This is an automated email from the ASF dual-hosted git repository.
rpuch pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/ignite-3.git
The following commit(s) were added to refs/heads/main by this push:
new 2c586ec04d4 IGNITE-26117 Use non-durable Raft storages destruction for
table destruction (#6363)
2c586ec04d4 is described below
commit 2c586ec04d43629fc14bb107bc3f0c264381fa98
Author: Roman Puchkovskiy <roman.puchkovs...@gmail.com>
AuthorDate: Tue Aug 5 16:59:23 2025 +0400
IGNITE-26117 Use non-durable Raft storages destruction for table
destruction (#6363)
---
.../ignite/internal/replicator/ReplicaManager.java | 18 ++++++++-
.../internal/table/distributed/TableManager.java | 44 ++++++++++++++++------
2 files changed, 49 insertions(+), 13 deletions(-)
diff --git
a/modules/replicator/src/main/java/org/apache/ignite/internal/replicator/ReplicaManager.java
b/modules/replicator/src/main/java/org/apache/ignite/internal/replicator/ReplicaManager.java
index 9e9df0d4aa5..4910d788894 100644
---
a/modules/replicator/src/main/java/org/apache/ignite/internal/replicator/ReplicaManager.java
+++
b/modules/replicator/src/main/java/org/apache/ignite/internal/replicator/ReplicaManager.java
@@ -1225,12 +1225,26 @@ public class ReplicaManager extends
AbstractEventProducer<LocalReplicaEvent, Loc
*/
public void destroyReplicationProtocolStoragesOnStartup(ReplicationGroupId
replicaGrpId)
throws NodeStoppingException {
- RaftNodeId raftNodeId = new RaftNodeId(replicaGrpId, new
Peer(localNodeConsistentId));
// We use 'isVolatileStorage' of false because on startup it's not a
problem if the value is wrong. If it actually
// was volatile, the log storage is already destroyed on an earlier
phase of node startup, so we will just issue an excessive
// log storage destruction request, and it's not a problem as
persistent log storage with same table/zone ID cannot exist
// if the storage was volatile.
- RaftGroupOptions groupOptions = groupOptionsForPartition(false, null);
+ destroyReplicationProtocolStorages(replicaGrpId, false);
+ }
+
+ /**
+ * Destroys replication protocol storages for the given group ID.
+ *
+ * <p>No durability guarantees are provided. If a node crashes, the
storage may come to life.
+ *
+ * @param replicaGrpId Replication group ID.
+ * @param isVolatileStorage is table storage volatile?
+ * @throws NodeStoppingException If the node is being stopped.
+ */
+ public void destroyReplicationProtocolStorages(ReplicationGroupId
replicaGrpId, boolean isVolatileStorage)
+ throws NodeStoppingException {
+ RaftNodeId raftNodeId = new RaftNodeId(replicaGrpId, new
Peer(localNodeConsistentId));
+ RaftGroupOptions groupOptions =
groupOptionsForPartition(isVolatileStorage, null);
((Loza) raftManager).destroyRaftNodeStorages(raftNodeId, groupOptions);
}
diff --git
a/modules/table/src/main/java/org/apache/ignite/internal/table/distributed/TableManager.java
b/modules/table/src/main/java/org/apache/ignite/internal/table/distributed/TableManager.java
index 76b7429b14f..8a06ea77424 100644
---
a/modules/table/src/main/java/org/apache/ignite/internal/table/distributed/TableManager.java
+++
b/modules/table/src/main/java/org/apache/ignite/internal/table/distributed/TableManager.java
@@ -2942,7 +2942,7 @@ public class TableManager implements
IgniteTablesInternal, IgniteComponent {
var tablePartitionId = new
TablePartitionId(internalTable.tableId(), partitionId);
stopReplicaAndDestroyFutures[partitionId] =
resourcesUnloadFuture
- .thenCompose(v ->
stopAndDestroyTablePartition(tablePartitionId, table));
+ .thenCompose(v ->
stopAndDestroyTablePartition(tablePartitionId, table, true));
}
return allOf(stopReplicaAndDestroyFutures).whenComplete((res,
th) -> {
@@ -2972,13 +2972,17 @@ public class TableManager implements
IgniteTablesInternal, IgniteComponent {
TableImpl table = tables.get(tablePartitionId.tableId());
assert table != null : tablePartitionId;
- return stopAndDestroyTablePartition(tablePartitionId,
table);
+ return stopAndDestroyTablePartition(tablePartitionId,
table, false);
});
}
- private CompletableFuture<Void>
stopAndDestroyTablePartition(TablePartitionId tablePartitionId, TableImpl
table) {
+ private CompletableFuture<Void> stopAndDestroyTablePartition(
+ TablePartitionId tablePartitionId,
+ TableImpl table,
+ boolean destroyingWholeTable
+ ) {
return stopTablePartition(tablePartitionId, table)
- .thenComposeAsync(v ->
destroyPartitionStorages(tablePartitionId, table), ioExecutor);
+ .thenComposeAsync(v ->
destroyPartitionStorages(tablePartitionId, table, destroyingWholeTable),
ioExecutor);
}
/**
@@ -3009,7 +3013,7 @@ public class TableManager implements
IgniteTablesInternal, IgniteComponent {
return replicaMgr.weakStopReplica(
tablePartitionId,
WeakReplicaStopReason.RESTART,
- () -> stopAndDestroyTablePartition(tablePartitionId, table)
+ () -> stopAndDestroyTablePartition(tablePartitionId, table,
false)
);
}
@@ -3043,7 +3047,11 @@ public class TableManager implements
IgniteTablesInternal, IgniteComponent {
});
}
- private CompletableFuture<Void> destroyPartitionStorages(TablePartitionId
tablePartitionId, TableImpl table) {
+ private CompletableFuture<Void> destroyPartitionStorages(
+ TablePartitionId tablePartitionId,
+ TableImpl table,
+ boolean destroyingWholeTable
+ ) {
InternalTable internalTable = table.internalTable();
int partitionId = tablePartitionId.partitionId();
@@ -3065,22 +3073,36 @@ public class TableManager implements
IgniteTablesInternal, IgniteComponent {
destroyFutures.add(runAsync(() ->
internalTable.txStateStorage().destroyPartitionStorage(partitionId),
ioExecutor));
}
- destroyFutures.add(runAsync(() ->
destroyReplicationProtocolStorages(tablePartitionId, table), ioExecutor));
+ destroyFutures.add(
+ runAsync(() -> {
+ // No need for durability guarantees if destruction
reliability is guaranteed by destroying table storages
+ // on startup.
+ destroyReplicationProtocolStorages(tablePartitionId,
table, !destroyingWholeTable);
+ }, ioExecutor)
+ );
}
// TODO: IGNITE-24926 - reduce set in localPartsByTableId after
storages destruction.
return allOf(destroyFutures.toArray(new CompletableFuture[]{}));
}
- private void destroyReplicationProtocolStorages(TablePartitionId
tablePartitionId, TableImpl table) {
+ private void destroyReplicationProtocolStorages(TablePartitionId
tablePartitionId, TableImpl table, boolean destroyDurably) {
var internalTbl = (InternalTableImpl) table.internalTable();
- destroyReplicationProtocolStorages(tablePartitionId,
internalTbl.storage().isVolatile());
+ destroyReplicationProtocolStorages(tablePartitionId,
internalTbl.storage().isVolatile(), destroyDurably);
}
- private void destroyReplicationProtocolStorages(TablePartitionId
tablePartitionId, boolean isVolatileStorage) {
+ private void destroyReplicationProtocolStorages(
+ TablePartitionId tablePartitionId,
+ boolean isVolatileStorage,
+ boolean destroyDurably
+ ) {
try {
-
replicaMgr.destroyReplicationProtocolStoragesDurably(tablePartitionId,
isVolatileStorage);
+ if (destroyDurably) {
+
replicaMgr.destroyReplicationProtocolStoragesDurably(tablePartitionId,
isVolatileStorage);
+ } else {
+
replicaMgr.destroyReplicationProtocolStorages(tablePartitionId,
isVolatileStorage);
+ }
} catch (NodeStoppingException e) {
throw new IgniteInternalException(NODE_STOPPING_ERR, e);
}
