use of org.elasticsearch.indices.recovery.RecoveriesCollection.RecoveryRef 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.elasticsearch.indices.recovery.RecoveriesCollection.RecoveryRef in project crate by crate.
the class PeerRecoveryTargetService method doRecovery.
private void doRecovery(final long recoveryId) {
final StartRecoveryRequest request;
final RecoveryState.Timer timer;
CancellableThreads cancellableThreads;
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();
timer = recoveryTarget.state().getTimer();
cancellableThreads = recoveryTarget.cancellableThreads();
try {
assert recoveryTarget.sourceNode() != null : "can not do a recovery without a source node";
LOGGER.trace("{} preparing shard for peer recovery", recoveryTarget.shardId());
recoveryTarget.indexShard().prepareForIndexRecovery();
final long startingSeqNo = recoveryTarget.indexShard().recoverLocallyUpToGlobalCheckpoint();
assert startingSeqNo == UNASSIGNED_SEQ_NO || recoveryTarget.state().getStage() == RecoveryState.Stage.TRANSLOG : "unexpected recovery stage [" + recoveryTarget.state().getStage() + "] starting seqno [ " + startingSeqNo + "]";
request = getStartRecoveryRequest(LOGGER, clusterService.localNode(), recoveryTarget, startingSeqNo);
} 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;
}
}
Consumer<Exception> handleException = e -> {
if (LOGGER.isTraceEnabled()) {
LOGGER.trace(() -> new ParameterizedMessage("[{}][{}] Got exception on recovery", request.shardId().getIndex().getName(), request.shardId().id()), e);
}
Throwable cause = SQLExceptions.unwrap(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 = SQLExceptions.unwrap(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);
};
try {
LOGGER.trace("{} starting recovery from {}", request.shardId(), request.sourceNode());
cancellableThreads.executeIO(() -> transportService.sendRequest(request.sourceNode(), PeerRecoverySourceService.Actions.START_RECOVERY, request, new TransportResponseHandler<RecoveryResponse>() {
@Override
public void handleResponse(RecoveryResponse recoveryResponse) {
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);
}
}
@Override
public void handleException(TransportException e) {
handleException.accept(e);
}
@Override
public String executor() {
// we do some heavy work like refreshes in the response so fork off to the generic threadpool
return ThreadPool.Names.GENERIC;
}
@Override
public RecoveryResponse read(StreamInput in) throws IOException {
return new RecoveryResponse(in);
}
}));
} catch (CancellableThreads.ExecutionCancelledException e) {
LOGGER.trace("recovery cancelled", e);
} catch (Exception e) {
handleException.accept(e);
}
}
Aggregations