(reply, logInfo);
|
|
|
-
|
|
|
- DataTransferProtos.OpBlockChecksumResponseProto checksumData =
|
|
|
- reply.getChecksumResponse();
|
|
|
-
|
|
|
- //read byte-per-checksum
|
|
|
- final int bpc = checksumData.getBytesPerCrc();
|
|
|
- if (blockIdx == 0) { //first block
|
|
|
- setBytesPerCRC(bpc);
|
|
|
- } else if (bpc != getBytesPerCRC()) {
|
|
|
- throw new IOException("Byte-per-checksum not matched: bpc=" + bpc
|
|
|
- + " but bytesPerCRC=" + getBytesPerCRC());
|
|
|
- }
|
|
|
-
|
|
|
- //read crc-per-block
|
|
|
- final long cpb = checksumData.getCrcPerBlock();
|
|
|
- if (blockIdx == 0) {
|
|
|
- setCrcPerBlock(cpb);
|
|
|
- }
|
|
|
-
|
|
|
- //read md5
|
|
|
- final MD5Hash md5 = new MD5Hash(
|
|
|
- checksumData.getMd5().toByteArray());
|
|
|
- md5.write(md5writer);
|
|
|
-
|
|
|
- // read crc-type
|
|
|
- final DataChecksum.Type ct;
|
|
|
- if (checksumData.hasCrcType()) {
|
|
|
- ct = PBHelperClient.convert(checksumData.getCrcType());
|
|
|
- } else {
|
|
|
- LOG.debug("Retrieving checksum from an earlier-version DataNode: " +
|
|
|
- "inferring checksum by reading first byte");
|
|
|
- ct = DataChecksum.Type.DEFAULT;
|
|
|
- }
|
|
|
-
|
|
|
- if (blockIdx == 0) { // first block
|
|
|
- setCrcType(ct);
|
|
|
- } else if (getCrcType() != DataChecksum.Type.MIXED &&
|
|
|
- getCrcType() != ct) {
|
|
|
- // if crc types are mixed in a file
|
|
|
- setCrcType(DataChecksum.Type.MIXED);
|
|
|
- }
|
|
|
-
|
|
|
- if (LOG.isDebugEnabled()) {
|
|
|
- if (blockIdx == 0) {
|
|
|
- LOG.debug("set bytesPerCRC=" + getBytesPerCRC()
|
|
|
- + ", crcPerBlock=" + getCrcPerBlock());
|
|
|
- }
|
|
|
- LOG.debug("got reply from " + targetDatanode + ": md5=" + md5);
|
|
|
- }
|
|
|
- }
|
|
|
- }
|
|
|
- }
|
|
|
-}
|
|
|
+}
|
Arpit Agarwal