Browse Source

Merge pull request #3 from chrislusf/master

sync
pull/1404/head
hilimd 4 years ago
committed by GitHub
parent
commit
6ea4ce7227
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
  1. 23
      other/java/client/src/main/java/seaweedfs/client/ByteBufferPool.java
  2. 135
      other/java/client/src/main/java/seaweedfs/client/FileChunkManifest.java
  3. 86
      other/java/client/src/main/java/seaweedfs/client/FilerClient.java
  4. 17
      other/java/client/src/main/java/seaweedfs/client/SeaweedRead.java
  5. 9
      other/java/client/src/main/java/seaweedfs/client/SeaweedUtil.java
  6. 36
      other/java/client/src/main/java/seaweedfs/client/SeaweedWrite.java
  7. 5
      other/java/client/src/test/java/seaweedfs/client/SeaweedReadTest.java
  8. 3
      other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedFileSystem.java
  9. 7
      other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedInputStream.java
  10. 8
      other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedOutputStream.java
  11. 3
      other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedFileSystem.java
  12. 7
      other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedInputStream.java
  13. 8
      other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedOutputStream.java
  14. 142
      weed/filer2/filechunk_manifest.go
  15. 113
      weed/filer2/filechunk_manifest_test.go
  16. 32
      weed/filer2/filechunks.go
  17. 10
      weed/filer2/filechunks_test.go
  18. 17
      weed/filer2/reader_at.go
  19. 28
      weed/filer2/stream.go
  20. 50
      weed/filesys/dirty_page.go
  21. 2
      weed/filesys/file.go
  22. 13
      weed/filesys/filehandle.go
  23. 23
      weed/filesys/meta_cache/meta_cache.go
  24. 66
      weed/filesys/wfs_write.go
  25. 2
      weed/replication/sink/azuresink/azure_sink.go
  26. 2
      weed/replication/sink/b2sink/b2_sink.go
  27. 26
      weed/replication/sink/filersink/filer_sink.go
  28. 2
      weed/replication/sink/gcssink/gcs_sink.go
  29. 2
      weed/replication/sink/s3sink/s3_sink.go
  30. 73
      weed/server/filer_grpc_server.go
  31. 5
      weed/server/filer_server_handlers_write.go
  32. 28
      weed/server/filer_server_handlers_write_autochunk.go
  33. 2
      weed/server/filer_server_handlers_write_cipher.go
  34. 2
      weed/server/webdav_server.go
  35. 8
      weed/shell/command_volume_fsck.go
  36. 10
      weed/storage/needle/volume_ttl.go

23
other/java/client/src/main/java/seaweedfs/client/ByteBufferPool.java

@ -1,22 +1,39 @@
package seaweedfs.client;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.nio.ByteBuffer;
import java.util.ArrayList;
import java.util.List;
public class ByteBufferPool {
static List<ByteBuffer> bufferList = new ArrayList<>();
private static final int MIN_BUFFER_SIZE = 8 * 1024 * 1024;
private static final Logger LOG = LoggerFactory.getLogger(ByteBufferPool.class);
private static final List<ByteBuffer> bufferList = new ArrayList<>();
public static synchronized ByteBuffer request(int bufferSize) {
if (bufferSize < MIN_BUFFER_SIZE) {
bufferSize = MIN_BUFFER_SIZE;
}
if (bufferList.isEmpty()) {
return ByteBuffer.allocate(bufferSize);
}
return bufferList.remove(bufferList.size()-1);
ByteBuffer buffer = bufferList.remove(bufferList.size() - 1);
if (buffer.capacity() >= bufferSize) {
return buffer;
}
LOG.info("add new buffer from {} to {}", buffer.capacity(), bufferSize);
bufferList.add(0, buffer);
return ByteBuffer.allocate(bufferSize);
}
public static synchronized void release(ByteBuffer obj) {
bufferList.add(obj);
bufferList.add(0, obj);
}
}

135
other/java/client/src/main/java/seaweedfs/client/FileChunkManifest.java

@ -0,0 +1,135 @@
package seaweedfs.client;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
public class FileChunkManifest {
private static final Logger LOG = LoggerFactory.getLogger(FileChunkManifest.class);
private static final int mergeFactor = 1000;
public static boolean hasChunkManifest(List<FilerProto.FileChunk> chunks) {
for (FilerProto.FileChunk chunk : chunks) {
if (chunk.getIsChunkManifest()) {
return true;
}
}
return false;
}
public static List<FilerProto.FileChunk> resolveChunkManifest(
final FilerGrpcClient filerGrpcClient, List<FilerProto.FileChunk> chunks) throws IOException {
List<FilerProto.FileChunk> dataChunks = new ArrayList<>();
for (FilerProto.FileChunk chunk : chunks) {
if (!chunk.getIsChunkManifest()) {
dataChunks.add(chunk);
continue;
}
// IsChunkManifest
LOG.debug("fetching chunk manifest:{}", chunk);
byte[] data = fetchChunk(filerGrpcClient, chunk);
FilerProto.FileChunkManifest m = FilerProto.FileChunkManifest.newBuilder().mergeFrom(data).build();
List<FilerProto.FileChunk> resolvedChunks = new ArrayList<>();
for (FilerProto.FileChunk t : m.getChunksList()) {
// avoid deprecated chunk.getFileId()
resolvedChunks.add(t.toBuilder().setFileId(FilerClient.toFileId(t.getFid())).build());
}
dataChunks.addAll(resolveChunkManifest(filerGrpcClient, resolvedChunks));
}
return dataChunks;
}
private static byte[] fetchChunk(final FilerGrpcClient filerGrpcClient, FilerProto.FileChunk chunk) throws IOException {
FilerProto.LookupVolumeRequest.Builder lookupRequest = FilerProto.LookupVolumeRequest.newBuilder();
String vid = "" + chunk.getFid().getVolumeId();
lookupRequest.addVolumeIds(vid);
FilerProto.LookupVolumeResponse lookupResponse = filerGrpcClient
.getBlockingStub().lookupVolume(lookupRequest.build());
Map<String, FilerProto.Locations> vid2Locations = lookupResponse.getLocationsMapMap();
FilerProto.Locations locations = vid2Locations.get(vid);
SeaweedRead.ChunkView chunkView = new SeaweedRead.ChunkView(
FilerClient.toFileId(chunk.getFid()), // avoid deprecated chunk.getFileId()
0,
-1,
0,
true,
chunk.getCipherKey().toByteArray(),
chunk.getIsCompressed());
byte[] chunkData = SeaweedRead.chunkCache.getChunk(chunkView.fileId);
if (chunkData == null) {
LOG.debug("doFetchFullChunkData:{}", chunkView);
chunkData = SeaweedRead.doFetchFullChunkData(chunkView, locations);
}
LOG.debug("chunk {} size {}", chunkView.fileId, chunkData.length);
SeaweedRead.chunkCache.setChunk(chunkView.fileId, chunkData);
return chunkData;
}
public static List<FilerProto.FileChunk> maybeManifestize(
final FilerGrpcClient filerGrpcClient, List<FilerProto.FileChunk> inputChunks) throws IOException {
// the return variable
List<FilerProto.FileChunk> chunks = new ArrayList<>();
List<FilerProto.FileChunk> dataChunks = new ArrayList<>();
for (FilerProto.FileChunk chunk : inputChunks) {
if (!chunk.getIsChunkManifest()) {
dataChunks.add(chunk);
} else {
chunks.add(chunk);
}
}
int remaining = dataChunks.size();
for (int i = 0; i + mergeFactor < dataChunks.size(); i += mergeFactor) {
FilerProto.FileChunk chunk = mergeIntoManifest(filerGrpcClient, dataChunks.subList(i, i + mergeFactor));
chunks.add(chunk);
remaining -= mergeFactor;
}
// remaining
for (int i = dataChunks.size() - remaining; i < dataChunks.size(); i++) {
chunks.add(dataChunks.get(i));
}
return chunks;
}
private static FilerProto.FileChunk mergeIntoManifest(final FilerGrpcClient filerGrpcClient, List<FilerProto.FileChunk> dataChunks) throws IOException {
// create and serialize the manifest
dataChunks = FilerClient.beforeEntrySerialization(dataChunks);
FilerProto.FileChunkManifest.Builder m = FilerProto.FileChunkManifest.newBuilder().addAllChunks(dataChunks);
byte[] data = m.build().toByteArray();
long minOffset = Long.MAX_VALUE;
long maxOffset = -1;
for (FilerProto.FileChunk chunk : dataChunks) {
minOffset = Math.min(minOffset, chunk.getOffset());
maxOffset = Math.max(maxOffset, chunk.getSize() + chunk.getOffset());
}
FilerProto.FileChunk.Builder manifestChunk = SeaweedWrite.writeChunk(
filerGrpcClient.getReplication(),
filerGrpcClient,
minOffset,
data, 0, data.length);
manifestChunk.setIsChunkManifest(true);
manifestChunk.setSize(maxOffset - minOffset);
return manifestChunk.build();
}
}

86
other/java/client/src/main/java/seaweedfs/client/FilerClient.java

@ -24,6 +24,67 @@ public class FilerClient {
this.filerGrpcClient = filerGrpcClient;
}
public static String toFileId(FilerProto.FileId fid) {
if (fid == null) {
return null;
}
return String.format("%d,%x%08x", fid.getVolumeId(), fid.getFileKey(), fid.getCookie());
}
public static FilerProto.FileId toFileIdObject(String fileIdStr) {
if (fileIdStr == null || fileIdStr.length() == 0) {
return null;
}
int commaIndex = fileIdStr.lastIndexOf(',');
String volumeIdStr = fileIdStr.substring(0, commaIndex);
String fileKeyStr = fileIdStr.substring(commaIndex + 1, fileIdStr.length() - 8);
String cookieStr = fileIdStr.substring(fileIdStr.length() - 8);
return FilerProto.FileId.newBuilder()
.setVolumeId(Integer.parseInt(volumeIdStr))
.setFileKey(Long.parseLong(fileKeyStr, 16))
.setCookie((int) Long.parseLong(cookieStr, 16))
.build();
}
public static List<FilerProto.FileChunk> beforeEntrySerialization(List<FilerProto.FileChunk> chunks) {
List<FilerProto.FileChunk> cleanedChunks = new ArrayList<>();
for (FilerProto.FileChunk chunk : chunks) {
FilerProto.FileChunk.Builder chunkBuilder = chunk.toBuilder();
chunkBuilder.clearFileId();
chunkBuilder.clearSourceFileId();
chunkBuilder.setFid(toFileIdObject(chunk.getFileId()));
FilerProto.FileId sourceFid = toFileIdObject(chunk.getSourceFileId());
if (sourceFid != null) {
chunkBuilder.setSourceFid(sourceFid);
}
cleanedChunks.add(chunkBuilder.build());
}
return cleanedChunks;
}
public static FilerProto.Entry afterEntryDeserialization(FilerProto.Entry entry) {
if (entry.getChunksList().size() <= 0) {
return entry;
}
String fileId = entry.getChunks(0).getFileId();
if (fileId != null && fileId.length() != 0) {
return entry;
}
FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
entryBuilder.clearChunks();
for (FilerProto.FileChunk chunk : entry.getChunksList()) {
FilerProto.FileChunk.Builder chunkBuilder = chunk.toBuilder();
chunkBuilder.setFileId(toFileId(chunk.getFid()));
String sourceFileId = toFileId(chunk.getSourceFid());
if (sourceFileId != null) {
chunkBuilder.setSourceFileId(sourceFileId);
}
entryBuilder.addChunks(chunkBuilder);
}
return entryBuilder.build();
}
public boolean mkdirs(String path, int mode) {
String currentUser = System.getProperty("user.name");
return mkdirs(path, mode, 0, 0, currentUser, new String[]{});
@ -184,7 +245,7 @@ public class FilerClient {
List<FilerProto.Entry> entries = new ArrayList<>();
while (iter.hasNext()) {
FilerProto.ListEntriesResponse resp = iter.next();
entries.add(fixEntryAfterReading(resp.getEntry()));
entries.add(afterEntryDeserialization(resp.getEntry()));
}
return entries;
}
@ -199,7 +260,7 @@ public class FilerClient {
if (entry == null) {
return null;
}
return fixEntryAfterReading(entry);
return afterEntryDeserialization(entry);
} catch (Exception e) {
if (e.getMessage().indexOf("filer: no entry is found in filer store") > 0) {
return null;
@ -209,7 +270,6 @@ public class FilerClient {
}
}
public boolean createEntry(String parent, FilerProto.Entry entry) {
try {
filerGrpcClient.getBlockingStub().createEntry(FilerProto.CreateEntryRequest.newBuilder()
@ -267,24 +327,4 @@ public class FilerClient {
return true;
}
private FilerProto.Entry fixEntryAfterReading(FilerProto.Entry entry) {
if (entry.getChunksList().size() <= 0) {
return entry;
}
String fileId = entry.getChunks(0).getFileId();
if (fileId != null && fileId.length() != 0) {
return entry;
}
FilerProto.Entry.Builder entryBuilder = entry.toBuilder();
entryBuilder.clearChunks();
for (FilerProto.FileChunk chunk : entry.getChunksList()) {
FilerProto.FileChunk.Builder chunkBuilder = chunk.toBuilder();
FilerProto.FileId fid = chunk.getFid();
fileId = String.format("%d,%d%x", fid.getVolumeId(), fid.getFileKey(), fid.getCookie());
chunkBuilder.setFileId(fileId);
entryBuilder.addChunks(chunkBuilder);
}
return entryBuilder.build();
}
}

17
other/java/client/src/main/java/seaweedfs/client/SeaweedRead.java

@ -2,16 +2,12 @@ package seaweedfs.client;
import org.apache.http.HttpEntity;
import org.apache.http.HttpHeaders;
import org.apache.http.HttpResponse;
import org.apache.http.client.HttpClient;
import org.apache.http.client.methods.CloseableHttpResponse;
import org.apache.http.client.methods.HttpGet;
import org.apache.http.impl.client.DefaultHttpClient;
import org.apache.http.util.EntityUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.Closeable;
import java.io.IOException;
import java.util.*;
@ -19,7 +15,7 @@ public class SeaweedRead {
private static final Logger LOG = LoggerFactory.getLogger(SeaweedRead.class);
static ChunkCache chunkCache = new ChunkCache(16);
static ChunkCache chunkCache = new ChunkCache(0);
// returns bytesRead
public static long read(FilerGrpcClient filerGrpcClient, List<VisibleInterval> visibleIntervals,
@ -44,7 +40,8 @@ public class SeaweedRead {
int startOffset = bufferOffset;
for (ChunkView chunkView : chunkViews) {
FilerProto.Locations locations = vid2Locations.get(parseVolumeId(chunkView.fileId));
if (locations.getLocationsCount() == 0) {
if (locations == null || locations.getLocationsCount() == 0) {
LOG.error("failed to locate {}", chunkView.fileId);
// log here!
return 0;
}
@ -77,7 +74,7 @@ public class SeaweedRead {
return len;
}
private static byte[] doFetchFullChunkData(ChunkView chunkView, FilerProto.Locations locations) throws IOException {
public static byte[] doFetchFullChunkData(ChunkView chunkView, FilerProto.Locations locations) throws IOException {
HttpGet request = new HttpGet(
String.format("http://%s/%s", locations.getLocations(0).getUrl(), chunkView.fileId));
@ -138,7 +135,11 @@ public class SeaweedRead {
return views;
}
public static List<VisibleInterval> nonOverlappingVisibleIntervals(List<FilerProto.FileChunk> chunkList) {
public static List<VisibleInterval> nonOverlappingVisibleIntervals(
final FilerGrpcClient filerGrpcClient, List<FilerProto.FileChunk> chunkList) throws IOException {
chunkList = FileChunkManifest.resolveChunkManifest(filerGrpcClient, chunkList);
FilerProto.FileChunk[] chunks = chunkList.toArray(new FilerProto.FileChunk[0]);
Arrays.sort(chunks, new Comparator<FilerProto.FileChunk>() {
@Override

9
other/java/client/src/main/java/seaweedfs/client/SeaweedUtil.java

@ -9,19 +9,22 @@ import org.apache.http.impl.conn.PoolingHttpClientConnectionManager;
public class SeaweedUtil {
static PoolingHttpClientConnectionManager cm = new PoolingHttpClientConnectionManager();
static CloseableHttpClient httpClient;
static {
// Increase max total connection to 200
cm.setMaxTotal(200);
// Increase default max connection per route to 20
cm.setDefaultMaxPerRoute(20);
}
public static CloseableHttpClient getClosableHttpClient() {
return HttpClientBuilder.create()
httpClient = HttpClientBuilder.create()
.setConnectionManager(cm)
.setConnectionReuseStrategy(DefaultConnectionReuseStrategy.INSTANCE)
.setKeepAliveStrategy(DefaultConnectionKeepAliveStrategy.INSTANCE)
.build();
}
public static CloseableHttpClient getClosableHttpClient() {
return httpClient;
}
}

36
other/java/client/src/main/java/seaweedfs/client/SeaweedWrite.java

@ -1,8 +1,6 @@
package seaweedfs.client;
import com.google.protobuf.ByteString;
import org.apache.http.HttpResponse;
import org.apache.http.client.HttpClient;
import org.apache.http.client.methods.CloseableHttpResponse;
import org.apache.http.client.methods.HttpPost;
import org.apache.http.entity.mime.HttpMultipartMode;
@ -10,10 +8,10 @@ import org.apache.http.entity.mime.MultipartEntityBuilder;
import org.apache.http.util.EntityUtils;
import java.io.ByteArrayInputStream;
import java.io.Closeable;
import java.io.IOException;
import java.io.InputStream;
import java.security.SecureRandom;
import java.util.List;
public class SeaweedWrite {
@ -25,6 +23,17 @@ public class SeaweedWrite {
final long offset,
final byte[] bytes,
final long bytesOffset, final long bytesLength) throws IOException {
synchronized (entry) {
entry.addChunks(writeChunk(replication, filerGrpcClient, offset, bytes, bytesOffset, bytesLength));
}
}
public static FilerProto.FileChunk.Builder writeChunk(final String replication,
final FilerGrpcClient filerGrpcClient,
final long offset,
final byte[] bytes,
final long bytesOffset,
final long bytesLength) throws IOException {
FilerProto.AssignVolumeResponse response = filerGrpcClient.getBlockingStub().assignVolume(
FilerProto.AssignVolumeRequest.newBuilder()
.setCollection(filerGrpcClient.getCollection())
@ -46,25 +55,28 @@ public class SeaweedWrite {
String etag = multipartUpload(targetUrl, auth, bytes, bytesOffset, bytesLength, cipherKey);
synchronized (entry) {
entry.addChunks(FilerProto.FileChunk.newBuilder()
// cache fileId ~ bytes
SeaweedRead.chunkCache.setChunk(fileId, bytes);
return FilerProto.FileChunk.newBuilder()
.setFileId(fileId)
.setOffset(offset)
.setSize(bytesLength)
.setMtime(System.currentTimeMillis() / 10000L)
.setETag(etag)
.setCipherKey(cipherKeyString)
);
.setCipherKey(cipherKeyString);
}
// cache fileId ~ bytes
SeaweedRead.chunkCache.setChunk(fileId, bytes);
public static void writeMeta(final FilerGrpcClient filerGrpcClient,
final String parentDirectory,
final FilerProto.Entry.Builder entry) throws IOException {
}
int chunkSize = entry.getChunksCount();
List<FilerProto.FileChunk> chunks = FileChunkManifest.maybeManifestize(filerGrpcClient, entry.getChunksList());
public static void writeMeta(final FilerGrpcClient filerGrpcClient,
final String parentDirectory, final FilerProto.Entry.Builder entry) {
synchronized (entry) {
entry.clearChunks();
entry.addAllChunks(chunks);
filerGrpcClient.getBlockingStub().createEntry(
FilerProto.CreateEntryRequest.newBuilder()
.setDirectory(parentDirectory)

5
other/java/client/src/test/java/seaweedfs/client/SeaweedReadTest.java

@ -3,13 +3,14 @@ package seaweedfs.client;
import org.junit.Assert;
import org.junit.Test;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
public class SeaweedReadTest {
@Test
public void testNonOverlappingVisibleIntervals() {
public void testNonOverlappingVisibleIntervals() throws IOException {
List<FilerProto.FileChunk> chunks = new ArrayList<>();
chunks.add(FilerProto.FileChunk.newBuilder()
.setFileId("aaa")
@ -24,7 +25,7 @@ public class SeaweedReadTest {
.setMtime(2000)
.build());
List<SeaweedRead.VisibleInterval> visibleIntervals = SeaweedRead.nonOverlappingVisibleIntervals(chunks);
List<SeaweedRead.VisibleInterval> visibleIntervals = SeaweedRead.nonOverlappingVisibleIntervals(null, chunks);
for (SeaweedRead.VisibleInterval visibleInterval : visibleIntervals) {
System.out.println("visible:" + visibleInterval);
}

3
other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedFileSystem.java

@ -30,7 +30,6 @@ public class SeaweedFileSystem extends FileSystem {
public static final String FS_SEAWEED_FILER_PORT = "fs.seaweed.filer.port";
private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
private static int BUFFER_SIZE = 16 * 1024 * 1024;
private URI uri;
private Path workingDirectory = new Path("/");
@ -61,8 +60,6 @@ public class SeaweedFileSystem extends FileSystem {
port = (port == -1) ? FS_SEAWEED_DEFAULT_PORT : port;
conf.setInt(FS_SEAWEED_FILER_PORT, port);
conf.setInt(IO_FILE_BUFFER_SIZE_KEY, BUFFER_SIZE);
setConf(conf);
this.uri = uri;

7
other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedInputStream.java

@ -2,7 +2,6 @@ package seaweed.hdfs;
// based on org.apache.hadoop.fs.azurebfs.services.AbfsInputStream
import com.google.common.base.Preconditions;
import org.apache.hadoop.fs.FSExceptionMessages;
import org.apache.hadoop.fs.FSInputStream;
import org.apache.hadoop.fs.FileSystem.Statistics;
@ -37,7 +36,7 @@ public class SeaweedInputStream extends FSInputStream {
final Statistics statistics,
final String path,
final FilerProto.Entry entry,
final int bufferSize) {
final int bufferSize) throws IOException {
this.filerGrpcClient = filerGrpcClient;
this.statistics = statistics;
this.path = path;
@ -45,7 +44,7 @@ public class SeaweedInputStream extends FSInputStream {
this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
this.bufferSize = bufferSize;
this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(entry.getChunksList());
this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(filerGrpcClient, entry.getChunksList());
LOG.debug("new path:{} entry:{} visibleIntervalList:{}", path, entry, visibleIntervalList);
@ -100,7 +99,7 @@ public class SeaweedInputStream extends FSInputStream {
}
}
return (int)bytesRead;
return (int) bytesRead;
}

8
other/java/hdfs2/src/main/java/seaweed/hdfs/SeaweedOutputStream.java

@ -56,12 +56,12 @@ public class SeaweedOutputStream extends OutputStream {
this.outputIndex = 0;
this.writeOperations = new ConcurrentLinkedDeque<>();
this.maxConcurrentRequestCount = 4 * Runtime.getRuntime().availableProcessors();
this.maxConcurrentRequestCount = Runtime.getRuntime().availableProcessors();
this.threadExecutor
= new ThreadPoolExecutor(maxConcurrentRequestCount,
maxConcurrentRequestCount,
10L,
120L,
TimeUnit.SECONDS,
new LinkedBlockingQueue<Runnable>());
this.completionService = new ExecutorCompletionService<>(this.threadExecutor);
@ -109,7 +109,7 @@ public class SeaweedOutputStream extends OutputStream {
break;
}
// System.out.println(path + " [" + (outputIndex + currentOffset) + "," + ((outputIndex + currentOffset) + writableBytes) + ")");
// System.out.println(path + " [" + (outputIndex + currentOffset) + "," + ((outputIndex + currentOffset) + writableBytes) + ") " + buffer.capacity());
buffer.put(data, currentOffset, writableBytes);
outputIndex += writableBytes;
currentOffset += writableBytes;
@ -180,7 +180,7 @@ public class SeaweedOutputStream extends OutputStream {
bufferToWrite.flip();
int bytesLength = bufferToWrite.limit() - bufferToWrite.position();
if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount * 2) {
if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount) {
waitForTaskToComplete();
}
final Future<Void> job = completionService.submit(() -> {

3
other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedFileSystem.java

@ -30,7 +30,6 @@ public class SeaweedFileSystem extends FileSystem {
public static final String FS_SEAWEED_FILER_PORT = "fs.seaweed.filer.port";
private static final Logger LOG = LoggerFactory.getLogger(SeaweedFileSystem.class);
private static int BUFFER_SIZE = 16 * 1024 * 1024;
private URI uri;
private Path workingDirectory = new Path("/");
@ -61,8 +60,6 @@ public class SeaweedFileSystem extends FileSystem {
port = (port == -1) ? FS_SEAWEED_DEFAULT_PORT : port;
conf.setInt(FS_SEAWEED_FILER_PORT, port);
conf.setInt(IO_FILE_BUFFER_SIZE_KEY, BUFFER_SIZE);
setConf(conf);
this.uri = uri;

7
other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedInputStream.java

@ -2,7 +2,6 @@ package seaweed.hdfs;
// based on org.apache.hadoop.fs.azurebfs.services.AbfsInputStream
import com.google.common.base.Preconditions;
import org.apache.hadoop.fs.FSExceptionMessages;
import org.apache.hadoop.fs.FSInputStream;
import org.apache.hadoop.fs.FileSystem.Statistics;
@ -37,7 +36,7 @@ public class SeaweedInputStream extends FSInputStream {
final Statistics statistics,
final String path,
final FilerProto.Entry entry,
final int bufferSize) {
final int bufferSize) throws IOException {
this.filerGrpcClient = filerGrpcClient;
this.statistics = statistics;
this.path = path;
@ -45,7 +44,7 @@ public class SeaweedInputStream extends FSInputStream {
this.contentLength = SeaweedRead.totalSize(entry.getChunksList());
this.bufferSize = bufferSize;
this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(entry.getChunksList());
this.visibleIntervalList = SeaweedRead.nonOverlappingVisibleIntervals(filerGrpcClient, entry.getChunksList());
LOG.debug("new path:{} entry:{} visibleIntervalList:{}", path, entry, visibleIntervalList);
@ -100,7 +99,7 @@ public class SeaweedInputStream extends FSInputStream {
}
}
return (int)bytesRead;
return (int) bytesRead;
}

8
other/java/hdfs3/src/main/java/seaweed/hdfs/SeaweedOutputStream.java

@ -60,12 +60,12 @@ public class SeaweedOutputStream extends OutputStream implements Syncable, Strea
this.outputIndex = 0;
this.writeOperations = new ConcurrentLinkedDeque<>();
this.maxConcurrentRequestCount = 4 * Runtime.getRuntime().availableProcessors();
this.maxConcurrentRequestCount = Runtime.getRuntime().availableProcessors();
this.threadExecutor
= new ThreadPoolExecutor(maxConcurrentRequestCount,
maxConcurrentRequestCount,
10L,
120L,
TimeUnit.SECONDS,
new LinkedBlockingQueue<Runnable>());
this.completionService = new ExecutorCompletionService<>(this.threadExecutor);
@ -113,7 +113,7 @@ public class SeaweedOutputStream extends OutputStream implements Syncable, Strea
break;
}
// System.out.println(path + " [" + (outputIndex + currentOffset) + "," + ((outputIndex + currentOffset) + writableBytes) + ")");
// System.out.println(path + " [" + (outputIndex + currentOffset) + "," + ((outputIndex + currentOffset) + writableBytes) + ") " + buffer.capacity());
buffer.put(data, currentOffset, writableBytes);
outputIndex += writableBytes;
currentOffset += writableBytes;
@ -227,7 +227,7 @@ public class SeaweedOutputStream extends OutputStream implements Syncable, Strea
bufferToWrite.flip();
int bytesLength = bufferToWrite.limit() - bufferToWrite.position();
if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount * 2) {
if (threadExecutor.getQueue().size() >= maxConcurrentRequestCount) {
waitForTaskToComplete();
}
final Future<Void> job = completionService.submit(() -> {

142
weed/filer2/filechunk_manifest.go

@ -0,0 +1,142 @@
package filer2
import (
"bytes"
"fmt"
"io"
"math"
"github.com/golang/protobuf/proto"
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/util"
)
const (
ManifestBatch = 1000
)
func HasChunkManifest(chunks []*filer_pb.FileChunk) bool {
for _, chunk := range chunks {
if chunk.IsChunkManifest {
return true
}
}
return false
}
func ResolveChunkManifest(lookupFileIdFn LookupFileIdFunctionType, chunks []*filer_pb.FileChunk) (dataChunks, manifestChunks []*filer_pb.FileChunk, manefestResolveErr error) {
// TODO maybe parallel this
for _, chunk := range chunks {
if !chunk.IsChunkManifest {
dataChunks = append(dataChunks, chunk)
continue
}
// IsChunkManifest
data, err := fetchChunk(lookupFileIdFn, chunk.FileId, chunk.CipherKey, chunk.IsCompressed)
if err != nil {
return chunks, nil, fmt.Errorf("fail to read manifest %s: %v", chunk.FileId, err)
}
m := &filer_pb.FileChunkManifest{}
if err := proto.Unmarshal(data, m); err != nil {
return chunks, nil, fmt.Errorf("fail to unmarshal manifest %s: %v", chunk.FileId, err)
}
manifestChunks = append(manifestChunks, chunk)
// recursive
filer_pb.AfterEntryDeserialization(m.Chunks)
dchunks, mchunks, subErr := ResolveChunkManifest(lookupFileIdFn, m.Chunks)
if subErr != nil {
return chunks, nil, subErr
}
dataChunks = append(dataChunks, dchunks...)
manifestChunks = append(manifestChunks, mchunks...)
}
return
}
// TODO fetch from cache for weed mount?
func fetchChunk(lookupFileIdFn LookupFileIdFunctionType, fileId string, cipherKey []byte, isGzipped bool) ([]byte, error) {
urlString, err := lookupFileIdFn(fileId)
if err != nil {
glog.Errorf("operation LookupFileId %s failed, err: %v", fileId, err)
return nil, err
}
var buffer bytes.Buffer
err = util.ReadUrlAsStream(urlString, cipherKey, isGzipped, true, 0, 0, func(data []byte) {
buffer.Write(data)
})
if err != nil {
glog.V(0).Infof("read %s failed, err: %v", fileId, err)
return nil, err
}
return buffer.Bytes(), nil
}
func MaybeManifestize(saveFunc SaveDataAsChunkFunctionType, inputChunks []*filer_pb.FileChunk) (chunks []*filer_pb.FileChunk, err error) {
return doMaybeManifestize(saveFunc, inputChunks, ManifestBatch, mergeIntoManifest)
}
func doMaybeManifestize(saveFunc SaveDataAsChunkFunctionType, inputChunks []*filer_pb.FileChunk, mergeFactor int, mergefn func(saveFunc SaveDataAsChunkFunctionType, dataChunks []*filer_pb.FileChunk) (manifestChunk *filer_pb.FileChunk, err error)) (chunks []*filer_pb.FileChunk, err error) {
var dataChunks []*filer_pb.FileChunk
for _, chunk := range inputChunks {
if !chunk.IsChunkManifest {
dataChunks = append(dataChunks, chunk)
} else {
chunks = append(chunks, chunk)
}
}
remaining := len(dataChunks)
for i := 0; i+mergeFactor <= len(dataChunks); i += mergeFactor {
chunk, err := mergefn(saveFunc, dataChunks[i:i+mergeFactor])
if err != nil {
return dataChunks, err
}
chunks = append(chunks, chunk)
remaining -= mergeFactor
}
// remaining
for i := len(dataChunks) - remaining; i < len(dataChunks); i++ {
chunks = append(chunks, dataChunks[i])
}
return
}
func mergeIntoManifest(saveFunc SaveDataAsChunkFunctionType, dataChunks []*filer_pb.FileChunk) (manifestChunk *filer_pb.FileChunk, err error) {
filer_pb.BeforeEntrySerialization(dataChunks)
// create and serialize the manifest
data, serErr := proto.Marshal(&filer_pb.FileChunkManifest{
Chunks: dataChunks,
})
if serErr != nil {
return nil, fmt.Errorf("serializing manifest: %v", serErr)
}
minOffset, maxOffset := int64(math.MaxInt64), int64(math.MinInt64)
for _, chunk := range dataChunks {
if minOffset > int64(chunk.Offset) {
minOffset = chunk.Offset
}
if maxOffset < int64(chunk.Size)+chunk.Offset {
maxOffset = int64(chunk.Size) + chunk.Offset
}
}
manifestChunk, _, _, err = saveFunc(bytes.NewReader(data), "", 0)
if err != nil {
return nil, err
}
manifestChunk.IsChunkManifest = true
manifestChunk.Offset = minOffset
manifestChunk.Size = uint64(maxOffset - minOffset)
return
}
type SaveDataAsChunkFunctionType func(reader io.Reader, name string, offset int64) (chunk *filer_pb.FileChunk, collection, replication string, err error)

113
weed/filer2/filechunk_manifest_test.go

@ -0,0 +1,113 @@
package filer2
import (
"bytes"
"math"
"testing"
"github.com/stretchr/testify/assert"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
)
func TestDoMaybeManifestize(t *testing.T) {
var manifestTests = []struct {
inputs []*filer_pb.FileChunk
expected []*filer_pb.FileChunk
}{
{
inputs: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: false},
{FileId: "2", IsChunkManifest: false},
{FileId: "3", IsChunkManifest: false},
{FileId: "4", IsChunkManifest: false},
},
expected: []*filer_pb.FileChunk{
{FileId: "12", IsChunkManifest: true},
{FileId: "34", IsChunkManifest: true},
},
},
{
inputs: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: true},
{FileId: "2", IsChunkManifest: false},
{FileId: "3", IsChunkManifest: false},
{FileId: "4", IsChunkManifest: false},
},
expected: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: true},
{FileId: "23", IsChunkManifest: true},
{FileId: "4", IsChunkManifest: false},
},
},
{
inputs: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: false},
{FileId: "2", IsChunkManifest: true},
{FileId: "3", IsChunkManifest: false},
{FileId: "4", IsChunkManifest: false},
},
expected: []*filer_pb.FileChunk{
{FileId: "2", IsChunkManifest: true},
{FileId: "13", IsChunkManifest: true},
{FileId: "4", IsChunkManifest: false},
},
},
{
inputs: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: true},
{FileId: "2", IsChunkManifest: true},
{FileId: "3", IsChunkManifest: false},
{FileId: "4", IsChunkManifest: false},
},
expected: []*filer_pb.FileChunk{
{FileId: "1", IsChunkManifest: true},
{FileId: "2", IsChunkManifest: true},
{FileId: "34", IsChunkManifest: true},
},
},
}
for i, mtest := range manifestTests {
println("test", i)
actual, _ := doMaybeManifestize(nil, mtest.inputs, 2, mockMerge)
assertEqualChunks(t, mtest.expected, actual)
}
}
func assertEqualChunks(t *testing.T, expected, actual []*filer_pb.FileChunk) {
assert.Equal(t, len(expected), len(actual))
for i := 0; i < len(actual); i++ {
assertEqualChunk(t, actual[i], expected[i])
}
}
func assertEqualChunk(t *testing.T, expected, actual *filer_pb.FileChunk) {
assert.Equal(t, expected.FileId, actual.FileId)
assert.Equal(t, expected.IsChunkManifest, actual.IsChunkManifest)
}
func mockMerge(saveFunc SaveDataAsChunkFunctionType, dataChunks []*filer_pb.FileChunk) (manifestChunk *filer_pb.FileChunk, err error) {
var buf bytes.Buffer
minOffset, maxOffset := int64(math.MaxInt64), int64(math.MinInt64)
for k := 0; k < len(dataChunks); k++ {
chunk := dataChunks[k]
buf.WriteString(chunk.FileId)
if minOffset > int64(chunk.Offset) {
minOffset = chunk.Offset
}
if maxOffset < int64(chunk.Size)+chunk.Offset {
maxOffset = int64(chunk.Size) + chunk.Offset
}
}
manifestChunk = &filer_pb.FileChunk{
FileId: buf.String(),
}
manifestChunk.IsChunkManifest = true
manifestChunk.Offset = minOffset
manifestChunk.Size = uint64(maxOffset - minOffset)
return
}

32
weed/filer2/filechunks.go

@ -46,9 +46,9 @@ func ETagChunks(chunks []*filer_pb.FileChunk) (etag string) {
return fmt.Sprintf("%x", h.Sum32())
}
func CompactFileChunks(chunks []*filer_pb.FileChunk) (compacted, garbage []*filer_pb.FileChunk) {
func CompactFileChunks(lookupFileIdFn LookupFileIdFunctionType, chunks []*filer_pb.FileChunk) (compacted, garbage []*filer_pb.FileChunk) {
visibles := NonOverlappingVisibleIntervals(chunks)
visibles, _ := NonOverlappingVisibleIntervals(lookupFileIdFn, chunks)
fileIds := make(map[string]bool)
for _, interval := range visibles {
@ -65,7 +65,23 @@ func CompactFileChunks(chunks []*filer_pb.FileChunk) (compacted, garbage []*file
return
}
func MinusChunks(as, bs []*filer_pb.FileChunk) (delta []*filer_pb.FileChunk) {
func MinusChunks(lookupFileIdFn LookupFileIdFunctionType, as, bs []*filer_pb.FileChunk) (delta []*filer_pb.FileChunk, err error) {
aData, aMeta, aErr := ResolveChunkManifest(lookupFileIdFn, as)
if aErr != nil {
return nil, aErr
}
bData, bMeta, bErr := ResolveChunkManifest(lookupFileIdFn, bs)
if bErr != nil {
return nil, bErr
}
delta = append(delta, DoMinusChunks(aData, bData)...)
delta = append(delta, DoMinusChunks(aMeta, bMeta)...)
return
}
func DoMinusChunks(as, bs []*filer_pb.FileChunk) (delta []*filer_pb.FileChunk) {
fileIds := make(map[string]bool)
for _, interval := range bs {
@ -94,9 +110,9 @@ func (cv *ChunkView) IsFullChunk() bool {
return cv.Size == cv.ChunkSize
}
func ViewFromChunks(chunks []*filer_pb.FileChunk, offset int64, size int64) (views []*ChunkView) {
func ViewFromChunks(lookupFileIdFn LookupFileIdFunctionType, chunks []*filer_pb.FileChunk, offset int64, size int64) (views []*ChunkView) {
visibles := NonOverlappingVisibleIntervals(chunks)
visibles, _ := NonOverlappingVisibleIntervals(lookupFileIdFn, chunks)
return ViewFromVisibleIntervals(visibles, offset, size)
@ -190,7 +206,11 @@ func MergeIntoVisibles(visibles, newVisibles []VisibleInterval, chunk *filer_pb.
return newVisibles
}
func NonOverlappingVisibleIntervals(chunks []*filer_pb.FileChunk) (visibles []VisibleInterval) {
// NonOverlappingVisibleIntervals translates the file chunk into VisibleInterval in memory
// If the file chunk content is a chunk manifest
func NonOverlappingVisibleIntervals(lookupFileIdFn LookupFileIdFunctionType, chunks []*filer_pb.FileChunk) (visibles []VisibleInterval, err error) {
chunks, _, err = ResolveChunkManifest(lookupFileIdFn, chunks)
sort.Slice(chunks, func(i, j int) bool {
return chunks[i].Mtime < chunks[j].Mtime

10
weed/filer2/filechunks_test.go

@ -16,7 +16,7 @@ func TestCompactFileChunks(t *testing.T) {
{Offset: 110, Size: 200, FileId: "jkl", Mtime: 300},
}
compacted, garbage := CompactFileChunks(chunks)
compacted, garbage := CompactFileChunks(nil, chunks)
if len(compacted) != 3 {
t.Fatalf("unexpected compacted: %d", len(compacted))
@ -49,7 +49,7 @@ func TestCompactFileChunks2(t *testing.T) {
})
}
compacted, garbage := CompactFileChunks(chunks)
compacted, garbage := CompactFileChunks(nil, chunks)
if len(compacted) != 4 {
t.Fatalf("unexpected compacted: %d", len(compacted))
@ -186,7 +186,7 @@ func TestIntervalMerging(t *testing.T) {
for i, testcase := range testcases {
log.Printf("++++++++++ merged test case %d ++++++++++++++++++++", i)
intervals := NonOverlappingVisibleIntervals(testcase.Chunks)
intervals, _ := NonOverlappingVisibleIntervals(nil, testcase.Chunks)
for x, interval := range intervals {
log.Printf("test case %d, interval %d, start=%d, stop=%d, fileId=%s",
i, x, interval.start, interval.stop, interval.fileId)
@ -371,7 +371,7 @@ func TestChunksReading(t *testing.T) {
for i, testcase := range testcases {
log.Printf("++++++++++ read test case %d ++++++++++++++++++++", i)
chunks := ViewFromChunks(testcase.Chunks, testcase.Offset, testcase.Size)
chunks := ViewFromChunks(nil, testcase.Chunks, testcase.Offset, testcase.Size)
for x, chunk := range chunks {
log.Printf("read case %d, chunk %d, offset=%d, size=%d, fileId=%s",
i, x, chunk.Offset, chunk.Size, chunk.FileId)
@ -415,6 +415,6 @@ func BenchmarkCompactFileChunks(b *testing.B) {
}
for n := 0; n < b.N; n++ {
CompactFileChunks(chunks)
CompactFileChunks(nil, chunks)
}
}

17
weed/filer2/reader_at.go

@ -1,7 +1,6 @@
package filer2
import (
"bytes"
"context"
"fmt"
"io"
@ -9,7 +8,6 @@ import (
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/util"
"github.com/chrislusf/seaweedfs/weed/util/chunk_cache"
"github.com/chrislusf/seaweedfs/weed/wdclient"
)
@ -144,19 +142,6 @@ func (c *ChunkReadAt) fetchChunkData(chunkView *ChunkView) (data []byte, err err
func (c *ChunkReadAt) doFetchFullChunkData(fileId string, cipherKey []byte, isGzipped bool) ([]byte, error) {
urlString, err := c.lookupFileId(fileId)
if err != nil {
glog.V(1).Infof("operation LookupFileId %s failed, err: %v", fileId, err)
return nil, err
}
var buffer bytes.Buffer
err = util.ReadUrlAsStream(urlString, cipherKey, isGzipped, true, 0, 0, func(data []byte) {
buffer.Write(data)
})
if err != nil {
glog.V(0).Infof("read %s failed, err: %v", fileId, err)
return nil, err
}
return fetchChunk(c.lookupFileId, fileId, cipherKey, isGzipped)
return buffer.Bytes(), nil
}

28
weed/filer2/stream.go

@ -2,6 +2,7 @@ package filer2
import (
"bytes"
"fmt"
"io"
"math"
"strings"
@ -14,7 +15,8 @@ import (
func StreamContent(masterClient *wdclient.MasterClient, w io.Writer, chunks []*filer_pb.FileChunk, offset int64, size int64) error {
chunkViews := ViewFromChunks(chunks, offset, size)
fmt.Printf("start to stream content for chunks: %+v\n", chunks)
chunkViews := ViewFromChunks(masterClient.LookupFileId, chunks, offset, size)
fileId2Url := make(map[string]string)
@ -50,14 +52,14 @@ func ReadAll(masterClient *wdclient.MasterClient, chunks []*filer_pb.FileChunk)
buffer := bytes.Buffer{}
chunkViews := ViewFromChunks(chunks, 0, math.MaxInt32)
lookupFileId := func(fileId string) (targetUrl string, err error) {
lookupFileIdFn := func(fileId string) (targetUrl string, err error) {
return masterClient.LookupFileId(fileId)
}
chunkViews := ViewFromChunks(lookupFileIdFn, chunks, 0, math.MaxInt64)
for _, chunkView := range chunkViews {
urlString, err := lookupFileId(chunkView.FileId)
urlString, err := lookupFileIdFn(chunkView.FileId)
if err != nil {
glog.V(1).Infof("operation LookupFileId %s failed, err: %v", chunkView.FileId, err)
return nil, err
@ -88,23 +90,27 @@ var _ = io.ReadSeeker(&ChunkStreamReader{})
func NewChunkStreamReaderFromFiler(masterClient *wdclient.MasterClient, chunks []*filer_pb.FileChunk) *ChunkStreamReader {
chunkViews := ViewFromChunks(chunks, 0, math.MaxInt32)
lookupFileIdFn := func(fileId string) (targetUrl string, err error) {
return masterClient.LookupFileId(fileId)
}
chunkViews := ViewFromChunks(lookupFileIdFn, chunks, 0, math.MaxInt64)
return &ChunkStreamReader{
chunkViews: chunkViews,
lookupFileId: func(fileId string) (targetUrl string, err error) {
return masterClient.LookupFileId(fileId)
},
lookupFileId: lookupFileIdFn,
}
}
func NewChunkStreamReader(filerClient filer_pb.FilerClient, chunks []*filer_pb.FileChunk) *ChunkStreamReader {
chunkViews := ViewFromChunks(chunks, 0, math.MaxInt32)
lookupFileIdFn := LookupFn(filerClient)
chunkViews := ViewFromChunks(lookupFileIdFn, chunks, 0, math.MaxInt64)
return &ChunkStreamReader{
chunkViews: chunkViews,
lookupFileId: LookupFn(filerClient),
lookupFileId: lookupFileIdFn,
}
}

50
weed/filesys/dirty_page.go

@ -2,16 +2,12 @@ package filesys
import (
"bytes"
"context"
"fmt"
"io"
"sync"
"time"
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/operation"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/security"
)
type ContinuousDirtyPages struct {
@ -141,53 +137,15 @@ func (pages *ContinuousDirtyPages) saveExistingLargestPageToStorage() (chunk *fi
func (pages *ContinuousDirtyPages) saveToStorage(reader io.Reader, offset int64, size int64) (*filer_pb.FileChunk, error) {
var fileId, host string
var auth security.EncodedJwt
dir, _ := pages.f.fullpath().DirAndName()
if err := pages.f.wfs.WithFilerClient(func(client filer_pb.SeaweedFilerClient) error {
request := &filer_pb.AssignVolumeRequest{
Count: 1,
Replication: pages.f.wfs.option.Replication,
Collection: pages.f.wfs.option.Collection,
TtlSec: pages.f.wfs.option.TtlSec,
DataCenter: pages.f.wfs.option.DataCenter,
ParentPath: dir,
}
resp, err := client.AssignVolume(context.Background(), request)
if err != nil {
glog.V(0).Infof("assign volume failure %v: %v", request, err)
return err
}
if resp.Error != "" {
return fmt.Errorf("assign volume failure %v: %v", request, resp.Error)
}
fileId, host, auth = resp.FileId, resp.Url, security.EncodedJwt(resp.Auth)
host = pages.f.wfs.AdjustedUrl(host)
pages.collection, pages.replication = resp.Collection, resp.Replication
return nil
}); err != nil {
return nil, fmt.Errorf("filerGrpcAddress assign volume: %v", err)
}
fileUrl := fmt.Sprintf("http://%s/%s", host, fileId)
uploadResult, err, data := operation.Upload(fileUrl, pages.f.Name, pages.f.wfs.option.Cipher, reader, false, "", nil, auth)
chunk, collection, replication, err := pages.f.wfs.saveDataAsChunk(dir)(reader, pages.f.Name, offset)
if err != nil {
glog.V(0).Infof("upload data %v to %s: %v", pages.f.Name, fileUrl, err)
return nil, fmt.Errorf("upload data: %v", err)
}
if uploadResult.Error != "" {
glog.V(0).Infof("upload failure %v to %s: %v", pages.f.Name, fileUrl, err)
return nil, fmt.Errorf("upload result: %v", uploadResult.Error)
return nil, err
}
pages.f.wfs.chunkCache.SetChunk(fileId, data)
pages.collection, pages.replication = collection, replication
return uploadResult.ToPbFileChunk(fileId, offset), nil
return chunk, nil
}

2
weed/filesys/file.go

@ -253,7 +253,7 @@ func (file *File) addChunks(chunks []*filer_pb.FileChunk) {
func (file *File) setEntry(entry *filer_pb.Entry) {
file.entry = entry
file.entryViewCache = filer2.NonOverlappingVisibleIntervals(file.entry.Chunks)
file.entryViewCache, _ = filer2.NonOverlappingVisibleIntervals(filer2.LookupFn(file.wfs), file.entry.Chunks)
file.reader = nil
}

13
weed/filesys/filehandle.go

@ -88,8 +88,12 @@ func (fh *FileHandle) readFromChunks(buff []byte, offset int64) (int64, error) {
return 0, nil
}
var chunkResolveErr error
if fh.f.entryViewCache == nil {
fh.f.entryViewCache = filer2.NonOverlappingVisibleIntervals(fh.f.entry.Chunks)
fh.f.entryViewCache, chunkResolveErr = filer2.NonOverlappingVisibleIntervals(filer2.LookupFn(fh.f.wfs), fh.f.entry.Chunks)
if chunkResolveErr != nil {
return 0, fmt.Errorf("fail to resolve chunk manifest: %v", chunkResolveErr)
}
fh.f.reader = nil
}
@ -206,7 +210,12 @@ func (fh *FileHandle) Flush(ctx context.Context, req *fuse.FlushRequest) error {
glog.V(3).Infof("%s chunks %d: %v [%d,%d)", fh.f.fullpath(), i, chunk.FileId, chunk.Offset, chunk.Offset+int64(chunk.Size))
}
chunks, garbages := filer2.CompactFileChunks(fh.f.entry.Chunks)
chunks, garbages := filer2.CompactFileChunks(filer2.LookupFn(fh.f.wfs), fh.f.entry.Chunks)
chunks, manifestErr := filer2.MaybeManifestize(fh.f.wfs.saveDataAsChunk(fh.f.dir.FullPath()), chunks)
if manifestErr != nil {
// not good, but should be ok
glog.V(0).Infof("MaybeManifestize: %v", manifestErr)
}
fh.f.entry.Chunks = chunks
// fh.f.entryViewCache = nil

23
weed/filesys/meta_cache/meta_cache.go

@ -8,10 +8,14 @@ import (
"github.com/chrislusf/seaweedfs/weed/filer2"
"github.com/chrislusf/seaweedfs/weed/filer2/leveldb"
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/util"
"github.com/chrislusf/seaweedfs/weed/util/bounded_tree"
)
// need to have logic similar to FilerStoreWrapper
// e.g. fill fileId field for chunks
type MetaCache struct {
actualStore filer2.FilerStore
sync.RWMutex
@ -46,6 +50,7 @@ func openMetaStore(dbFolder string) filer2.FilerStore {
func (mc *MetaCache) InsertEntry(ctx context.Context, entry *filer2.Entry) error {
mc.Lock()
defer mc.Unlock()
filer_pb.BeforeEntrySerialization(entry.Chunks)
return mc.actualStore.InsertEntry(ctx, entry)
}
@ -78,13 +83,19 @@ func (mc *MetaCache) AtomicUpdateEntry(ctx context.Context, oldPath util.FullPat
func (mc *MetaCache) UpdateEntry(ctx context.Context, entry *filer2.Entry) error {
mc.Lock()
defer mc.Unlock()
filer_pb.BeforeEntrySerialization(entry.Chunks)
return mc.actualStore.UpdateEntry(ctx, entry)
}
func (mc *MetaCache) FindEntry(ctx context.Context, fp util.FullPath) (entry *filer2.Entry, err error) {
mc.RLock()
defer mc.RUnlock()
return mc.actualStore.FindEntry(ctx, fp)
entry, err = mc.actualStore.FindEntry(ctx, fp)
if err != nil {
return nil, err
}
filer_pb.AfterEntryDeserialization(entry.Chunks)
return
}
func (mc *MetaCache) DeleteEntry(ctx context.Context, fp util.FullPath) (err error) {
@ -96,7 +107,15 @@ func (mc *MetaCache) DeleteEntry(ctx context.Context, fp util.FullPath) (err err
func (mc *MetaCache) ListDirectoryEntries(ctx context.Context, dirPath util.FullPath, startFileName string, includeStartFile bool, limit int) ([]*filer2.Entry, error) {
mc.RLock()
defer mc.RUnlock()
return mc.actualStore.ListDirectoryEntries(ctx, dirPath, startFileName, includeStartFile, limit)
entries, err := mc.actualStore.ListDirectoryEntries(ctx, dirPath, startFileName, includeStartFile, limit)
if err != nil {
return nil, err
}
for _, entry := range entries {
filer_pb.AfterEntryDeserialization(entry.Chunks)
}
return entries, err
}
func (mc *MetaCache) Shutdown() {

66
weed/filesys/wfs_write.go

@ -0,0 +1,66 @@
package filesys
import (
"context"
"fmt"
"io"
"github.com/chrislusf/seaweedfs/weed/filer2"
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/operation"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/security"
)
func (wfs *WFS) saveDataAsChunk(dir string) filer2.SaveDataAsChunkFunctionType {
return func(reader io.Reader, filename string, offset int64) (chunk *filer_pb.FileChunk, collection, replication string, err error) {
var fileId, host string
var auth security.EncodedJwt
if err := wfs.WithFilerClient(func(client filer_pb.SeaweedFilerClient) error {
request := &filer_pb.AssignVolumeRequest{
Count: 1,
Replication: wfs.option.Replication,
Collection: wfs.option.Collection,
TtlSec: wfs.option.TtlSec,
DataCenter: wfs.option.DataCenter,
ParentPath: dir,
}
resp, err := client.AssignVolume(context.Background(), request)
if err != nil {
glog.V(0).Infof("assign volume failure %v: %v", request, err)
return err
}
if resp.Error != "" {
return fmt.Errorf("assign volume failure %v: %v", request, resp.Error)
}
fileId, host, auth = resp.FileId, resp.Url, security.EncodedJwt(resp.Auth)
host = wfs.AdjustedUrl(host)
collection, replication = resp.Collection, resp.Replication
return nil
}); err != nil {
return nil, "", "", fmt.Errorf("filerGrpcAddress assign volume: %v", err)
}
fileUrl := fmt.Sprintf("http://%s/%s", host, fileId)
uploadResult, err, data := operation.Upload(fileUrl, filename, wfs.option.Cipher, reader, false, "", nil, auth)
if err != nil {
glog.V(0).Infof("upload data %v to %s: %v", filename, fileUrl, err)
return nil, "", "", fmt.Errorf("upload data: %v", err)
}
if uploadResult.Error != "" {
glog.V(0).Infof("upload failure %v to %s: %v", filename, fileUrl, err)
return nil, "", "", fmt.Errorf("upload result: %v", uploadResult.Error)
}
wfs.chunkCache.SetChunk(fileId, data)
chunk = uploadResult.ToPbFileChunk(fileId, offset)
return chunk, "", "", nil
}
}

2
weed/replication/sink/azuresink/azure_sink.go

@ -96,7 +96,7 @@ func (g *AzureSink) CreateEntry(key string, entry *filer_pb.Entry) error {
}
totalSize := filer2.TotalSize(entry.Chunks)
chunkViews := filer2.ViewFromChunks(entry.Chunks, 0, int64(totalSize))
chunkViews := filer2.ViewFromChunks(g.filerSource.LookupFileId, entry.Chunks, 0, int64(totalSize))
// Create a URL that references a to-be-created blob in your
// Azure Storage account's container.

2
weed/replication/sink/b2sink/b2_sink.go

@ -85,7 +85,7 @@ func (g *B2Sink) CreateEntry(key string, entry *filer_pb.Entry) error {
}
totalSize := filer2.TotalSize(entry.Chunks)
chunkViews := filer2.ViewFromChunks(entry.Chunks, 0, int64(totalSize))
chunkViews := filer2.ViewFromChunks(g.filerSource.LookupFileId, entry.Chunks, 0, int64(totalSize))
bucket, err := g.client.Bucket(context.Background(), g.bucket)
if err != nil {

26
weed/replication/sink/filersink/filer_sink.go

@ -167,12 +167,15 @@ func (fs *FilerSink) UpdateEntry(key string, oldEntry *filer_pb.Entry, newParent
glog.V(0).Infof("already replicated %s", key)
} else {
// find out what changed
deletedChunks, newChunks := compareChunks(oldEntry, newEntry)
deletedChunks, newChunks, err := compareChunks(filer2.LookupFn(fs), oldEntry, newEntry)
if err != nil {
return true, fmt.Errorf("replicte %s compare chunks error: %v", key, err)
}
// delete the chunks that are deleted from the source
if deleteIncludeChunks {
// remove the deleted chunks. Actual data deletion happens in filer UpdateEntry FindUnusedFileChunks
existingEntry.Chunks = filer2.MinusChunks(existingEntry.Chunks, deletedChunks)
existingEntry.Chunks = filer2.DoMinusChunks(existingEntry.Chunks, deletedChunks)
}
// replicate the chunks that are new in the source
@ -200,8 +203,21 @@ func (fs *FilerSink) UpdateEntry(key string, oldEntry *filer_pb.Entry, newParent
})
}
func compareChunks(oldEntry, newEntry *filer_pb.Entry) (deletedChunks, newChunks []*filer_pb.FileChunk) {
deletedChunks = filer2.MinusChunks(oldEntry.Chunks, newEntry.Chunks)
newChunks = filer2.MinusChunks(newEntry.Chunks, oldEntry.Chunks)
func compareChunks(lookupFileIdFn filer2.LookupFileIdFunctionType, oldEntry, newEntry *filer_pb.Entry) (deletedChunks, newChunks []*filer_pb.FileChunk, err error) {
aData, aMeta, aErr := filer2.ResolveChunkManifest(lookupFileIdFn, oldEntry.Chunks)
if aErr != nil {
return nil, nil, aErr
}
bData, bMeta, bErr := filer2.ResolveChunkManifest(lookupFileIdFn, newEntry.Chunks)
if bErr != nil {
return nil, nil, bErr
}
deletedChunks = append(deletedChunks, filer2.DoMinusChunks(aData, bData)...)
deletedChunks = append(deletedChunks, filer2.DoMinusChunks(aMeta, bMeta)...)
newChunks = append(newChunks, filer2.DoMinusChunks(bData, aData)...)
newChunks = append(newChunks, filer2.DoMinusChunks(bMeta, aMeta)...)
return
}

2
weed/replication/sink/gcssink/gcs_sink.go

@ -90,7 +90,7 @@ func (g *GcsSink) CreateEntry(key string, entry *filer_pb.Entry) error {
}
totalSize := filer2.TotalSize(entry.Chunks)
chunkViews := filer2.ViewFromChunks(entry.Chunks, 0, int64(totalSize))
chunkViews := filer2.ViewFromChunks(g.filerSource.LookupFileId, entry.Chunks, 0, int64(totalSize))
wc := g.client.Bucket(g.bucket).Object(key).NewWriter(context.Background())

2
weed/replication/sink/s3sink/s3_sink.go

@ -108,7 +108,7 @@ func (s3sink *S3Sink) CreateEntry(key string, entry *filer_pb.Entry) error {
}
totalSize := filer2.TotalSize(entry.Chunks)
chunkViews := filer2.ViewFromChunks(entry.Chunks, 0, int64(totalSize))
chunkViews := filer2.ViewFromChunks(s3sink.filerSource.LookupFileId, entry.Chunks, 0, int64(totalSize))
parts := make([]*s3.CompletedPart, len(chunkViews))

73
weed/server/filer_grpc_server.go

@ -14,6 +14,7 @@ import (
"github.com/chrislusf/seaweedfs/weed/operation"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/pb/master_pb"
"github.com/chrislusf/seaweedfs/weed/storage/needle"
"github.com/chrislusf/seaweedfs/weed/util"
)
@ -137,13 +138,28 @@ func (fs *FilerServer) LookupVolume(ctx context.Context, req *filer_pb.LookupVol
return resp, nil
}
func (fs *FilerServer) lookupFileId(fileId string) (targetUrl string, err error) {
fid, err := needle.ParseFileIdFromString(fileId)
if err != nil {
return "", err
}
locations, found := fs.filer.MasterClient.GetLocations(uint32(fid.VolumeId))
if !found || len(locations) == 0 {
return "", fmt.Errorf("not found volume %d in %s", fid.VolumeId, fileId)
}
return fmt.Sprintf("http://%s/%s", locations[0].Url, fileId), nil
}
func (fs *FilerServer) CreateEntry(ctx context.Context, req *filer_pb.CreateEntryRequest) (resp *filer_pb.CreateEntryResponse, err error) {
glog.V(4).Infof("CreateEntry %v", req)
resp = &filer_pb.CreateEntryResponse{}
chunks, garbages := filer2.CompactFileChunks(req.Entry.Chunks)
chunks, garbage, err2 := fs.cleanupChunks(nil, req.Entry)
if err2 != nil {
return &filer_pb.CreateEntryResponse{}, fmt.Errorf("CreateEntry cleanupChunks %s %s: %v", req.Directory, req.Entry.Name, err2)
}
if req.Entry.Attributes == nil {
glog.V(3).Infof("CreateEntry %s: nil attributes", filepath.Join(req.Directory, req.Entry.Name))
@ -158,7 +174,7 @@ func (fs *FilerServer) CreateEntry(ctx context.Context, req *filer_pb.CreateEntr
}, req.OExcl, req.IsFromOtherCluster)
if createErr == nil {
fs.filer.DeleteChunks(garbages)
fs.filer.DeleteChunks(garbage)
} else {
glog.V(3).Infof("CreateEntry %s: %v", filepath.Join(req.Directory, req.Entry.Name), createErr)
resp.Error = createErr.Error()
@ -177,10 +193,10 @@ func (fs *FilerServer) UpdateEntry(ctx context.Context, req *filer_pb.UpdateEntr
return &filer_pb.UpdateEntryResponse{}, fmt.Errorf("not found %s: %v", fullpath, err)
}
// remove old chunks if not included in the new ones
unusedChunks := filer2.MinusChunks(entry.Chunks, req.Entry.Chunks)
chunks, garbages := filer2.CompactFileChunks(req.Entry.Chunks)
chunks, garbage, err2 := fs.cleanupChunks(entry, req.Entry)
if err2 != nil {
return &filer_pb.UpdateEntryResponse{}, fmt.Errorf("UpdateEntry cleanupChunks %s: %v", fullpath, err2)
}
newEntry := &filer2.Entry{
FullPath: util.JoinPath(req.Directory, req.Entry.Name),
@ -214,8 +230,7 @@ func (fs *FilerServer) UpdateEntry(ctx context.Context, req *filer_pb.UpdateEntr
}
if err = fs.filer.UpdateEntry(ctx, entry, newEntry); err == nil {
fs.filer.DeleteChunks(unusedChunks)
fs.filer.DeleteChunks(garbages)
fs.filer.DeleteChunks(garbage)
} else {
glog.V(3).Infof("UpdateEntry %s: %v", filepath.Join(req.Directory, req.Entry.Name), err)
}
@ -225,6 +240,37 @@ func (fs *FilerServer) UpdateEntry(ctx context.Context, req *filer_pb.UpdateEntr
return &filer_pb.UpdateEntryResponse{}, err
}
func (fs *FilerServer) cleanupChunks(existingEntry *filer2.Entry, newEntry *filer_pb.Entry) (chunks, garbage []*filer_pb.FileChunk, err error) {
chunks = newEntry.Chunks
// remove old chunks if not included in the new ones
if existingEntry != nil {
garbage, err = filer2.MinusChunks(fs.lookupFileId, existingEntry.Chunks, newEntry.Chunks)
if err != nil {
return chunks, nil, fmt.Errorf("MinusChunks: %v", err)
}
}
// files with manifest chunks are usually large and append only, skip calculating covered chunks
var coveredChunks []*filer_pb.FileChunk
if !filer2.HasChunkManifest(newEntry.Chunks) {
chunks, coveredChunks = filer2.CompactFileChunks(fs.lookupFileId, newEntry.Chunks)
garbage = append(garbage, coveredChunks...)
}
chunks, err = filer2.MaybeManifestize(fs.saveAsChunk(
newEntry.Attributes.Replication,
newEntry.Attributes.Collection,
"",
needle.SecondsToTTL(newEntry.Attributes.TtlSec),
false), chunks)
if err != nil {
// not good, but should be ok
glog.V(0).Infof("MaybeManifestize: %v", err)
}
return
}
func (fs *FilerServer) AppendToEntry(ctx context.Context, req *filer_pb.AppendToEntryRequest) (*filer_pb.AppendToEntryResponse, error) {
glog.V(4).Infof("AppendToEntry %v", req)
@ -254,6 +300,17 @@ func (fs *FilerServer) AppendToEntry(ctx context.Context, req *filer_pb.AppendTo
entry.Chunks = append(entry.Chunks, req.Chunks...)
entry.Chunks, err = filer2.MaybeManifestize(fs.saveAsChunk(
entry.Replication,
entry.Collection,
"",
needle.SecondsToTTL(entry.TtlSec),
false), entry.Chunks)
if err != nil {
// not good, but should be ok
glog.V(0).Infof("MaybeManifestize: %v", err)
}
err = fs.filer.CreateEntry(context.Background(), entry, false, false)
return &filer_pb.AppendToEntryResponse{}, err

5
weed/server/filer_server_handlers_write.go

@ -40,7 +40,7 @@ type FilerPostResult struct {
Url string `json:"url,omitempty"`
}
func (fs *FilerServer) assignNewFileInfo(w http.ResponseWriter, r *http.Request, replication, collection, dataCenter, ttlString string, fsync bool) (fileId, urlLocation string, auth security.EncodedJwt, err error) {
func (fs *FilerServer) assignNewFileInfo(replication, collection, dataCenter, ttlString string, fsync bool) (fileId, urlLocation string, auth security.EncodedJwt, err error) {
stats.FilerRequestCounter.WithLabelValues("assign").Inc()
start := time.Now()
@ -67,7 +67,6 @@ func (fs *FilerServer) assignNewFileInfo(w http.ResponseWriter, r *http.Request,
assignResult, ae := operation.Assign(fs.filer.GetMaster(), fs.grpcDialOption, ar, altRequest)
if ae != nil {
glog.Errorf("failing to assign a file id: %v", ae)
writeJsonError(w, r, http.StatusInternalServerError, ae)
err = ae
return
}
@ -114,7 +113,7 @@ func (fs *FilerServer) PostHandler(w http.ResponseWriter, r *http.Request) {
return
}
fileId, urlLocation, auth, err := fs.assignNewFileInfo(w, r, replication, collection, dataCenter, ttlString, fsync)
fileId, urlLocation, auth, err := fs.assignNewFileInfo(replication, collection, dataCenter, ttlString, fsync)
if err != nil || fileId == "" || urlLocation == "" {
glog.V(0).Infof("fail to allocate volume for %s, collection:%s, datacenter:%s", r.URL.Path, collection, dataCenter)

28
weed/server/filer_server_handlers_write_autochunk.go

@ -102,7 +102,7 @@ func (fs *FilerServer) doAutoChunk(ctx context.Context, w http.ResponseWriter, r
limitedReader := io.LimitReader(partReader, int64(chunkSize))
// assign one file id for one chunk
fileId, urlLocation, auth, assignErr := fs.assignNewFileInfo(w, r, replication, collection, dataCenter, ttlString, fsync)
fileId, urlLocation, auth, assignErr := fs.assignNewFileInfo(replication, collection, dataCenter, ttlString, fsync)
if assignErr != nil {
return nil, assignErr
}
@ -132,6 +132,12 @@ func (fs *FilerServer) doAutoChunk(ctx context.Context, w http.ResponseWriter, r
}
}
fileChunks, replyerr = filer2.MaybeManifestize(fs.saveAsChunk(replication, collection, dataCenter, ttlString, fsync), fileChunks)
if replyerr != nil {
glog.V(0).Infof("manifestize %s: %v", r.RequestURI, replyerr)
return
}
path := r.URL.Path
if strings.HasSuffix(path, "/") {
if fileName != "" {
@ -184,3 +190,23 @@ func (fs *FilerServer) doUpload(urlLocation string, w http.ResponseWriter, r *ht
uploadResult, err, _ := operation.Upload(urlLocation, fileName, fs.option.Cipher, limitedReader, false, contentType, pairMap, auth)
return uploadResult, err
}
func (fs *FilerServer) saveAsChunk(replication string, collection string, dataCenter string, ttlString string, fsync bool) filer2.SaveDataAsChunkFunctionType {
return func(reader io.Reader, name string, offset int64) (*filer_pb.FileChunk, string, string, error) {
// assign one file id for one chunk
fileId, urlLocation, auth, assignErr := fs.assignNewFileInfo(replication, collection, dataCenter, ttlString, fsync)
if assignErr != nil {
return nil, "", "", assignErr
}
// upload the chunk to the volume server
uploadResult, uploadErr, _ := operation.Upload(urlLocation, name, fs.option.Cipher, reader, false, "", nil, auth)
if uploadErr != nil {
return nil, "", "", uploadErr
}
return uploadResult.ToPbFileChunk(fileId, offset), collection, replication, nil
}
}

2
weed/server/filer_server_handlers_write_cipher.go

@ -19,7 +19,7 @@ import (
func (fs *FilerServer) encrypt(ctx context.Context, w http.ResponseWriter, r *http.Request,
replication string, collection string, dataCenter string, ttlSeconds int32, ttlString string, fsync bool) (filerResult *FilerPostResult, err error) {
fileId, urlLocation, auth, err := fs.assignNewFileInfo(w, r, replication, collection, dataCenter, ttlString, fsync)
fileId, urlLocation, auth, err := fs.assignNewFileInfo(replication, collection, dataCenter, ttlString, fsync)
if err != nil || fileId == "" || urlLocation == "" {
return nil, fmt.Errorf("fail to allocate volume for %s, collection:%s, datacenter:%s", r.URL.Path, collection, dataCenter)

2
weed/server/webdav_server.go

@ -474,7 +474,7 @@ func (f *WebDavFile) Read(p []byte) (readSize int, err error) {
return 0, io.EOF
}
if f.entryViewCache == nil {
f.entryViewCache = filer2.NonOverlappingVisibleIntervals(f.entry.Chunks)
f.entryViewCache, _ = filer2.NonOverlappingVisibleIntervals(filer2.LookupFn(f.fs), f.entry.Chunks)
f.reader = nil
}
if f.reader == nil {

8
weed/shell/command_volume_fsck.go

@ -11,6 +11,7 @@ import (
"path/filepath"
"sync"
"github.com/chrislusf/seaweedfs/weed/filer2"
"github.com/chrislusf/seaweedfs/weed/operation"
"github.com/chrislusf/seaweedfs/weed/pb/filer_pb"
"github.com/chrislusf/seaweedfs/weed/pb/master_pb"
@ -196,7 +197,12 @@ func (c *commandVolumeFsck) collectFilerFileIds(tempFolder string, volumeIdToSer
files[i.vid].Write(buffer)
}
}, func(entry *filer_pb.FullEntry, outputChan chan interface{}) (err error) {
for _, chunk := range entry.Entry.Chunks {
dChunks, mChunks, resolveErr := filer2.ResolveChunkManifest(filer2.LookupFn(c.env), entry.Entry.Chunks)
if resolveErr != nil {
return nil
}
dChunks = append(dChunks, mChunks...)
for _, chunk := range dChunks {
outputChan <- &Item{
vid: chunk.Fid.VolumeId,
fileKey: chunk.Fid.FileKey,

10
weed/storage/needle/volume_ttl.go

@ -1,11 +1,12 @@
package needle
import (
"fmt"
"strconv"
)
const (
//stored unit types
// stored unit types
Empty byte = iota
Minute
Hour
@ -139,3 +140,10 @@ func (t TTL) Minutes() uint32 {
}
return 0
}
func SecondsToTTL(seconds int32) string {
if seconds == 0 {
return ""
}
return fmt.Sprintf("%dm", seconds/60)
}
Loading…
Cancel
Save