目录
5.2.1. hbase.master.info.port 4
5.2.2. hbase.master.info.bindAddress 4
1. 前言
本文将HBase-0.98.0安装在Hadoop-2.4.0上，关于Hadoop-2.4.0的安装，请参见《Hadoop-2.4.0分布式安装手册》一文。安装环境为64位SuSE-Linux 10.1版本。
本文将在HBase官方提供的quickstart.html文件的指导下进行，在docs/getting_started目录下可找到quickstart.html，或直接浏览在线的：http://hbase.apache.org/book/quickstart.html。
安装使用外置的ZooKeeper，有关ZooKeeper的安装，请参见《ZooKeeper-3.4.6分布式安装指南》一文。
关于分布式安装，请浏览：http://hbase.apache.org/book/standalone_dist.html#distributed，关于HBase使用外置的ZooKeeper配置，请浏览：http://hbase.apache.org/book/zookeeper.html。
所有在线的文档，均会出现在二进制安装包解压后的docs目录下。本文的安装环境为64位SuSE 10.1 Linux。
2. 约定
假设将Hadoop-2.4.0安装在/data/hadoop/current目录，而/data/hadoop/current实际是到/data/hadoop/hadoop-2.4.0的软链接。
HBase安装目录为/data/hadoop/hbase，而/data/hadoop/hbase实际是到hbase-0.98.0-hadoop2的软链接。
建议JDK为1.7版本，在JDK1.8上可以成功安装Hadoop-2.4.0，但基于JDK1.8编译Hadoop-2.4.0则会遇到困难。
3. 相关端口
|
2888 |
ZooKeeper，如果是Leader，用来监听Follower的连接 |
|
3888 |
ZooKeeper，用于Leader选举 |
|
2181 |
ZooKeeper，用来监听客户端的连接 |
|
60010 |
hbase.master.info.port，HMaster的http端口 |
|
60000 |
hbase.master.port，HMaster的RPC端口 |
|
60030 |
hbase.regionserver.info.port，HRegionServer的http端口 |
|
60020 |
hbase.regionserver.port，HRegionServer的RPC端口 |
|
8080 |
hbase.rest.port，HBase REST server的端口 |
4. 下载HBase
官网：http://hbase.apache.org/，在这里即可找到下载HBase的链接。
下载国内映像站点：http://mirror.bit.edu.cn/apache/hbase/，HBase-0.98.0版本的下载网址：http://mirror.bit.edu.cn/apache/hbase/hbase-0.98.0/。选择下载hbase-0.98.0-hadoop2-bin.tar.gz。
5. 安装步骤
5.1. 修改conf/regionservers
regionservers类似于Hadoop的slaves文件，不需要在RegionServer机器上执行些修改。
将所有HRegionServers的IP或主机名一行一行的例举在在regionservers文件中，注意必须一行一个，不能一行多个。本文配置如下：
|
hadoop@VM_40_171_sles10_64:~/hbase/conf> cat regionservers 10.12.154.77 10.12.154.78 10.12.154.79 |
5.2. 修改conf/hbase-site.xml
需要在所有机器上做同样的操作，可以借助scp命令，先配置好一台，然后复制过去，如：scp hbase-site.xml hadoop@10.12.154.79:/data/hadoop/hbase/conf/。
hbase-site.xml是HBase的配置文件。默认的hbase-site.xml是空的，如下所示：
|
<?xml version="1.0"?> <?xml-stylesheet type="text/xsl" href="configuration.xsl"?> <!-- /** * * 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. */ --> <configuration> </configuration> |
没关系，就用它。不要用docs目录下的hbase-default.xml，这个会让你看得难受。
编辑hbase-site.xml，添加如下内容（摘自standalone_dist.html，搜索“Fully-distributed”）：
|
<configuration> <property> <name>hbase.rootdir</name> <value>hdfs://172.25.40.171:9001/hbase</value> <description>The directory shared by RegionServers.</description> </property> <property> <name>hbase.cluster.distributed</name> <value>true</value> <description>The mode the cluster will be in. Possible values are false: standalone and pseudo-distributed setups with managed Zookeeper true: fully-distributed with unmanaged Zookeeper Quorum (see hbase-env.sh) </description> </property> <property> <name>hbase.zookeeper.quorum</name> <value>DEVNET-154-77,DEVNET-154-70,DEVNET-154-79</value> <description>Comma separated list of servers in the ZooKeeper Quorum. For example, "host1.mydomain.com,host2.mydomain.com,host3.mydomain.com". By default this is set to localhost for local and pseudo-distributed modes of operation. For a fully-distributed setup, this should be set to a full list of ZooKeeper quorum servers. If HBASE_MANAGES_ZK is set in hbase-env.sh this is the list of servers which we will start/stop ZooKeeper on. </description> </property> <property> <name>hbase.master.maxclockskew</name> <value>600000</value> <description>Time(ms) difference of regionserver from master</description> </property> </configuration> |
“hbase.zookeeper.quorum”可以填写IP列表。hdfs://172.25.40.171:9001对应hdfs-site.xml中的“dfs.namenode.rpc-address”。“hbase.zookeeper.quorum”配置为ZooKeeper集群各节点主机名或IP。
更多的信息，可以浏览：http://hbase.apache.org/book/config.files.html。
5.2.1. hbase.master.info.port
用于指定HMaster的http端口。
5.2.2. hbase.master.info.bindAddress
用于指定HMaster的http的IP地址，如果不设定该值，可能使用IPv6地址。
5.3. 修改conf/hbase-env.sh
需要在所有机器上做同样的操作，可以借助scp命令，先配置好一台，然后复制过去，如：scp hbase-site.xml hadoop@10.12.154.79:/data/hadoop/hbase/conf/，修改内容如下：
1) 设置JAVA_HOME
|
# The java implementation to use. Java 1.6 required. export JAVA_HOME=/data/jdk |
上述/data/jdk是JDK的安装目录。
2) 设置HBASE_MANAGES_ZK
|
# Tell HBase whether it should manage it's own instance of Zookeeper or not. export HBASE_MANAGES_ZK=false |
如果HBASE_MANAGES_ZK值为true，则表示使用HBase自带的ZooKeeper，建议单独部署ZooKeeper，这样便于ZooKeeper同时为其它系统提供服务。
3) 设置HBASE_CLASSPATH
|
# Extra Java CLASSPATH elements. Optional. export HBASE_CLASSPATH=/data/hadoop/current/etc/hadoop |
6. 启动运行
进入HBASE_HOME/bin目录，执行start-hbase.sh即可启动HBase。请使用JDK提供的jps命令，分别查看HMaster和HRegionServer进程是否已经起来，同时检查日志文件是否有错误。
7. 基本的HBase命令
通过执行“hbase shell”进入命令行操作界面。详细请浏览官方文档：quickstart.html。
|
hbase(main):003:0> create 'test', 'cf' # 创建表test，一个列族cf 0 row(s) in 1.2200 seconds hbase(main):003:0> list 'test' .. 1 row(s) in 0.0550 seconds hbase(main):004:0> put 'test', 'row1', 'cf:a', 'value1' # 往表test的cf列族的a字段插入值value1 0 row(s) in 0.0560 seconds hbase(main):005:0> put 'test', 'row2', 'cf:b', 'value2' 0 row(s) in 0.0370 seconds hbase(main):006:0> put 'test', 'row3', 'cf:c', 'value3' 0 row(s) in 0.0450 seconds hbase(main):007:0> scan 'test' # 扫描表test ROW COLUMN+CELL row1 column=cf:a, timestamp=1288380727188, value=value1 row2 column=cf:b, timestamp=1288380738440, value=value2 row3 column=cf:c, timestamp=1288380747365, value=value3 3 row(s) in 0.0590 seconds hbase(main):008:0> get 'test', 'row1' # 从表test取一行数据 COLUMN CELL cf:a timestamp=1288380727188, value=value1 1 row(s) in 0.0400 seconds hbase(main):012:0> disable 'test' 0 row(s) in 1.0930 seconds hbase(main):013:0> drop 'test' 0 row(s) in 0.0770 seconds hbase(main):014:0> exit |
8. 备HMaster配置
备HMaster可以有0到多个，配置和主HMaster完全相同，所以只需要复制一份已配置好的HMaster过去即可，然后同样的命令启动。启动好后，一样可以执行HBase shell命令。
9. 常见错误
本文的实践过程中遇到了如下一些错误：
1) 错误1：Host key not found from database
遇到如下错误，说明不能免密码登录DEVNET-154-70、DEVNET-154-77和DEVNET-154-79，假设用户名为hadoop，则可以试试ssh hadoop@DEVNET-154-70来检查是否能免密码登录：
|
./start-hbase.sh DEVNET-154-70: Host key not found from database. DEVNET-154-70: Key fingerprint: DEVNET-154-70: xihad-rotuf-lykeh-mapup-kylin-kybub-sohid-bucaf-gafyg-vecuc-tyxux DEVNET-154-70: You can get a public key's fingerprint by running DEVNET-154-70: % ssh-keygen -F publickey.pub DEVNET-154-70: on the keyfile. DEVNET-154-70: warning: tcgetattr failed in ssh_rl_set_tty_modes_for_fd: fd 1: Invalid argument DEVNET-154-77: Host key not found from database. DEVNET-154-77: Key fingerprint: DEVNET-154-77: xuhog-tavip-donon-vuvac-tycyh-sysyz-zacur-didoz-fugif-vosar-ruxyx DEVNET-154-77: You can get a public key's fingerprint by running DEVNET-154-77: % ssh-keygen -F publickey.pub DEVNET-154-77: on the keyfile. DEVNET-154-77: warning: tcgetattr failed in ssh_rl_set_tty_modes_for_fd: fd 1: Invalid argument DEVNET-154-79: Host key not found from database. DEVNET-154-79: Key fingerprint: DEVNET-154-79: xolim-mysyg-bozes-zilyz-futaf-tatig-zaryn-pilaf-betyf-meduf-tixux DEVNET-154-79: You can get a public key's fingerprint by running DEVNET-154-79: % ssh-keygen -F publickey.pub DEVNET-154-79: on the keyfile. DEVNET-154-79: warning: tcgetattr failed in ssh_rl_set_tty_modes_for_fd: fd 1: Invalid argument |
2) 错误2：Failed deleting my ephemeral node
原因可能是因为之前配置错误，比如使用HBase自带的ZooKeeper启动过，后改为使用外围的ZooKeeper再启动。
|
2014-04-22 16:26:17,452 WARN [regionserver60020] zookeeper.RecoverableZooKeeper: Node /hbase/rs/DEVNET-154-79,60020,1398155173411 already deleted, retry=false 2014-04-22 16:26:17,453 WARN [regionserver60020] regionserver.HRegionServer: Failed deleting my ephemeral node org.apache.zookeeper.KeeperException$NoNodeException: KeeperErrorCode = NoNode for /hbase/rs/DEVNET-154-79,60020,1398155173411 at org.apache.zookeeper.KeeperException.create(KeeperException.java:111) at org.apache.zookeeper.KeeperException.create(KeeperException.java:51) at org.apache.zookeeper.ZooKeeper.delete(ZooKeeper.java:873) at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.delete(RecoverableZooKeeper.java:156) at org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode(ZKUtil.java:1273) at org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode(ZKUtil.java:1262) at org.apache.hadoop.hbase.regionserver.HRegionServer.deleteMyEphemeralNode(HRegionServer.java:1273) at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:1003) at java.lang.Thread.run(Thread.java:744) |
3) 错误3：Master rejected startup because clock is out of sync
来自RegionServer端的日志，HMaster拒绝RegionServer的连接。这个错误是因为HMaster上的时间和RegionServer上的时间相差超过30秒。两种解决办法：一是同步时间，二是修改hbase-site.xml中的hbase.master.maxclockskew（HMaster端的hdfs-site.xml文件）：。
|
2014-04-22 16:34:36,701 FATAL [regionserver60020] regionserver.HRegionServer: Master rejected startup because clock is out of sync org.apache.hadoop.hbase.ClockOutOfSyncException: org.apache.hadoop.hbase.ClockOutOfSyncException: Server DEVNET-154-79,60020,1398155672511 has been rejected; Reported time is too far out of sync with master. Time difference of 175968ms > max allowed of 30000ms at org.apache.hadoop.hbase.master.ServerManager.checkClockSkew(ServerManager.java:316) at org.apache.hadoop.hbase.master.ServerManager.regionServerStartup(ServerManager.java:216) at org.apache.hadoop.hbase.master.HMaster.regionServerStartup(HMaster.java:1281) at org.apache.hadoop.hbase.protobuf.generated.RegionServerStatusProtos$RegionServerStatusService$2.callBlockingMethod(RegionServerStatusProtos.java:5085) at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:2008) at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:92) at org.apache.hadoop.hbase.ipc.FifoRpcScheduler$1.run(FifoRpcScheduler.java:73) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:744) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:408) at org.apache.hadoop.ipc.RemoteException.instantiateException(RemoteException.java:106) at org.apache.hadoop.ipc.RemoteException.unwrapRemoteException(RemoteException.java:95) at org.apache.hadoop.hbase.protobuf.ProtobufUtil.getRemoteException(ProtobufUtil.java:284) at org.apache.hadoop.hbase.regionserver.HRegionServer.reportForDuty(HRegionServer.java:1998) at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:839) at java.lang.Thread.run(Thread.java:744) |
将hbase.master.maxclockskew改成可以容忍10分钟：
|
<property> <name>hbase.master.maxclockskew</name> <value>600000</value> <description>Time(ms) difference of regionserver from master</description> </property> |
10. 相关文档
《HBase-0.98.0分布式安装指南》
《Hive 0.12.0安装指南》
《ZooKeeper-3.4.6分布式安装指南》
《Hadoop 2.3.0源码反向工程》
《在Linux上编译Hadoop-2.4.0》
《Accumulo-1.5.1安装指南》
《Drill 1.0.0安装指南》
《Shark 0.9.1安装指南》
更多，敬请关注技术博客：http://aquester.cublog.cn。
附：phoenix的安装
Phoenix的安装非常简单。官网有说明（http://phoenix.incubator.apache.org/download.html），二进制安装包可从http://www.apache.org/dyn/closer.cgi/incubator/phoenix/上下载，本文下载的是phoenix-4.0.0-incubating.tar.gz，注意和HBase的兼容关系：
|
Phoenix版本 |
HBase版本 |
|
Phoenix 2.x |
HBase 0.94.x |
|
Phoenix 3.x |
HBase 0.94.x |
|
Phoenix 4.x |
HBase 0.98.1+ |
安装步骤为：
1) 将phoenix-4.0.0-incubating.tar.gz上传到Phoenix客户端机器，假设安装到/data/hadoop
2) 解压phoenix-4.0.0-incubating.tar.gz，解压后生成phoenix-4.0.0-incubating目录
3) 建立软链接：ln -s phoenix-4.0.0-incubating phoenix
4) 将/data/hadoop/phoenix/hadoop-2/phoenix-4.0.0-incubating-client.jar添加到CLASSPATH
5) 将phoenix/common目录下的phoenix-core-4.0.0-incubating.jar复制到所有HBase region server的的CLASSPATH中，比如HBase的lib目录
6) 重启HBase集群
运行phoenix也非常简单，命令格式为：
|
sqlline.py zookeeper file.sql |
示例：
|
hadoop@VM-40-171-sles10-64:~/phoenix/bin> ./sqlline.py 10.12.154.78 Setting property: [isolation, TRANSACTION_READ_COMMITTED] issuing: !connect jdbc:phoenix:10.12.154.78 none none org.apache.phoenix.jdbc.PhoenixDriver Connecting to jdbc:phoenix:10.12.154.78 Connected to: Phoenix (version 4.0) Driver: org.apache.phoenix.jdbc.PhoenixDriver (version 4.0) Autocommit status: true Transaction isolation: TRANSACTION_READ_COMMITTED Building list of tables and columns for tab-completion (set fastconnect to true to skip)... 53/53 (100%) Done Done sqlline version 1.1.2 0: jdbc:phoenix:10.12.154.78> select * from test; Error: ERROR 1012 (42M03): Table undefined. tableName=TEST (state=42M03,code=1012) 0: jdbc:phoenix:10.12.154.78> create table test ( a int, b string); Error: ERROR 601 (42P00): Syntax error. Unsupported sql type: INT (state=42P00,code=601) 0: jdbc:phoenix:10.12.154.78> create table test (a integer, b integer); Error: ERROR 509 (42888): The table does not have a primary key. tableName=TEST (state=42888,code=509) 0: jdbc:phoenix:10.12.154.78> create table test (a integer primary key, b integer) ; No rows affected (1.424 seconds) 0: jdbc:phoenix:10.12.154.78> UPSERT INTO TEST VALUES (1, 1); 1 row affected (0.099 seconds) 0: jdbc:phoenix:10.12.154.78> UPSERT INTO TEST VALUES (2, 12); 1 row affected (0.02 seconds) 0: jdbc:phoenix:10.12.154.78> select * from test; +------------+------------+ | A | B | +------------+------------+ | 1 | 1 | | 2 | 12 | +------------+------------+ 2 rows selected (0.116 seconds) 0: jdbc:phoenix:10.12.154.78> |
有关语法请浏览：http://phoenix.incubator.apache.org/language/index.html，有关数据类型请浏览：http://phoenix.incubator.apache.org/language/datatypes.html。