Copilot commented on code in PR #10617:
URL: https://github.com/apache/ozone/pull/10617#discussion_r3557007650
##########
hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/container/AbstractContainerReportHandler.java:
##########
@@ -362,6 +362,24 @@ private boolean updateContainerState(final DatanodeDetails
datanode,
}
}
+ /**
+ * Apply a container lifecycle state transition, but only on the leader SCM.
+ * On a follower the underlying {@code containerManager.updateContainerState}
+ * is a Ratis write and would throw {@code NotLeaderException}, which would
+ * abort {@code processContainerReplica} and skip recording the replica
+ * location. Skipping the state change on a follower is safe: the leader
Review Comment:
The leader-only guard added for updateContainerState avoids
NotLeaderException for lifecycle events, but this handler still performs other
replicated container state transitions during report handling (eg
transitionDeletingOrDeletedToTargetState in the DELETING/DELETED branch). On a
follower that call can also throw NotLeaderException (an IOException) and abort
replica processing, reintroducing the same “skip recording replica location”
failure mode for deleting containers.
##########
hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/ha/SCMStateMachine.java:
##########
@@ -386,16 +401,91 @@ public void notifyTermIndexUpdated(long term, long index)
{
}
}
- if (currentLeaderTerm.get() == term) {
- // This means after a restart, all pending transactions have been
applied.
+ // As committed entries are applied (e.g. a restarted follower catching
up),
+ // start the datanode protocol server once we are caught up with the
leader's
+ // committed index. No-op once the server has already been started.
+ tryStartDNServerAndRefreshSafeMode();
+ }
+
+ /**
+ * Start the DatanodeProtocolServer and re-evaluate safe-mode rules, but only
+ * when this SCM is safe to accept datanode reports: it is the leader, or it
+ * is a follower whose state machine has caught up with the leader's
committed
+ * log. Guarded by {@code isStateMachineReady} (CAS) so the non-idempotent
+ * {@code DatanodeProtocolServer.start()} runs exactly once.
+ *
+ * <p>In HA mode {@link StorageContainerManager#start()} deliberately does
not
+ * start the datanode protocol server; it is deferred to here so datanode
+ * container reports are processed against the up-to-date container/pipeline
+ * state rather than a stale, mid-replay snapshot.
+ */
+ private void tryStartDNServerAndRefreshSafeMode() {
+ if (isStateMachineReady.get()) {
+ return;
+ }
+ if (scm.getScmContext().isLeader() || isFollowerCaughtUp()) {
if (isStateMachineReady.compareAndSet(false, true)) {
- // Refresh Safemode rules state if not already done.
+ scm.getDatanodeProtocolServer().start();
scm.getScmSafeModeManager().refreshAndValidate();
}
- currentLeaderTerm.set(-1L);
}
}
+ /**
+ * @return true if this follower's last applied index has reached the
leader's
+ * committed index captured when it (re)joined, i.e. all transactions the
+ * leader had committed at that point have been replayed. Comparing against a
+ * fixed target avoids chasing the leader's ever-advancing live commit index.
+ */
+ private boolean isFollowerCaughtUp() {
+ try {
+ long target = leaderCommitIndexOnStart;
+ if (target < 0) {
+ // Not captured at leader-change time yet; capture the leader's current
+ // commit index once here so we still compare against a fixed target.
+ target = getLeaderCommitIndex();
+ if (target < 0) {
+ LOG.warn("Leader commit index not available yet");
+ return false;
+ }
Review Comment:
isFollowerCaughtUp() logs a WARN every time leader commit info is
unavailable. During startup/catch-up, tryStartDNServerAndRefreshSafeMode() can
be invoked repeatedly (eg via notifyTermIndexUpdated), so this can flood logs
even though it’s a normal transient condition. Consider lowering to DEBUG (or
rate-limiting) until the leader commit index is known.
##########
hadoop-ozone/integration-test/src/test/java/org/apache/hadoop/hdds/scm/TestSCMFollowerCatchupWithContainerReport.java:
##########
@@ -0,0 +1,336 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.hadoop.hdds.scm;
+
+import static java.nio.charset.StandardCharsets.UTF_8;
+import static
org.apache.hadoop.hdds.protocol.proto.HddsProtos.LifeCycleState.CLOSED;
+import static
org.apache.hadoop.hdds.protocol.proto.HddsProtos.ReplicationFactor.THREE;
+import static org.junit.jupiter.api.Assertions.assertArrayEquals;
+import static org.junit.jupiter.api.Assertions.assertEquals;
+import static org.junit.jupiter.api.Assertions.assertFalse;
+
+import java.io.IOException;
+import java.util.LinkedHashSet;
+import java.util.Set;
+import java.util.concurrent.TimeUnit;
+import java.util.function.BooleanSupplier;
+import org.apache.hadoop.hdds.client.RatisReplicationConfig;
+import org.apache.hadoop.hdds.conf.OzoneConfiguration;
+import org.apache.hadoop.hdds.protocol.proto.HddsProtos.LifeCycleState;
+import org.apache.hadoop.hdds.scm.container.ContainerID;
+import org.apache.hadoop.hdds.scm.server.StorageContainerManager;
+import org.apache.hadoop.ozone.MiniOzoneCluster;
+import org.apache.hadoop.ozone.MiniOzoneHAClusterImpl;
+import org.apache.hadoop.ozone.TestDataUtil;
+import org.apache.hadoop.ozone.client.ObjectStore;
+import org.apache.hadoop.ozone.client.OzoneBucket;
+import org.apache.hadoop.ozone.client.OzoneClient;
+import org.apache.hadoop.ozone.client.OzoneKeyDetails;
+import org.apache.hadoop.ozone.client.OzoneVolume;
+import org.apache.hadoop.ozone.client.io.OzoneInputStream;
+import org.apache.ozone.test.GenericTestUtils;
+import org.junit.jupiter.api.AfterAll;
+import org.junit.jupiter.api.BeforeAll;
+import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.Timeout;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+/**
+ * Verifies that a follower SCM correctly rebuilds container replica locations
+ * for containers that were <em>created</em> while it was offline. After the
+ * follower restarts, catches up its Ratis log, and is promoted to leader, all
+ * such containers must still have the expected replica count and all keys must
+ * be readable.
+ *
+ * <p>This is the create-while-down counterpart of
+ * {@code TestSCMFollowerCatchupWithContainerClose} (HDDS-14989). It exercises
Review Comment:
Class-level Javadoc references TestSCMFollowerCatchupWithContainerClose, but
that class does not exist in this module. This makes the documentation
misleading and harder to maintain.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
