use git in obsidian
This commit is contained in:
194
Hadoop/HDFS搭建.md
Normal file
194
Hadoop/HDFS搭建.md
Normal file
@@ -0,0 +1,194 @@
|
||||
#### 相关命令
|
||||
- hadoop-daemon.sh start journalnode 或 hdfs --daemon start journalnode
|
||||
- hadoop-daemon.sh start namenode 或 hdfs --daemon start namenode
|
||||
- hdfs namenode -format 第一个namenode格式化
|
||||
- hdfs namenode -bootstrapStandby 第二、三个namenode格式化
|
||||
- hdfs zkfc -formatZK 格式化zkfc ,仅在nn1执行
|
||||
- hdfs --daemon start zkfc 在 nn1、nn2、nn3执行
|
||||
>`core-site.xml`
|
||||
|
||||
```xml
|
||||
<configuration>
|
||||
<property>
|
||||
<name>fs.defaultFS</name>
|
||||
<value>hdfs://ns1</value>
|
||||
<description>默认文件服务的协议和NS逻辑名称，和hdfs-site.xml里的对应此配置替代了1.0里的fs.default.name</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>hadoop.tmp.dir</name>
|
||||
<value>/data/tmp</value>
|
||||
<description>数据存储目录</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>hadoop.proxyuser.root.groups</name>
|
||||
<value>hadoop</value>
|
||||
<description>
|
||||
hdfs dfsadmin –refreshSuperUserGroupsConfiguration,
|
||||
yarn rmadmin –refreshSuperUserGroupsConfiguration
|
||||
使用这两个命令不用重启就能刷新
|
||||
</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>hadoop.proxyuser.root.hosts</name>
|
||||
<value>localhost</value>
|
||||
<description>本地代理</description>
|
||||
</property>
|
||||
|
||||
<!-- zkfc的配置 -->
|
||||
<property>
|
||||
<name>ha.zookeeper.quorum</name>
|
||||
<value>nn1:2181,nn2:2181,nn3:2181</value>
|
||||
<description>HA使用的zookeeper地址</description>
|
||||
</property>
|
||||
</configuration>
|
||||
```
|
||||
>`hdfs-site.xml`
|
||||
|
||||
```xml
|
||||
<configuration>
|
||||
<property>
|
||||
<name>dfs.namenode.name.dir</name>
|
||||
<value>/data/namenode</value>
|
||||
<description>namenode本地文件存放地址</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.nameservices</name>
|
||||
<value>ns1</value>
|
||||
<description>提供服务的NS逻辑名称，与core-site.xml里的对应</description>
|
||||
</property>
|
||||
|
||||
<!-- namenode的配置 -->
|
||||
<!-- 主要的 -->
|
||||
<property>
|
||||
<name>dfs.ha.namenodes.ns1</name>
|
||||
<value>nn1,nn2,nn3</value>
|
||||
<description>列出该逻辑名称下的NameNode逻辑名称</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.rpc-address.ns1.nn1</name>
|
||||
<value>nn1:9000</value>
|
||||
<description>指定NameNode的RPC位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.http-address.ns1.nn1</name>
|
||||
<value>nn1:50070</value>
|
||||
<description>指定NameNode的Web Server位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.rpc-address.ns1.nn2</name>
|
||||
<value>nn2:9000</value>
|
||||
<description>指定NameNode的RPC位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.http-address.ns1.nn2</name>
|
||||
<value>nn2:50070</value>
|
||||
<description>指定NameNode的Web Server位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.rpc-address.ns1.nn3</name>
|
||||
<value>nn3:9000</value>
|
||||
<description>指定NameNode的RPC位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.http-address.ns1.nn3</name>
|
||||
<value>nn3:50070</value>
|
||||
<description>指定NameNode的Web Server位置</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.namenode.handler.count</name>
|
||||
<value>77</value>
|
||||
<description>namenode的工作线程数</description>
|
||||
</property>
|
||||
|
||||
<!-- journaldata配置,使得其他两个namenode同步第一个namenode数据 -->
|
||||
<property>
|
||||
<name>dfs.namenode.shared.edits.dir</name>
|
||||
<value>qjournal://nn1:8485;nn2:8485;nn3:8485/ns1</value>
|
||||
<description>指定用于HA存放edits的共享存储，通常是namenode的所在机器</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.journalnode.edits.dir</name>
|
||||
<value>/data/journaldata/</value>
|
||||
<description>journaldata服务存放文件的地址</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>ipc.client.connect.max.retries</name>
|
||||
<value>10</value>
|
||||
<description>namenode和journalnode的链接重试次数10次</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>ipc.client.connect.retry.interval</name>
|
||||
<value>10000</value>
|
||||
<description>重试的间隔时间10s</description>
|
||||
</property>
|
||||
|
||||
<!-- zkfc的配置 -->
|
||||
<property>
|
||||
<name>dfs.ha.fencing.methods</name>
|
||||
<value>sshfence</value>
|
||||
<description>指定HA做隔离的方法，缺省是ssh，可设为shell，稍后详述</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.ha.fencing.ssh.private-key-files</name>
|
||||
<value>/home/hadoop/.ssh/id_rsa</value>
|
||||
<description>杀死命令脚本的免密配置秘钥</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.client.failover.proxy.provider.ns1</name>
|
||||
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
|
||||
<description>指定客户端用于HA切换的代理类，不同的NS可以用不同的代理类以上示例为Hadoop 2.0自带的缺省代理类</description>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.client.failover.proxy.provider.auto-ha</name>
|
||||
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
|
||||
</property>
|
||||
|
||||
<property>
|
||||
<name>dfs.ha.automatic-failover.enabled</name>
|
||||
<value>true</value>
|
||||
</property>
|
||||
<!-- datanode配置 -->
|
||||
<property>
|
||||
<name>dfs.datanode.data.dir</name>
|
||||
<value>/data/datanode</value>
|
||||
<description>datanode本地文件存放地址</description>
|
||||
</property>
|
||||
<property>
|
||||
<name>dfs.replication</name>
|
||||
<value>3</value>
|
||||
<description>文件复本数</description>
|
||||
</property>
|
||||
<property>
|
||||
<name>dfs.namenode.datanode.registration.ip-hostname-check</name>
|
||||
<value>false</value>
|
||||
</property>
|
||||
<property>
|
||||
<name>dfs.client.use.datanode.hostname</name>
|
||||
<value>true</value>
|
||||
</property>
|
||||
<property>
|
||||
<name>dfs.datanode.use.datanode.hostname</name>
|
||||
<value>true</value>
|
||||
</property>
|
||||
</configuration>
|
||||
|
||||
```
|
||||
|
||||
|
||||
32
Hadoop/Zookeeper.md
Normal file
32
Hadoop/Zookeeper.md
Normal file
@@ -0,0 +1,32 @@
|
||||
> **选举机制**
|
||||
ZooKeeper 不是一个传统意义上的文件系统。它是一个分布式协调服务，主要用于分布式系统中解决数据一致性、分布式锁管理、集群配置管理等问题。
|
||||
- 过半选举
|
||||
|
||||
>linux的前台和后台任务
|
||||
- `&` 后台运行
|
||||
```bash
|
||||
./f.sh &
|
||||
```
|
||||
`nohup` 将结果存放到 日志文件,而不是在控制台输出
|
||||
```bash
|
||||
nohup ./f.sh &
|
||||
```
|
||||
`2&>1` 将stdin、stdout、stderror输出到 `/tmp/log.txt`
|
||||
```bash
|
||||
nohup ./f.sh >> /tmp/log.txt 2>&1
|
||||
```
|
||||
`/dev/null` 将结果忽略
|
||||
```bash
|
||||
nohup ./f.sh >> /dev/null 2>&1
|
||||
```
|
||||
程序运行的`stdin、stdout、stderror`查看
|
||||
先使用`jps`查看zookeeper的进程ID
|
||||
然后`cd /proc/进程ID/fd` 即可
|
||||
|
||||
>相关命令
|
||||
|
||||
- `ssh_all_zookeeper.sh /usr/local/zookeeper/bin/zkServer.sh start`
|
||||
- `ssh_all_zookeeper.sh /usr/local/zookeeper/bin/zkServer.sh status`
|
||||
- `ssh_all_zookeeper.sh /usr/local/zookeeper/bin/zkServer.sh stop`
|
||||
- `zkCli.sh start`
|
||||
|
||||
27
Hadoop/脚本.md
Normal file
27
Hadoop/脚本.md
Normal file
@@ -0,0 +1,27 @@
|
||||
### 集群启动脚本
|
||||
> 也可以直接使用 `start-dfs.sh` ，需要先启动`zookeeper`
|
||||
|
||||
```bash
|
||||
ssh_all_zookeeper.sh /usr/local/zookeeper/bin/zkServer.sh start
|
||||
ssh_all_zookeeper.sh hdfs --daemon start journalnode
|
||||
ssh_all_zookeeper.sh hdfs --daemon start namenode
|
||||
ssh_all_zookeeper.sh hdfs --daemon start zkfc
|
||||
ssh hadoop@s1 hdfs --daemon start datanode
|
||||
ssh hadoop@s2 hdfs --daemon start datanode
|
||||
ssh hadoop@s3 hdfs --daemon start datanode
|
||||
```
|
||||
|
||||
### 集群停止脚本
|
||||
> 也可以直接使用 `stop-dfs.sh`
|
||||
```bash
|
||||
ssh_all_zookeeper.sh hdfs --daemon stop namenode
|
||||
ssh hadoop@s1 hdfs --daemon stop datanode
|
||||
ssh hadoop@s2 hdfs --daemon stop datanode
|
||||
ssh hadoop@s3 hdfs --daemon stop datanode
|
||||
ssh_all_zookeeper.sh hdfs --daemon stop journalnode
|
||||
ssh_all_zookeeper.sh hdfs --daemon stop zkfc
|
||||
ssh_all_zookeeper.sh /usr/local/zookeeper/bin/zkServer.sh start
|
||||
```
|
||||
集群搭建注意事项
|
||||
- `/data`这个目录六台机器均需要创建，且将所有者、所属组改为`hadoop`
|
||||
-
|
||||
Reference in New Issue
Block a user