Skip to content

[segment replication]Introducing common Replication interfaces for segment replication and recovery code paths #3234

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 10 commits into from
May 23, 2022
Merged

[segment replication]Introducing common Replication interfaces for segment replication and recovery code paths #3234

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
9d381c0
RecoveryState inherits from ReplicationState + RecoveryTarget inherit…
Poojita-Raj May 3, 2022
ca0106a
Refactoring: mixedClusterVersion error fix + move Stage to Replicatio…
Poojita-Raj May 10, 2022
ba51c17
pull ReplicationListener into a top level class + add javadocs + addr…
Poojita-Raj May 12, 2022
d7fc756
fix javadoc
Poojita-Raj May 13, 2022
b84c184
review changes
Poojita-Raj May 17, 2022
8447624
Refactoring the hierarchy relationship between repl and recovery
Poojita-Raj May 18, 2022
5edeae3
style fix
Poojita-Raj May 18, 2022
47379df
move package common under replication
Poojita-Raj May 18, 2022
233acbc
rename to replication
Poojita-Raj May 18, 2022
50c4e6e
rename and doc changes
Poojita-Raj May 19, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion server/src/internalClusterTest/java/org/opensearch/indices/recovery/IndexRecoveryIT.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,8 +101,8 @@
import org.opensearch.indices.IndicesService;
import org.opensearch.indices.NodeIndicesStats;
import org.opensearch.indices.analysis.AnalysisModule;
import org.opensearch.indices.recovery.RecoveryState.Stage;
import org.opensearch.indices.replication.common.ReplicationLuceneIndex;
import org.opensearch.indices.recovery.RecoveryState.Stage;
import org.opensearch.node.NodeClosedException;
import org.opensearch.node.RecoverySettingsChunkSizePlugin;
import org.opensearch.plugins.AnalysisPlugin;
Expand Down
11 changes: 6 additions & 5 deletions server/src/main/java/org/opensearch/index/shard/IndexShard.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -157,6 +157,7 @@
import org.opensearch.indices.cluster.IndicesClusterStateService;
import org.opensearch.indices.recovery.PeerRecoveryTargetService;
import org.opensearch.indices.recovery.RecoveryFailedException;
import org.opensearch.indices.recovery.RecoveryListener;
import org.opensearch.indices.recovery.RecoveryState;
import org.opensearch.indices.recovery.RecoveryTarget;
import org.opensearch.repositories.RepositoriesService;
Expand Down Expand Up @@ -2876,7 +2877,7 @@ protected Engine getEngineOrNull() {
public void startRecovery(
RecoveryState recoveryState,
PeerRecoveryTargetService recoveryTargetService,
PeerRecoveryTargetService.RecoveryListener recoveryListener,
RecoveryListener recoveryListener,
RepositoriesService repositoriesService,
Consumer<MappingMetadata> mappingUpdateConsumer,
IndicesService indicesService
Expand Down Expand Up @@ -2909,7 +2910,7 @@ public void startRecovery(
recoveryTargetService.startRecovery(this, recoveryState.getSourceNode(), recoveryListener);
} catch (Exception e) {
failShard("corrupted preexisting index", e);
recoveryListener.onRecoveryFailure(recoveryState, new RecoveryFailedException(recoveryState, null, e), true);
recoveryListener.onFailure(recoveryState, new RecoveryFailedException(recoveryState, null, e), true);
}
break;
case SNAPSHOT:
Expand Down Expand Up @@ -2984,15 +2985,15 @@ public void startRecovery(
private void executeRecovery(
String reason,
RecoveryState recoveryState,
PeerRecoveryTargetService.RecoveryListener recoveryListener,
RecoveryListener recoveryListener,
CheckedConsumer<ActionListener<Boolean>, Exception> action
) {
markAsRecovering(reason, recoveryState); // mark the shard as recovering on the cluster state thread
threadPool.generic().execute(ActionRunnable.wrap(ActionListener.wrap(r -> {
if (r) {
recoveryListener.onRecoveryDone(recoveryState);
recoveryListener.onDone(recoveryState);
}
}, e -> recoveryListener.onRecoveryFailure(recoveryState, new RecoveryFailedException(recoveryState, null, e), true)), action));
}, e -> recoveryListener.onFailure(recoveryState, new RecoveryFailedException(recoveryState, null, e), true)), action));
}

/**
Expand Down
3 changes: 2 additions & 1 deletion server/src/main/java/org/opensearch/indices/IndicesService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -136,6 +136,7 @@
import org.opensearch.indices.fielddata.cache.IndicesFieldDataCache;
import org.opensearch.indices.mapper.MapperRegistry;
import org.opensearch.indices.recovery.PeerRecoveryTargetService;
import org.opensearch.indices.recovery.RecoveryListener;
import org.opensearch.indices.recovery.RecoveryState;
import org.opensearch.node.Node;
import org.opensearch.plugins.IndexStorePlugin;
Expand Down Expand Up @@ -839,7 +840,7 @@ public synchronized void verifyIndexMetadata(IndexMetadata metadata, IndexMetada
public IndexShard createShard(
final ShardRouting shardRouting,
final PeerRecoveryTargetService recoveryTargetService,
final PeerRecoveryTargetService.RecoveryListener recoveryListener,
final RecoveryListener recoveryListener,
final RepositoriesService repositoriesService,
final Consumer<IndexShard.ShardFailure> onShardFailure,
final Consumer<ShardId> globalCheckpointSyncer,
Expand Down
42 changes: 10 additions & 32 deletions server/src/main/java/org/opensearch/indices/cluster/IndicesClusterStateService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,8 +78,9 @@
import org.opensearch.indices.IndicesService;
import org.opensearch.indices.recovery.PeerRecoverySourceService;
import org.opensearch.indices.recovery.PeerRecoveryTargetService;
import org.opensearch.indices.recovery.RecoveryFailedException;
import org.opensearch.indices.recovery.RecoveryListener;
import org.opensearch.indices.recovery.RecoveryState;
import org.opensearch.indices.replication.common.ReplicationState;
import org.opensearch.repositories.RepositoriesService;
import org.opensearch.search.SearchService;
import org.opensearch.snapshots.SnapshotShardsService;
Expand Down Expand Up @@ -624,7 +625,7 @@ private void createShard(DiscoveryNodes nodes, RoutingTable routingTable, ShardR
indicesService.createShard(
shardRouting,
recoveryTargetService,
new RecoveryListener(shardRouting, primaryTerm),
new RecoveryListener(shardRouting, primaryTerm, this),
repositoriesService,
failedShardHandler,
globalCheckpointSyncer,
Expand Down Expand Up @@ -739,39 +740,16 @@ private static DiscoveryNode findSourceNodeForPeerRecovery(
return sourceNode;
}

private class RecoveryListener implements PeerRecoveryTargetService.RecoveryListener {

/**
* ShardRouting with which the shard was created
*/
private final ShardRouting shardRouting;

/**
* Primary term with which the shard was created
*/
private final long primaryTerm;

private RecoveryListener(final ShardRouting shardRouting, final long primaryTerm) {
this.shardRouting = shardRouting;
this.primaryTerm = primaryTerm;
}

@Override
public void onRecoveryDone(final RecoveryState state) {
shardStateAction.shardStarted(shardRouting, primaryTerm, "after " + state.getRecoverySource(), SHARD_STATE_ACTION_LISTENER);
}

@Override
public void onRecoveryFailure(RecoveryState state, RecoveryFailedException e, boolean sendShardFailure) {
handleRecoveryFailure(shardRouting, sendShardFailure, e);
}
}

// package-private for testing
synchronized void handleRecoveryFailure(ShardRouting shardRouting, boolean sendShardFailure, Exception failure) {
public synchronized void handleRecoveryFailure(ShardRouting shardRouting, boolean sendShardFailure, Exception failure) {
failAndRemoveShard(shardRouting, sendShardFailure, "failed recovery", failure, clusterService.state());
}

public void handleRecoveryDone(ReplicationState state, ShardRouting shardRouting, long primaryTerm) {
RecoveryState RecState = (RecoveryState) state;
Copy link
Member

@kartg kartg May 19, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think you can change the method parameter type to RecoveryState, and this casting can be removed

shardStateAction.shardStarted(shardRouting, primaryTerm, "after " + RecState.getRecoverySource(), SHARD_STATE_ACTION_LISTENER);
}

private void failAndRemoveShard(
ShardRouting shardRouting,
boolean sendShardFailure,
Expand Down Expand Up @@ -1004,7 +982,7 @@ U createIndex(IndexMetadata indexMetadata, List<IndexEventListener> builtInIndex
T createShard(
ShardRouting shardRouting,
PeerRecoveryTargetService recoveryTargetService,
PeerRecoveryTargetService.RecoveryListener recoveryListener,
RecoveryListener recoveryListener,
RepositoriesService repositoriesService,
Consumer<IndexShard.ShardFailure> onShardFailure,
Consumer<ShardId> globalCheckpointSyncer,
Expand Down
Loading