keith-turner commented on code in PR #3327: URL: https://github.com/apache/accumulo/pull/3327#discussion_r1174031465
########## server/manager/src/main/java/org/apache/accumulo/manager/tableOps/goal/SetHostingGoal.java: ########## @@ -0,0 +1,139 @@ +/* + * 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 + * + * https://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.accumulo.manager.tableOps.goal; + +import java.util.ArrayList; +import java.util.List; + +import org.apache.accumulo.core.client.AccumuloException; +import org.apache.accumulo.core.client.admin.TabletHostingGoal; +import org.apache.accumulo.core.clientImpl.ClientContext; +import org.apache.accumulo.core.clientImpl.thrift.TableOperation; +import org.apache.accumulo.core.data.Key; +import org.apache.accumulo.core.data.NamespaceId; +import org.apache.accumulo.core.data.PartialKey; +import org.apache.accumulo.core.data.Range; +import org.apache.accumulo.core.data.TableId; +import org.apache.accumulo.core.dataImpl.KeyExtent; +import org.apache.accumulo.core.dataImpl.thrift.TRange; +import org.apache.accumulo.core.fate.Repo; +import org.apache.accumulo.core.metadata.schema.Ample.TabletsMutator; +import org.apache.accumulo.core.metadata.schema.TabletMetadata; +import org.apache.accumulo.core.metadata.schema.TabletMetadata.Location; +import org.apache.accumulo.core.metadata.schema.TabletsMetadata; +import org.apache.accumulo.manager.Manager; +import org.apache.accumulo.manager.tableOps.ManagerRepo; +import org.apache.accumulo.manager.tableOps.Utils; +import org.apache.hadoop.io.Text; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +public class SetHostingGoal extends ManagerRepo { + + private static final long serialVersionUID = 1L; + private static final Logger LOG = LoggerFactory.getLogger(SetHostingGoal.class); + + private final TableId tableId; + private final NamespaceId namespaceId; + private final TRange range; + private final TabletHostingGoal goal; + + public SetHostingGoal(TableId tableId, NamespaceId namespaceId, TRange range, + TabletHostingGoal goal) { + this.tableId = tableId; + this.namespaceId = namespaceId; + this.range = range; + this.goal = goal; + } + + @Override + public long isReady(long tid, Manager manager) throws Exception { + return Utils.reserveNamespace(manager, namespaceId, tid, false, true, + TableOperation.SET_HOSTING_GOAL) + + Utils.reserveTable(manager, tableId, tid, true, true, TableOperation.SET_HOSTING_GOAL); + } + + @Override + public Repo<Manager> call(long tid, Manager manager) throws Exception { + final Range r = new Range(range); + final List<KeyExtent> extents = findExtentsForRange(manager.getContext(), tableId, r); + LOG.debug("Setting tablet hosting goal to {} requested for: {} ", goal, extents); + try (TabletsMutator mutator = manager.getContext().getAmple().mutateTablets()) { + extents.forEach(e -> mutator.mutateTablet(e).setHostingGoal(goal).mutate()); Review Comment: I think the closest example is bulk import. It makes rpcs to the tablet and this helps avoid concurrent problems with splits. The tablet on the tablet server writes the data to the metadata table. The fate op keeps scanning until it sees the data in all tablets in the range. The table lock helps avoid concurrent problems with destructive operations like delete table. So it avoids problems with concurrent operations with a combo of repeated metadata scans, tablet RPCs, and a table lock. So in the old model of how Accumulo worked we probably would have added an RPC to the tablet to write the new hosting goal. -- 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]
