use of org.apache.lucene.store.AlreadyClosedException in project elasticsearch by elastic.
the class PeerRecoveryTargetService method doRecovery.
private void doRecovery(final long recoveryId) {
final StartRecoveryRequest request;
final CancellableThreads cancellableThreads;
final RecoveryState.Timer timer;
try (RecoveryRef recoveryRef = onGoingRecoveries.getRecovery(recoveryId)) {
if (recoveryRef == null) {
logger.trace("not running recovery with id [{}] - can not find it (probably finished)", recoveryId);
return;
}
final RecoveryTarget recoveryTarget = recoveryRef.target();
cancellableThreads = recoveryTarget.cancellableThreads();
timer = recoveryTarget.state().getTimer();
try {
assert recoveryTarget.sourceNode() != null : "can not do a recovery without a source node";
request = getStartRecoveryRequest(recoveryTarget);
logger.trace("{} preparing shard for peer recovery", recoveryTarget.shardId());
recoveryTarget.indexShard().prepareForIndexRecovery();
} catch (final Exception e) {
// this will be logged as warning later on...
logger.trace("unexpected error while preparing shard for peer recovery, failing recovery", e);
onGoingRecoveries.failRecovery(recoveryId, new RecoveryFailedException(recoveryTarget.state(), "failed to prepare shard for recovery", e), true);
return;
}
}
try {
logger.trace("{} starting recovery from {}", request.shardId(), request.sourceNode());
final AtomicReference<RecoveryResponse> responseHolder = new AtomicReference<>();
cancellableThreads.execute(() -> responseHolder.set(transportService.submitRequest(request.sourceNode(), PeerRecoverySourceService.Actions.START_RECOVERY, request, new FutureTransportResponseHandler<RecoveryResponse>() {
@Override
public RecoveryResponse newInstance() {
return new RecoveryResponse();
}
}).txGet()));
final RecoveryResponse recoveryResponse = responseHolder.get();
final TimeValue recoveryTime = new TimeValue(timer.time());
// do this through ongoing recoveries to remove it from the collection
onGoingRecoveries.markRecoveryAsDone(recoveryId);
if (logger.isTraceEnabled()) {
StringBuilder sb = new StringBuilder();
sb.append('[').append(request.shardId().getIndex().getName()).append(']').append('[').append(request.shardId().id()).append("] ");
sb.append("recovery completed from ").append(request.sourceNode()).append(", took[").append(recoveryTime).append("]\n");
sb.append(" phase1: recovered_files [").append(recoveryResponse.phase1FileNames.size()).append("]").append(" with " + "total_size of [").append(new ByteSizeValue(recoveryResponse.phase1TotalSize)).append("]").append(", took [").append(timeValueMillis(recoveryResponse.phase1Time)).append("], throttling_wait [").append(timeValueMillis(recoveryResponse.phase1ThrottlingWaitTime)).append(']').append("\n");
sb.append(" : reusing_files [").append(recoveryResponse.phase1ExistingFileNames.size()).append("] with " + "total_size of [").append(new ByteSizeValue(recoveryResponse.phase1ExistingTotalSize)).append("]\n");
sb.append(" phase2: start took [").append(timeValueMillis(recoveryResponse.startTime)).append("]\n");
sb.append(" : recovered [").append(recoveryResponse.phase2Operations).append("]").append(" transaction log " + "operations").append(", took [").append(timeValueMillis(recoveryResponse.phase2Time)).append("]").append("\n");
logger.trace("{}", sb);
} else {
logger.debug("{} recovery done from [{}], took [{}]", request.shardId(), request.sourceNode(), recoveryTime);
}
} catch (CancellableThreads.ExecutionCancelledException e) {
logger.trace("recovery cancelled", e);
} catch (Exception e) {
if (logger.isTraceEnabled()) {
logger.trace((Supplier<?>) () -> new ParameterizedMessage("[{}][{}] Got exception on recovery", request.shardId().getIndex().getName(), request.shardId().id()), e);
}
Throwable cause = ExceptionsHelper.unwrapCause(e);
if (cause instanceof CancellableThreads.ExecutionCancelledException) {
// this can also come from the source wrapped in a RemoteTransportException
onGoingRecoveries.failRecovery(recoveryId, new RecoveryFailedException(request, "source has canceled the recovery", cause), false);
return;
}
if (cause instanceof RecoveryEngineException) {
// unwrap an exception that was thrown as part of the recovery
cause = cause.getCause();
}
// do it twice, in case we have double transport exception
cause = ExceptionsHelper.unwrapCause(cause);
if (cause instanceof RecoveryEngineException) {
// unwrap an exception that was thrown as part of the recovery
cause = cause.getCause();
}
if (cause instanceof IllegalIndexShardStateException || cause instanceof IndexNotFoundException || cause instanceof ShardNotFoundException) {
// if the target is not ready yet, retry
retryRecovery(recoveryId, "remote shard not ready", recoverySettings.retryDelayStateSync(), recoverySettings.activityTimeout());
return;
}
if (cause instanceof DelayRecoveryException) {
retryRecovery(recoveryId, cause, recoverySettings.retryDelayStateSync(), recoverySettings.activityTimeout());
return;
}
if (cause instanceof ConnectTransportException) {
logger.debug("delaying recovery of {} for [{}] due to networking error [{}]", request.shardId(), recoverySettings.retryDelayNetwork(), cause.getMessage());
retryRecovery(recoveryId, cause.getMessage(), recoverySettings.retryDelayNetwork(), recoverySettings.activityTimeout());
return;
}
if (cause instanceof AlreadyClosedException) {
onGoingRecoveries.failRecovery(recoveryId, new RecoveryFailedException(request, "source shard is closed", cause), false);
return;
}
onGoingRecoveries.failRecovery(recoveryId, new RecoveryFailedException(request, e), true);
}
}
use of org.apache.lucene.store.AlreadyClosedException in project elasticsearch by elastic.
the class RemoteClusterConnectionTests method testCloseWhileConcurrentlyConnecting.
public void testCloseWhileConcurrentlyConnecting() throws IOException, InterruptedException, BrokenBarrierException {
List<DiscoveryNode> knownNodes = new CopyOnWriteArrayList<>();
try (MockTransportService seedTransport = startTransport("seed_node", knownNodes, Version.CURRENT);
MockTransportService seedTransport1 = startTransport("seed_node_1", knownNodes, Version.CURRENT);
MockTransportService discoverableTransport = startTransport("discoverable_node", knownNodes, Version.CURRENT)) {
DiscoveryNode seedNode = seedTransport.getLocalDiscoNode();
DiscoveryNode seedNode1 = seedTransport1.getLocalDiscoNode();
knownNodes.add(seedTransport.getLocalDiscoNode());
knownNodes.add(discoverableTransport.getLocalDiscoNode());
knownNodes.add(seedTransport1.getLocalDiscoNode());
Collections.shuffle(knownNodes, random());
List<DiscoveryNode> seedNodes = Arrays.asList(seedNode1, seedNode);
Collections.shuffle(seedNodes, random());
try (MockTransportService service = MockTransportService.createNewService(Settings.EMPTY, Version.CURRENT, threadPool, null)) {
service.start();
service.acceptIncomingRequests();
try (RemoteClusterConnection connection = new RemoteClusterConnection(Settings.EMPTY, "test-cluster", seedNodes, service, Integer.MAX_VALUE, n -> true)) {
int numThreads = randomIntBetween(4, 10);
Thread[] threads = new Thread[numThreads];
CyclicBarrier barrier = new CyclicBarrier(numThreads + 1);
for (int i = 0; i < threads.length; i++) {
final int numConnectionAttempts = randomIntBetween(10, 100);
threads[i] = new Thread() {
@Override
public void run() {
try {
barrier.await();
CountDownLatch latch = new CountDownLatch(numConnectionAttempts);
for (int i = 0; i < numConnectionAttempts; i++) {
AtomicReference<RuntimeException> executed = new AtomicReference<>();
ActionListener<Void> listener = ActionListener.wrap(x -> {
if (executed.compareAndSet(null, new RuntimeException())) {
latch.countDown();
} else {
throw new AssertionError("shit's been called twice", executed.get());
}
}, x -> {
if (executed.compareAndSet(null, new RuntimeException())) {
latch.countDown();
} else {
throw new AssertionError("shit's been called twice", executed.get());
}
if (x instanceof RejectedExecutionException || x instanceof AlreadyClosedException || x instanceof CancellableThreads.ExecutionCancelledException) {
} else {
throw new AssertionError(x);
}
});
connection.updateSeedNodes(seedNodes, listener);
}
latch.await();
} catch (Exception ex) {
throw new AssertionError(ex);
}
}
};
threads[i].start();
}
barrier.await();
connection.close();
}
}
}
}
use of org.apache.lucene.store.AlreadyClosedException in project elasticsearch by elastic.
the class UnicastZenPing method ping.
/**
* a variant of {@link #ping(Consumer, TimeValue)}, but allows separating the scheduling duration
* from the duration used for request level time outs. This is useful for testing
*/
protected void ping(final Consumer<PingCollection> resultsConsumer, final TimeValue scheduleDuration, final TimeValue requestDuration) {
final List<DiscoveryNode> seedNodes;
try {
seedNodes = resolveHostsLists(unicastZenPingExecutorService, logger, configuredHosts, limitPortCounts, transportService, UNICAST_NODE_PREFIX, resolveTimeout);
} catch (InterruptedException e) {
throw new RuntimeException(e);
}
seedNodes.addAll(hostsProvider.buildDynamicNodes());
final DiscoveryNodes nodes = contextProvider.nodes();
// add all possible master nodes that were active in the last known cluster configuration
for (ObjectCursor<DiscoveryNode> masterNode : nodes.getMasterNodes().values()) {
seedNodes.add(masterNode.value);
}
final ConnectionProfile connectionProfile = ConnectionProfile.buildSingleChannelProfile(TransportRequestOptions.Type.REG, requestDuration, requestDuration);
final PingingRound pingingRound = new PingingRound(pingingRoundIdGenerator.incrementAndGet(), seedNodes, resultsConsumer, nodes.getLocalNode(), connectionProfile);
activePingingRounds.put(pingingRound.id(), pingingRound);
final AbstractRunnable pingSender = new AbstractRunnable() {
@Override
public void onFailure(Exception e) {
if (e instanceof AlreadyClosedException == false) {
logger.warn("unexpected error while pinging", e);
}
}
@Override
protected void doRun() throws Exception {
sendPings(requestDuration, pingingRound);
}
};
threadPool.generic().execute(pingSender);
threadPool.schedule(TimeValue.timeValueMillis(scheduleDuration.millis() / 3), ThreadPool.Names.GENERIC, pingSender);
threadPool.schedule(TimeValue.timeValueMillis(scheduleDuration.millis() / 3 * 2), ThreadPool.Names.GENERIC, pingSender);
threadPool.schedule(scheduleDuration, ThreadPool.Names.GENERIC, new AbstractRunnable() {
@Override
protected void doRun() throws Exception {
finishPingingRound(pingingRound);
}
@Override
public void onFailure(Exception e) {
logger.warn("unexpected error while finishing pinging round", e);
}
});
}
use of org.apache.lucene.store.AlreadyClosedException in project elasticsearch by elastic.
the class ShadowEngineTests method testFailEngineOnCorruption.
public void testFailEngineOnCorruption() throws IOException {
ParsedDocument doc = testParsedDocument("1", "test", null, testDocumentWithTextField(), B_1, null);
primaryEngine.index(indexForDoc(doc));
primaryEngine.flush();
MockDirectoryWrapper leaf = DirectoryUtils.getLeaf(replicaEngine.config().getStore().directory(), MockDirectoryWrapper.class);
leaf.setRandomIOExceptionRate(1.0);
leaf.setRandomIOExceptionRateOnOpen(1.0);
try {
replicaEngine.refresh("foo");
fail("exception expected");
} catch (Exception ex) {
}
try {
Engine.Searcher searchResult = replicaEngine.acquireSearcher("test");
MatcherAssert.assertThat(searchResult, EngineSearcherTotalHitsMatcher.engineSearcherTotalHits(1));
MatcherAssert.assertThat(searchResult, EngineSearcherTotalHitsMatcher.engineSearcherTotalHits(new TermQuery(new Term("value", "test")), 1));
searchResult.close();
fail("exception expected");
} catch (AlreadyClosedException ex) {
// all is well
}
}
use of org.apache.lucene.store.AlreadyClosedException in project elasticsearch by elastic.
the class InternalEngineTests method testForceMergeAndClose.
public void testForceMergeAndClose() throws IOException, InterruptedException {
int numIters = randomIntBetween(2, 10);
for (int j = 0; j < numIters; j++) {
try (Store store = createStore()) {
final InternalEngine engine = createEngine(store, createTempDir());
final CountDownLatch startGun = new CountDownLatch(1);
final CountDownLatch indexed = new CountDownLatch(1);
Thread thread = new Thread() {
@Override
public void run() {
try {
try {
startGun.await();
} catch (InterruptedException e) {
throw new RuntimeException(e);
}
int i = 0;
while (true) {
int numDocs = randomIntBetween(1, 20);
for (int j = 0; j < numDocs; j++) {
i++;
ParsedDocument doc = testParsedDocument(Integer.toString(i), "test", null, testDocument(), B_1, null);
Engine.Index index = indexForDoc(doc);
engine.index(index);
}
engine.refresh("test");
indexed.countDown();
try {
engine.forceMerge(randomBoolean(), 1, false, randomBoolean(), randomBoolean());
} catch (IOException e) {
return;
}
}
} catch (AlreadyClosedException ex) {
// fine
} catch (IOException e) {
throw new AssertionError(e);
}
}
};
thread.start();
startGun.countDown();
int someIters = randomIntBetween(1, 10);
for (int i = 0; i < someIters; i++) {
engine.forceMerge(randomBoolean(), 1, false, randomBoolean(), randomBoolean());
}
indexed.await();
IOUtils.close(engine);
thread.join();
}
}
}
Aggregations