siddhantsangwan commented on code in PR #3802:
URL: https://github.com/apache/ozone/pull/3802#discussion_r991879550
##########
hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/container/replication/RatisContainerReplicaCount.java:
##########
@@ -247,7 +248,26 @@ public boolean isSufficientlyReplicated() {
}
/**
- * Return true is the container is over replicated. Decommission and
+ * Return true if the container is sufficiently replicated. Decommissioning
+ * and Decommissioned containers are ignored in this check, assuming they
will
+ * eventually be removed from the cluster.
+ * This check ignores inflight additions, if includePendingAdd is false,
+ * otherwise it will assume they complete ok.
+ *
+ * @return True if the container is sufficiently replicated and False
+ * otherwise.
+ */
+ public boolean isSufficientlyReplicated(boolean includePendingAdd) {
+ // Positive for under-rep, negative for over-rep
+ int delta = missingReplicas();
+ if (includePendingAdd) {
+ delta -= inFlightAdd;
+ }
+ return delta <= 0;
+ }
Review Comment:
Are we deliberately not considering pending deletes here?
##########
hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/container/replication/health/RatisReplicationCheckHandler.java:
##########
@@ -0,0 +1,198 @@
+/*
+ * 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
+ * <p/>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p/>
+ * 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.container.replication.health;
+
+import org.apache.hadoop.hdds.protocol.DatanodeDetails;
+import org.apache.hadoop.hdds.scm.ContainerPlacementStatus;
+import org.apache.hadoop.hdds.scm.PlacementPolicy;
+import org.apache.hadoop.hdds.scm.container.ContainerInfo;
+import org.apache.hadoop.hdds.scm.container.ContainerReplica;
+import org.apache.hadoop.hdds.scm.container.ReplicationManagerReport;
+import org.apache.hadoop.hdds.scm.container.replication.ContainerCheckRequest;
+import org.apache.hadoop.hdds.scm.container.replication.ContainerHealthResult;
+import org.apache.hadoop.hdds.scm.container.replication.ContainerReplicaOp;
+import
org.apache.hadoop.hdds.scm.container.replication.RatisContainerReplicaCount;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import static
org.apache.hadoop.hdds.protocol.proto.HddsProtos.ReplicationType.RATIS;
+
+/**
+ * Class to determine the health state of a Ratis Container. Given the
container
+ * and current replica details, along with replicas pending add and delete,
+ * this class will return a ContainerHealthResult indicating if the container
+ * is healthy, or under / over replicated etc.
+ */
+public class RatisReplicationCheckHandler extends AbstractCheck {
+ public static final Logger LOG =
+ LoggerFactory.getLogger(RatisReplicationCheckHandler.class);
+
+ /**
+ * PlacementPolicy which is used to identify where a container
+ * should be replicated.
+ */
+ private final PlacementPolicy ratisContainerPlacement;
+
+ public RatisReplicationCheckHandler(PlacementPolicy containerPlacement) {
+ this.ratisContainerPlacement = containerPlacement;
+ }
+
+ @Override
+ public boolean handle(ContainerCheckRequest request) {
+ if (request.getContainerInfo().getReplicationType() != RATIS) {
+ // This handler is only for Ratis containers.
+ return false;
+ }
+ ReplicationManagerReport report = request.getReport();
+ ContainerInfo container = request.getContainerInfo();
+ ContainerHealthResult health = checkHealth(request);
+ if (health.getHealthState() == ContainerHealthResult.HealthState.HEALTHY) {
+ // If the container is healthy, there is nothing else to do in this
+ // handler so return as unhandled so any further handlers will be tried.
+ return false;
+ }
+ if (health.getHealthState()
+ == ContainerHealthResult.HealthState.UNDER_REPLICATED) {
+ report.incrementAndSample(
+ ReplicationManagerReport.HealthState.UNDER_REPLICATED,
+ container.containerID());
+ ContainerHealthResult.UnderReplicatedHealthResult underHealth
+ = ((ContainerHealthResult.UnderReplicatedHealthResult) health);
+ if (underHealth.isUnrecoverable()) {
+ report.incrementAndSample(ReplicationManagerReport.HealthState.MISSING,
+ container.containerID());
+ }
+ if (underHealth.isMisReplicated()) {
+ report.incrementAndSample(
+ ReplicationManagerReport.HealthState.MIS_REPLICATED,
+ container.containerID());
+ }
+ // TODO - if it is unrecoverable, should we return false to other
+ // handlers can be tried?
+ if (!underHealth.isUnrecoverable() &&
+ (underHealth.isMisReplicatedAfterPending() ||
+ !underHealth.isSufficientlyReplicatedAfterPending())) {
+ request.getReplicationQueue().enqueue(underHealth);
+ }
+ return true;
+ }
Review Comment:
NIT: Let's add a new line after this brace for better readability.
--
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]
