/**
* Test whether we can delay the deletion of unknown blocks in DataNode's first several block
* reports.
*/
@Test
public void testPendingDeleteUnknownBlocks() throws Exception {
final int fileNum = 5; // 5 files
final Path[] files = new Path[fileNum];
final DataNodeProperties[] dnprops = new DataNodeProperties[REPLICATION];
// create a group of files, each file contains 1 block
for (int i = 0; i < fileNum; i++) {
files[i] = new Path("/file" + i);
DFSTestUtil.createFile(dfs, files[i], BLOCKSIZE, REPLICATION, i);
}
// wait until all DataNodes have replicas
waitForReplication();
for (int i = REPLICATION - 1; i >= 0; i--) {
dnprops[i] = cluster.stopDataNode(i);
}
Thread.sleep(2000);
// delete 2 files, we still have 3 files remaining so that we can cover
// every DN storage
for (int i = 0; i < 2; i++) {
dfs.delete(files[i], true);
}
// restart NameNode
cluster.restartNameNode(false);
InvalidateBlocks invalidateBlocks =
(InvalidateBlocks)
Whitebox.getInternalState(
cluster.getNamesystem().getBlockManager(), "invalidateBlocks");
InvalidateBlocks mockIb = Mockito.spy(invalidateBlocks);
Mockito.doReturn(1L).when(mockIb).getInvalidationDelay();
Whitebox.setInternalState(
cluster.getNamesystem().getBlockManager(), "invalidateBlocks", mockIb);
Assert.assertEquals(0L, cluster.getNamesystem().getPendingDeletionBlocks());
// restart DataNodes
for (int i = 0; i < REPLICATION; i++) {
cluster.restartDataNode(dnprops[i], true);
}
cluster.waitActive();
for (int i = 0; i < REPLICATION; i++) {
DataNodeTestUtils.triggerBlockReport(cluster.getDataNodes().get(i));
}
Thread.sleep(2000);
// make sure we have received block reports by checking the total block #
Assert.assertEquals(3, cluster.getNamesystem().getBlocksTotal());
Assert.assertEquals(4, cluster.getNamesystem().getPendingDeletionBlocks());
cluster.restartNameNode(true);
Thread.sleep(6000);
Assert.assertEquals(3, cluster.getNamesystem().getBlocksTotal());
Assert.assertEquals(0, cluster.getNamesystem().getPendingDeletionBlocks());
}
/*
* Since NameNode will not persist any locations of the block, addBlock()
* retry call after restart NN should re-select the locations and return to
* client. refer HDFS-5257
*/
@Test
public void testAddBlockRetryShouldReturnBlockWithLocations() throws Exception {
final String src = "/testAddBlockRetryShouldReturnBlockWithLocations";
NamenodeProtocols nameNodeRpc = cluster.getNameNodeRpc();
// create file
nameNodeRpc.create(
src,
FsPermission.getFileDefault(),
"clientName",
new EnumSetWritable<CreateFlag>(EnumSet.of(CreateFlag.CREATE)),
true,
(short) 3,
1024,
null);
// start first addBlock()
LOG.info("Starting first addBlock for " + src);
LocatedBlock lb1 =
nameNodeRpc.addBlock(src, "clientName", null, null, INodeId.GRANDFATHER_INODE_ID, null);
assertTrue("Block locations should be present", lb1.getLocations().length > 0);
cluster.restartNameNode();
nameNodeRpc = cluster.getNameNodeRpc();
LocatedBlock lb2 =
nameNodeRpc.addBlock(src, "clientName", null, null, INodeId.GRANDFATHER_INODE_ID, null);
assertEquals("Blocks are not equal", lb1.getBlock(), lb2.getBlock());
assertTrue("Wrong locations with retry", lb2.getLocations().length > 0);
}
/** Test that the NN re-learns of volume failures after restart. */
@Test
public void testVolFailureStatsPreservedOnNNRestart() throws Exception {
assumeTrue(!System.getProperty("os.name").startsWith("Windows"));
// Bring up two more datanodes that can tolerate 1 failure
cluster.startDataNodes(conf, 2, true, null, null);
cluster.waitActive();
final DatanodeManager dm = cluster.getNamesystem().getBlockManager().getDatanodeManager();
long origCapacity = DFSTestUtil.getLiveDatanodeCapacity(dm);
long dnCapacity = DFSTestUtil.getDatanodeCapacity(dm, 0);
// Fail the first volume on both datanodes (we have to keep the
// third healthy so one node in the pipeline will not fail).
File dn1Vol1 = new File(dataDir, "data" + (2 * 0 + 1));
File dn2Vol1 = new File(dataDir, "data" + (2 * 1 + 1));
assertTrue("Couldn't chmod local vol", FileUtil.setExecutable(dn1Vol1, false));
assertTrue("Couldn't chmod local vol", FileUtil.setExecutable(dn2Vol1, false));
Path file1 = new Path("/test1");
DFSTestUtil.createFile(fs, file1, 1024, (short) 2, 1L);
DFSTestUtil.waitReplication(fs, file1, (short) 2);
// The NN reports two volumes failures
DFSTestUtil.waitForDatanodeStatus(
dm, 3, 0, 2, origCapacity - (1 * dnCapacity), WAIT_FOR_HEARTBEATS);
// After restarting the NN it still see the two failures
cluster.restartNameNode(0);
cluster.waitActive();
DFSTestUtil.waitForDatanodeStatus(
dm, 3, 0, 2, origCapacity - (1 * dnCapacity), WAIT_FOR_HEARTBEATS);
}
@Test
public void testOpenFilesWithRename() throws Exception {
Path path = new Path("/test");
doWriteAndAbort(fs, path);
// check for zero sized blocks
Path fileWithEmptyBlock = new Path("/test/test/test4");
fs.create(fileWithEmptyBlock);
NamenodeProtocols nameNodeRpc = cluster.getNameNodeRpc();
String clientName = fs.getClient().getClientName();
// create one empty block
nameNodeRpc.addBlock(
fileWithEmptyBlock.toString(),
clientName,
null,
null,
HdfsConstants.GRANDFATHER_INODE_ID,
null);
fs.createSnapshot(path, "s2");
fs.rename(new Path("/test/test"), new Path("/test/test-renamed"));
fs.delete(new Path("/test/test-renamed"), true);
NameNode nameNode = cluster.getNameNode();
NameNodeAdapter.enterSafeMode(nameNode, false);
NameNodeAdapter.saveNamespace(nameNode);
NameNodeAdapter.leaveSafeMode(nameNode);
cluster.restartNameNode(true);
}
@Test
public void testParentDirWithUCFileDeleteWithSnapShot() throws Exception {
Path path = new Path("/test");
doWriteAndAbort(fs, path);
// delete parent directory
fs.delete(new Path("/test/test"), true);
cluster.restartNameNode();
}
@Test
public void testUCFileDeleteWithSnapShot() throws Exception {
Path path = new Path("/test");
doWriteAndAbort(fs, path);
// delete files separately
fs.delete(new Path("/test/test/test2"), true);
fs.delete(new Path("/test/test/test3"), true);
cluster.restartNameNode();
}
@Test
public void testPendingDeletion() throws Exception {
final Path foo = new Path("/foo");
DFSTestUtil.createFile(dfs, foo, BLOCKSIZE, REPLICATION, 0);
// restart NN
cluster.restartNameNode(true);
dfs.delete(foo, true);
Assert.assertEquals(0, cluster.getNamesystem().getBlocksTotal());
Assert.assertEquals(REPLICATION, cluster.getNamesystem().getPendingDeletionBlocks());
Thread.sleep(6000);
Assert.assertEquals(0, cluster.getNamesystem().getBlocksTotal());
Assert.assertEquals(0, cluster.getNamesystem().getPendingDeletionBlocks());
}
private void doTestMultipleSnapshots(boolean saveNamespace) throws IOException {
Path path = new Path("/test");
doWriteAndAbort(fs, path);
fs.createSnapshot(path, "s2");
fs.delete(new Path("/test/test"), true);
fs.deleteSnapshot(path, "s2");
cluster.triggerBlockReports();
if (saveNamespace) {
NameNode nameNode = cluster.getNameNode();
NameNodeAdapter.enterSafeMode(nameNode, false);
NameNodeAdapter.saveNamespace(nameNode);
NameNodeAdapter.leaveSafeMode(nameNode);
}
cluster.restartNameNode(true);
}
@Test
public void testWithCheckpoint() throws Exception {
Path path = new Path("/test");
doWriteAndAbort(fs, path);
fs.delete(new Path("/test/test"), true);
NameNode nameNode = cluster.getNameNode();
NameNodeAdapter.enterSafeMode(nameNode, false);
NameNodeAdapter.saveNamespace(nameNode);
NameNodeAdapter.leaveSafeMode(nameNode);
cluster.restartNameNode(true);
// read snapshot file after restart
String test2snapshotPath = Snapshot.getSnapshotPath(path.toString(), "s1/test/test2");
DFSTestUtil.readFile(fs, new Path(test2snapshotPath));
String test3snapshotPath = Snapshot.getSnapshotPath(path.toString(), "s1/test/test3");
DFSTestUtil.readFile(fs, new Path(test3snapshotPath));
}
/**
* Test to simulate interleaved checkpointing by 2 2NNs after a storage directory has been taken
* offline. The first will cause the directory to come back online, but it won't have any valid
* contents. The second 2NN will then try to perform a checkpoint. The NN should not serve up the
* image or edits from the restored (empty) dir.
*/
@Test
public void testCheckpointWithRestoredDirectory() throws IOException {
SecondaryNameNode secondary = null;
try {
cluster = new MiniDFSCluster(0, config, 1, true, false, true, null, null, null, null);
cluster.waitActive();
secondary = new SecondaryNameNode(config);
FSImage fsImage = cluster.getNameNode().getFSImage();
FileSystem fs = cluster.getFileSystem();
Path path1 = new Path("/", "test");
writeFile(fs, path1, 2);
// Take name3 offline
fsImage.getEditLog().removeEditsAndStorageDir(2);
// Simulate a 2NN beginning a checkpoint, but not finishing. This will
// cause name3 to be restored.
cluster.getNameNode().rollEditLog();
// Now another 2NN comes along to do a full checkpoint.
secondary.doCheckpoint();
// The created file should still exist in the in-memory FS state after the
// checkpoint.
assertTrue("File missing after checkpoint", fs.exists(path1));
secondary.shutdown();
// Restart the NN so it reloads the edits from on-disk.
cluster.restartNameNode();
// The created file should still exist after the restart.
assertTrue("path should still exist after restart", fs.exists(path1));
} finally {
if (cluster != null) {
cluster.shutdown();
}
if (secondary != null) {
secondary.shutdown();
}
}
}
/**
* Test case which restarts the standby node in such a way that, when it exits safemode, it will
* want to invalidate a bunch of over-replicated block replicas. Ensures that if we failover at
* this point it won't lose data.
*/
@Test
public void testNNClearsCommandsOnFailoverAfterStartup() throws Exception {
// Make lots of blocks to increase chances of triggering a bug.
DFSTestUtil.createFile(fs, TEST_FILE_PATH, 30 * SMALL_BLOCK, (short) 3, 1L);
banner("Shutting down NN2");
cluster.shutdownNameNode(1);
banner("Setting replication to 1, rolling edit log.");
nn1.getRpcServer().setReplication(TEST_FILE, (short) 1);
nn1.getRpcServer().rollEditLog();
// Start NN2 again. When it starts up, it will see all of the
// blocks as over-replicated, since it has the metadata for
// replication=1, but the DNs haven't yet processed the deletions.
banner("Starting NN2 again.");
cluster.restartNameNode(1);
nn2 = cluster.getNameNode(1);
banner("triggering BRs");
cluster.triggerBlockReports();
// We expect that both NN1 and NN2 will have some number of
// deletions queued up for the DNs.
banner("computing invalidation on nn1");
BlockManagerTestUtil.computeInvalidationWork(nn1.getNamesystem().getBlockManager());
banner("computing invalidation on nn2");
BlockManagerTestUtil.computeInvalidationWork(nn2.getNamesystem().getBlockManager());
// Dump some info for debugging purposes.
banner("Metadata immediately before failover");
doMetasave(nn2);
// Transition nn2 to active even though nn1 still thinks it's active
banner("Failing to NN2 but let NN1 continue to think it's active");
NameNodeAdapter.abortEditLogs(nn1);
NameNodeAdapter.enterSafeMode(nn1, false);
cluster.transitionToActive(1);
// Check that the standby picked up the replication change.
assertEquals(1, nn2.getRpcServer().getFileInfo(TEST_FILE).getReplication());
// Dump some info for debugging purposes.
banner("Metadata immediately after failover");
doMetasave(nn2);
banner("Triggering heartbeats and block reports so that fencing is completed");
cluster.triggerHeartbeats();
cluster.triggerBlockReports();
banner("Metadata after nodes have all block-reported");
doMetasave(nn2);
// The block should no longer be postponed.
assertEquals(0, nn2.getNamesystem().getPostponedMisreplicatedBlocks());
// Wait for NN2 to enact its deletions (replication monitor has to run, etc)
BlockManagerTestUtil.computeInvalidationWork(nn2.getNamesystem().getBlockManager());
HATestUtil.waitForNNToIssueDeletions(nn2);
cluster.triggerHeartbeats();
HATestUtil.waitForDNDeletions(cluster);
cluster.triggerDeletionReports();
assertEquals(0, nn2.getNamesystem().getUnderReplicatedBlocks());
assertEquals(0, nn2.getNamesystem().getPendingReplicationBlocks());
banner("Making sure the file is still readable");
FileSystem fs2 = cluster.getFileSystem(1);
DFSTestUtil.readFile(fs2, TEST_FILE_PATH);
}
protected void doTestRead(Configuration conf, MiniDFSCluster cluster, boolean isStriped)
throws Exception {
final int numDataNodes = cluster.getDataNodes().size();
final NameNode nn = cluster.getNameNode();
final NamenodeProtocols nnProto = nn.getRpcServer();
final BlockManager bm = nn.getNamesystem().getBlockManager();
final BlockTokenSecretManager sm = bm.getBlockTokenSecretManager();
// set a short token lifetime (1 second) initially
SecurityTestUtil.setBlockTokenLifetime(sm, 1000L);
Path fileToRead = new Path(FILE_TO_READ);
FileSystem fs = cluster.getFileSystem();
byte[] expected = generateBytes(FILE_SIZE);
createFile(fs, fileToRead, expected);
/*
* setup for testing expiration handling of cached tokens
*/
// read using blockSeekTo(). Acquired tokens are cached in in1
FSDataInputStream in1 = fs.open(fileToRead);
assertTrue(checkFile1(in1, expected));
// read using blockSeekTo(). Acquired tokens are cached in in2
FSDataInputStream in2 = fs.open(fileToRead);
assertTrue(checkFile1(in2, expected));
// read using fetchBlockByteRange(). Acquired tokens are cached in in3
FSDataInputStream in3 = fs.open(fileToRead);
assertTrue(checkFile2(in3, expected));
/*
* testing READ interface on DN using a BlockReader
*/
DFSClient client = null;
try {
client = new DFSClient(new InetSocketAddress("localhost", cluster.getNameNodePort()), conf);
} finally {
if (client != null) client.close();
}
List<LocatedBlock> locatedBlocks =
nnProto.getBlockLocations(FILE_TO_READ, 0, FILE_SIZE).getLocatedBlocks();
LocatedBlock lblock = locatedBlocks.get(0); // first block
// verify token is not expired
assertFalse(isBlockTokenExpired(lblock));
// read with valid token, should succeed
tryRead(conf, lblock, true);
/*
* wait till myToken and all cached tokens in in1, in2 and in3 expire
*/
while (!isBlockTokenExpired(lblock)) {
try {
Thread.sleep(10);
} catch (InterruptedException ignored) {
}
}
/*
* continue testing READ interface on DN using a BlockReader
*/
// verify token is expired
assertTrue(isBlockTokenExpired(lblock));
// read should fail
tryRead(conf, lblock, false);
// use a valid new token
bm.setBlockToken(lblock, BlockTokenIdentifier.AccessMode.READ);
// read should succeed
tryRead(conf, lblock, true);
// use a token with wrong blockID
long rightId = lblock.getBlock().getBlockId();
long wrongId = rightId + 1;
lblock.getBlock().setBlockId(wrongId);
bm.setBlockToken(lblock, BlockTokenIdentifier.AccessMode.READ);
lblock.getBlock().setBlockId(rightId);
// read should fail
tryRead(conf, lblock, false);
// use a token with wrong access modes
bm.setBlockToken(lblock, BlockTokenIdentifier.AccessMode.WRITE);
// read should fail
tryRead(conf, lblock, false);
// set a long token lifetime for future tokens
SecurityTestUtil.setBlockTokenLifetime(sm, 600 * 1000L);
/*
* testing that when cached tokens are expired, DFSClient will re-fetch
* tokens transparently for READ.
*/
// confirm all tokens cached in in1 are expired by now
List<LocatedBlock> lblocks = DFSTestUtil.getAllBlocks(in1);
for (LocatedBlock blk : lblocks) {
assertTrue(isBlockTokenExpired(blk));
}
// verify blockSeekTo() is able to re-fetch token transparently
in1.seek(0);
assertTrue(checkFile1(in1, expected));
// confirm all tokens cached in in2 are expired by now
List<LocatedBlock> lblocks2 = DFSTestUtil.getAllBlocks(in2);
for (LocatedBlock blk : lblocks2) {
assertTrue(isBlockTokenExpired(blk));
}
// verify blockSeekTo() is able to re-fetch token transparently (testing
// via another interface method)
if (isStriped) {
// striped block doesn't support seekToNewSource
in2.seek(0);
} else {
assertTrue(in2.seekToNewSource(0));
}
assertTrue(checkFile1(in2, expected));
// confirm all tokens cached in in3 are expired by now
List<LocatedBlock> lblocks3 = DFSTestUtil.getAllBlocks(in3);
for (LocatedBlock blk : lblocks3) {
assertTrue(isBlockTokenExpired(blk));
}
// verify fetchBlockByteRange() is able to re-fetch token transparently
assertTrue(checkFile2(in3, expected));
/*
* testing that after datanodes are restarted on the same ports, cached
* tokens should still work and there is no need to fetch new tokens from
* namenode. This test should run while namenode is down (to make sure no
* new tokens can be fetched from namenode).
*/
// restart datanodes on the same ports that they currently use
assertTrue(cluster.restartDataNodes(true));
cluster.waitActive();
assertEquals(numDataNodes, cluster.getDataNodes().size());
cluster.shutdownNameNode(0);
// confirm tokens cached in in1 are still valid
lblocks = DFSTestUtil.getAllBlocks(in1);
for (LocatedBlock blk : lblocks) {
assertFalse(isBlockTokenExpired(blk));
}
// verify blockSeekTo() still works (forced to use cached tokens)
in1.seek(0);
assertTrue(checkFile1(in1, expected));
// confirm tokens cached in in2 are still valid
lblocks2 = DFSTestUtil.getAllBlocks(in2);
for (LocatedBlock blk : lblocks2) {
assertFalse(isBlockTokenExpired(blk));
}
// verify blockSeekTo() still works (forced to use cached tokens)
if (isStriped) {
in2.seek(0);
} else {
in2.seekToNewSource(0);
}
assertTrue(checkFile1(in2, expected));
// confirm tokens cached in in3 are still valid
lblocks3 = DFSTestUtil.getAllBlocks(in3);
for (LocatedBlock blk : lblocks3) {
assertFalse(isBlockTokenExpired(blk));
}
// verify fetchBlockByteRange() still works (forced to use cached tokens)
assertTrue(checkFile2(in3, expected));
/*
* testing that when namenode is restarted, cached tokens should still
* work and there is no need to fetch new tokens from namenode. Like the
* previous test, this test should also run while namenode is down. The
* setup for this test depends on the previous test.
*/
// restart the namenode and then shut it down for test
cluster.restartNameNode(0);
cluster.shutdownNameNode(0);
// verify blockSeekTo() still works (forced to use cached tokens)
in1.seek(0);
assertTrue(checkFile1(in1, expected));
// verify again blockSeekTo() still works (forced to use cached tokens)
if (isStriped) {
in2.seek(0);
} else {
in2.seekToNewSource(0);
}
assertTrue(checkFile1(in2, expected));
// verify fetchBlockByteRange() still works (forced to use cached tokens)
assertTrue(checkFile2(in3, expected));
/*
* testing that after both namenode and datanodes got restarted (namenode
* first, followed by datanodes), DFSClient can't access DN without
* re-fetching tokens and is able to re-fetch tokens transparently. The
* setup of this test depends on the previous test.
*/
// restore the cluster and restart the datanodes for test
cluster.restartNameNode(0);
assertTrue(cluster.restartDataNodes(true));
cluster.waitActive();
assertEquals(numDataNodes, cluster.getDataNodes().size());
// shutdown namenode so that DFSClient can't get new tokens from namenode
cluster.shutdownNameNode(0);
// verify blockSeekTo() fails (cached tokens become invalid)
in1.seek(0);
assertFalse(checkFile1(in1, expected));
// verify fetchBlockByteRange() fails (cached tokens become invalid)
assertFalse(checkFile2(in3, expected));
// restart the namenode to allow DFSClient to re-fetch tokens
cluster.restartNameNode(0);
// verify blockSeekTo() works again (by transparently re-fetching
// tokens from namenode)
in1.seek(0);
assertTrue(checkFile1(in1, expected));
if (isStriped) {
in2.seek(0);
} else {
in2.seekToNewSource(0);
}
assertTrue(checkFile1(in2, expected));
// verify fetchBlockByteRange() works again (by transparently
// re-fetching tokens from namenode)
assertTrue(checkFile2(in3, expected));
/*
* testing that when datanodes are restarted on different ports, DFSClient
* is able to re-fetch tokens transparently to connect to them
*/
// restart datanodes on newly assigned ports
assertTrue(cluster.restartDataNodes(false));
cluster.waitActive();
assertEquals(numDataNodes, cluster.getDataNodes().size());
// verify blockSeekTo() is able to re-fetch token transparently
in1.seek(0);
assertTrue(checkFile1(in1, expected));
// verify blockSeekTo() is able to re-fetch token transparently
if (isStriped) {
in2.seek(0);
} else {
in2.seekToNewSource(0);
}
assertTrue(checkFile1(in2, expected));
// verify fetchBlockByteRange() is able to re-fetch token transparently
assertTrue(checkFile2(in3, expected));
}