View Javadoc

1   /**
2    * Licensed to the Apache Software Foundation (ASF) under one
3    * or more contributor license agreements.  See the NOTICE file
4    * distributed with this work for additional information
5    * regarding copyright ownership.  The ASF licenses this file
6    * to you under the Apache License, Version 2.0 (the
7    * "License"); you may not use this file except in compliance
8    * with the License.  You may obtain a copy of the License at
9    *
10   *     http://www.apache.org/licenses/LICENSE-2.0
11   *
12   * Unless required by applicable law or agreed to in writing, software
13   * distributed under the License is distributed on an "AS IS" BASIS,
14   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15   * See the License for the specific language governing permissions and
16   * limitations under the License.
17   */
18  package org.apache.hadoop.hbase.master.snapshot;
19  
20  import java.io.IOException;
21  import java.util.HashSet;
22  import java.util.List;
23  import java.util.Set;
24  
25  import org.apache.commons.logging.Log;
26  import org.apache.commons.logging.LogFactory;
27  import org.apache.hadoop.hbase.classification.InterfaceAudience;
28  import org.apache.hadoop.hbase.HRegionInfo;
29  import org.apache.hadoop.hbase.ServerName;
30  import org.apache.hadoop.hbase.client.RegionReplicaUtil;
31  import org.apache.hadoop.hbase.errorhandling.ForeignException;
32  import org.apache.hadoop.hbase.master.MasterServices;
33  import org.apache.hadoop.hbase.procedure.Procedure;
34  import org.apache.hadoop.hbase.procedure.ProcedureCoordinator;
35  import org.apache.hadoop.hbase.protobuf.generated.HBaseProtos.SnapshotDescription;
36  import org.apache.hadoop.hbase.snapshot.HBaseSnapshotException;
37  import org.apache.hadoop.hbase.util.Pair;
38  
39  import com.google.common.collect.Lists;
40  
41  /**
42   * Handle the master side of taking a snapshot of an online table, regardless of snapshot type.
43   * Uses a {@link Procedure} to run the snapshot across all the involved region servers.
44   * @see ProcedureCoordinator
45   */
46  @InterfaceAudience.Private
47  public class EnabledTableSnapshotHandler extends TakeSnapshotHandler {
48  
49    private static final Log LOG = LogFactory.getLog(EnabledTableSnapshotHandler.class);
50    private final ProcedureCoordinator coordinator;
51  
52    public EnabledTableSnapshotHandler(SnapshotDescription snapshot, MasterServices master,
53        final SnapshotManager manager) throws IOException {
54      super(snapshot, master, manager);
55      this.coordinator = manager.getCoordinator();
56    }
57  
58    @Override
59    public EnabledTableSnapshotHandler prepare() throws Exception {
60      return (EnabledTableSnapshotHandler) super.prepare();
61    }
62  
63    // TODO consider switching over to using regionnames, rather than server names. This would allow
64    // regions to migrate during a snapshot, and then be involved when they are ready. Still want to
65    // enforce a snapshot time constraints, but lets us be potentially a bit more robust.
66  
67    /**
68     * This method kicks off a snapshot procedure. Other than that it hangs around for various phases
69     * to complete.
70     */
71    @Override
72    protected void snapshotRegions(List<Pair<HRegionInfo, ServerName>> regions) throws IOException {
73      Set<String> regionServers = new HashSet<String>(regions.size());
74      for (Pair<HRegionInfo, ServerName> region : regions) {
75        if (region != null && region.getFirst() != null && region.getSecond() != null) {
76          HRegionInfo hri = region.getFirst();
77          if (hri.isOffline() && (hri.isSplit() || hri.isSplitParent())) continue;
78          regionServers.add(region.getSecond().toString());
79        }
80      }
81  
82      // start the snapshot on the RS
83      Procedure proc = coordinator.startProcedure(this.monitor, this.snapshot.getName(),
84        this.snapshot.toByteArray(), Lists.newArrayList(regionServers));
85      if (proc == null) {
86        String msg = "Failed to submit distributed procedure for snapshot '"
87            + snapshot.getName() + "'";
88        LOG.error(msg);
89        throw new HBaseSnapshotException(msg);
90      }
91  
92      try {
93        // wait for the snapshot to complete.  A timer thread is kicked off that should cancel this
94        // if it takes too long.
95        proc.waitForCompleted();
96        LOG.info("Done waiting - online snapshot for " + this.snapshot.getName());
97  
98        // Take the offline regions as disabled
99        for (Pair<HRegionInfo, ServerName> region : regions) {
100         HRegionInfo regionInfo = region.getFirst();
101         if (regionInfo.isOffline() && (regionInfo.isSplit() || regionInfo.isSplitParent()) &&
102             RegionReplicaUtil.isDefaultReplica(regionInfo)) {
103           LOG.info("Take disabled snapshot of offline region=" + regionInfo);
104           snapshotDisabledRegion(regionInfo);
105         }
106       }
107     } catch (InterruptedException e) {
108       ForeignException ee =
109           new ForeignException("Interrupted while waiting for snapshot to finish", e);
110       monitor.receive(ee);
111       Thread.currentThread().interrupt();
112     } catch (ForeignException e) {
113       monitor.receive(e);
114     }
115   }
116 }