hbase-client/src/main/java/org/apache/hadoop/hbase/client/Hbck.java

   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.client;
  19
  20 import java.io.Closeable;
  21 import java.io.IOException;
  22 import java.util.List;
  23 import java.util.Map;
  24 import java.util.stream.Collectors;
  25 import org.apache.hadoop.hbase.Abortable;
  26 import org.apache.hadoop.hbase.HBaseInterfaceAudience;
  27 import org.apache.hadoop.hbase.ServerName;
  28 import org.apache.hadoop.hbase.master.RegionState;
  29 import org.apache.yetus.audience.InterfaceAudience;
  30
  31 import org.apache.hadoop.hbase.shaded.protobuf.ProtobufUtil;
  32 import org.apache.hadoop.hbase.shaded.protobuf.generated.HBaseProtos;
  33
  34 /**
  35  * Hbck fixup tool APIs. Obtain an instance from {@link Connection#getHbck()} and call
  36  * {@link #close()} when done.
  37  * <p>WARNING: the below methods can damage the cluster. It may leave the cluster in an
  38  * indeterminate state, e.g. region not assigned, or some hdfs files left behind. After running
  39  * any of the below, operators may have to do some clean up on hdfs or schedule some assign
  40  * procedures to get regions back online. DO AT YOUR OWN RISK. For experienced users only.
  41  *
  42  * @see ConnectionFactory
  43  * @since 2.0.2, 2.1.1
  44  */
  45 @InterfaceAudience.LimitedPrivate(HBaseInterfaceAudience.HBCK)
  46 public interface Hbck extends Abortable, Closeable {
  47   /**
  48    * Update table state in Meta only. No procedures are submitted to open/assign or
  49    * close/unassign regions of the table.
  50    * @param state table state
  51    * @return previous state of the table in Meta
  52    */
  53   TableState setTableStateInMeta(TableState state) throws IOException;
  54
  55   /**
  56    * Update region state in Meta only. No procedures are submitted to manipulate the given region or
  57    * any other region from same table.
  58    * @param nameOrEncodedName2State list of all region states to be updated in meta
  59    * @return previous state of the region in Meta
  60    */
  61   Map<String, RegionState.State>
  62     setRegionStateInMeta(Map<String, RegionState.State> nameOrEncodedName2State) throws IOException;
  63
  64   /**
  65    * Like {@link Admin#assign(byte[])} but 'raw' in that it can do more than one Region at a time
  66    * -- good if many Regions to online -- and it will schedule the assigns even in the case where
  67    * Master is initializing (as long as the ProcedureExecutor is up). Does NOT call Coprocessor
  68    * hooks.
  69    * @param override You need to add the override for case where a region has previously been
  70    *              bypassed. When a Procedure has been bypassed, a Procedure will have completed
  71    *              but no other Procedure will be able to make progress on the target entity
  72    *              (intentionally). This override flag will override this fencing mechanism.
  73    * @param encodedRegionNames Region encoded names; e.g. 1588230740 is the hard-coded encoding
  74    *                           for hbase:meta region and de00010733901a05f5a2a3a382e27dd4 is an
  75    *                           example of what a random user-space encoded Region name looks like.
  76    */
  77   List<Long> assigns(List<String> encodedRegionNames, boolean override) throws IOException;
  78
  79   default List<Long> assigns(List<String> encodedRegionNames) throws IOException {
  80     return assigns(encodedRegionNames, false);
  81   }
  82
  83   /**
  84    * Like {@link Admin#unassign(byte[], boolean)} but 'raw' in that it can do more than one Region
  85    * at a time -- good if many Regions to offline -- and it will schedule the assigns even in the
  86    * case where Master is initializing (as long as the ProcedureExecutor is up). Does NOT call
  87    * Coprocessor hooks.
  88    * @param override You need to add the override for case where a region has previously been
  89    *              bypassed. When a Procedure has been bypassed, a Procedure will have completed
  90    *              but no other Procedure will be able to make progress on the target entity
  91    *              (intentionally). This override flag will override this fencing mechanism.
  92    * @param encodedRegionNames Region encoded names; e.g. 1588230740 is the hard-coded encoding
  93    *                           for hbase:meta region and de00010733901a05f5a2a3a382e27dd4 is an
  94    *                           example of what a random user-space encoded Region name looks like.
  95    */
  96   List<Long> unassigns(List<String> encodedRegionNames, boolean override) throws IOException;
  97
  98   default List<Long> unassigns(List<String> encodedRegionNames) throws IOException {
  99     return unassigns(encodedRegionNames, false);
 100   }
 101
 102   /**
 103    * Bypass specified procedure and move it to completion. Procedure is marked completed but
 104    * no actual work is done from the current state/step onwards. Parents of the procedure are
 105    * also marked for bypass.
 106    *
 107    * @param pids of procedures to complete.
 108    * @param waitTime wait time in ms for acquiring lock for a procedure
 109    * @param override if override set to true, we will bypass the procedure even if it is executing.
 110    *   This is for procedures which can't break out during execution (bugs?).
 111    * @param recursive If set, if a parent procedure, we will find and bypass children and then
 112    *   the parent procedure (Dangerous but useful in case where child procedure has been 'lost').
 113    *   Does not always work. Experimental.
 114    * @return true if procedure is marked for bypass successfully, false otherwise
 115    */
 116   List<Boolean> bypassProcedure(List<Long> pids, long waitTime, boolean override, boolean recursive)
 117       throws IOException;
 118
 119   /**
 120    * Use {@link #scheduleServerCrashProcedures(List)} instead.
 121    * @deprecated since 2.2.1. Will removed in 3.0.0.
 122    */
 123   @Deprecated
 124   default List<Long> scheduleServerCrashProcedure(List<HBaseProtos.ServerName> serverNames)
 125       throws IOException {
 126     return scheduleServerCrashProcedures(
 127         serverNames.stream().map(ProtobufUtil::toServerName).collect(Collectors.toList()));
 128   }
 129
 130   List<Long> scheduleServerCrashProcedures(List<ServerName> serverNames) throws IOException;
 131
 132   List<Long> scheduleSCPsForUnknownServers() throws IOException;
 133
 134   /**
 135    * Request HBCK chore to run at master side.
 136    *
 137    * @return <code>true</code> if HBCK chore ran, <code>false</code> if HBCK chore already running
 138    * @throws IOException if a remote or network exception occurs
 139    */
 140   boolean runHbckChore() throws IOException;
 141
 142   /**
 143    * Fix Meta.
 144    */
 145   void fixMeta() throws IOException;
 146 }