1. HDFS常用操作
1.1. 查询
1.1.1. 浏览器查询
1.1.2. 命令行查询
[yun@mini04 bin]$ hadoop fs -ls /
1.2. 上传文件
[yun@mini05 zhangliang]$ cat test.info [yun@mini05 zhangliang]$ hadoop fs -put test.info / # 上传文件
[yun@mini05 software]$ ll -h
total 641M
-rw-r--r-- yun yun 190M Jun : CentOS-.4_hadoop-2.7..tar.gz
[yun@mini05 software]$ hadoop fs -put CentOS-.4_hadoop-2.7..tar.gz / # 上传文件
[yun@mini05 software]$ hadoop fs -ls /
Found items
-rw-r--r-- yun supergroup -- : /CentOS-.4_hadoop-2.7..tar.gz
-rw-r--r-- yun supergroup -- : /test.info
1.2.1. 文件存放位置
[yun@mini04 subdir0]$ pwd # 根据之前的配置，总共3份相同的文件
/app/hadoop/tmp/dfs/data/current/BP--10.0.0.11-/current/finalized/subdir0/subdir0
[yun@mini04 subdir0]$ ll -h # 默认128M 切片
total 192M
-rw-rw-r-- yun yun Jun : blk_1073741825
-rw-rw-r-- yun yun Jun : blk_1073741825_1001.meta
-rw-rw-r-- yun yun 128M Jun : blk_1073741826
-rw-rw-r-- yun yun 1.1M Jun : blk_1073741826_1002.meta
-rw-rw-r-- yun yun 62M Jun : blk_1073741827
-rw-rw-r-- yun yun 493K Jun : blk_1073741827_1003.meta
[yun@mini04 subdir0]$ cat blk_1073741825 [yun@mini04 subdir0]$
1.2.2. 浏览器访问
1.3. 文件下载
[yun@mini04 zhangliang]$ pwd
/app/software/zhangliang
[yun@mini04 zhangliang]$ ll
total
[yun@mini04 zhangliang]$ hadoop fs -ls /
Found items
-rw-r--r-- yun supergroup -- : /CentOS-.4_hadoop-2.7..tar.gz
-rw-r--r-- yun supergroup -- : /test.info
[yun@mini04 zhangliang]$ hadoop fs -get /test.info
[yun@mini04 zhangliang]$ hadoop fs -get /CentOS-.4_hadoop-2.7..tar.gz
[yun@mini04 zhangliang]$ ll
total
-rw-r--r-- yun yun Jun : CentOS-.4_hadoop-2.7..tar.gz
-rw-r--r-- yun yun Jun : test.info
[yun@mini04 zhangliang]$ cat test.info [yun@mini04 zhangliang]$
2. 简单案例
2.1. 准备数据
[yun@mini05 zhangliang]$ pwd
/app/software/zhangliang
[yun@mini05 zhangliang]$ ll
total
-rw-rw-r-- yun yun Jun : test.info
-rw-rw-r-- yun yun Jun : zhang.info
[yun@mini05 zhangliang]$ cat test.info [yun@mini05 zhangliang]$ cat zhang.info
zxcvbnm
asdfghjkl
qwertyuiop
qwertyuiop qwertyuiop
[yun@mini05 zhangliang]$ hadoop fs -mkdir -p /wordcount/input
[yun@mini05 zhangliang]$ hadoop fs -put test.info zhang.info /wordcount/input
[yun@mini05 zhangliang]$ hadoop fs -ls /wordcount/input
Found items
-rw-r--r-- yun supergroup -- : /wordcount/input/test.info
-rw-r--r-- yun supergroup -- : /wordcount/input/zhang.info
2.1. 运行分析
[yun@mini04 mapreduce]$ pwd
/app/hadoop/share/hadoop/mapreduce
[yun@mini04 mapreduce]$ ll
total
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-app-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-common-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-core-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-hs-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-hs-plugins-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-jobclient-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-jobclient-2.7.-tests.jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-client-shuffle-2.7..jar
-rw-r--r-- yun yun Jun : hadoop-mapreduce-examples-2.7..jar
drwxr-xr-x yun yun Jun : lib
drwxr-xr-x yun yun Jun : lib-examples
drwxr-xr-x yun yun Jun : sources
[yun@mini04 mapreduce]$ hadoop jar hadoop-mapreduce-examples-2.7..jar wordcount /wordcount/input /wordcount/output
// :: INFO client.RMProxy: Connecting to ResourceManager at mini02/10.0.0.12:
// :: INFO input.FileInputFormat: Total input paths to process :
// :: INFO mapreduce.JobSubmitter: number of splits:
// :: INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1528589937101_0002
// :: INFO impl.YarnClientImpl: Submitted application application_1528589937101_0002
// :: INFO mapreduce.Job: The url to track the job: http://mini02:8088/proxy/application_1528589937101_0002/
// :: INFO mapreduce.Job: Running job: job_1528589937101_0002
// :: INFO mapreduce.Job: Job job_1528589937101_0002 running in uber mode : false
// :: INFO mapreduce.Job: map % reduce %
// :: INFO mapreduce.Job: map % reduce %
// :: INFO mapreduce.Job: map % reduce %
// :: INFO mapreduce.Job: Job job_1528589937101_0002 completed successfully
// :: INFO mapreduce.Job: Counters:
File System Counters
FILE: Number of bytes read=
FILE: Number of bytes written=
FILE: Number of read operations=
FILE: Number of large read operations=
FILE: Number of write operations=
HDFS: Number of bytes read=
HDFS: Number of bytes written=
HDFS: Number of read operations=
HDFS: Number of large read operations=
HDFS: Number of write operations=
Job Counters
Launched map tasks=
Launched reduce tasks=
Data-local map tasks=
Total time spent by all maps in occupied slots (ms)=
Total time spent by all reduces in occupied slots (ms)=
Total time spent by all map tasks (ms)=
Total time spent by all reduce tasks (ms)=
Total vcore-milliseconds taken by all map tasks=
Total vcore-milliseconds taken by all reduce tasks=
Total megabyte-milliseconds taken by all map tasks=
Total megabyte-milliseconds taken by all reduce tasks=
Map-Reduce Framework
Map input records=
Map output records=
Map output bytes=
Map output materialized bytes=
Input split bytes=
Combine input records=
Combine output records=
Reduce input groups=
Reduce shuffle bytes=
Reduce input records=
Reduce output records=
Spilled Records=
Shuffled Maps =
Failed Shuffles=
Merged Map outputs=
GC time elapsed (ms)=
CPU time spent (ms)=
Physical memory (bytes) snapshot=
Virtual memory (bytes) snapshot=
Total committed heap usage (bytes)=
Shuffle Errors
BAD_ID=
CONNECTION=
IO_ERROR=
WRONG_LENGTH=
WRONG_MAP=
WRONG_REDUCE=
File Input Format Counters
Bytes Read=
File Output Format Counters
Bytes Written=
[yun@mini05 zhangliang]$ hadoop fs -ls /wordcount/output
Found items
-rw-r--r-- yun supergroup -- : /wordcount/output/_SUCCESS
-rw-r--r-- yun supergroup -- : /wordcount/output/part-r-
[yun@mini05 zhangliang]$ hadoop fs -cat /wordcount/output/part-r- asdfghjkl
qwertyuiop
zxcvbnm
3. 案例：开发shell采集脚本
3.1. 需求说明
点击流日志每天都10T，在业务应用服务器上，需要准实时上传至数据仓库（Hadoop HDFS）上
3.2. 需求分析
一般上传文件都是在凌晨24点操作，由于很多种类的业务数据都要在晚上进行传输，为了减轻服务器的压力，避开高峰期。
如果需要伪实时的上传，则采用定时上传的方式。比如每小时上传一次。
3.3. web日志模拟
运行jar包模拟web日志
# 必要的目录
# 模拟web服务器目录 /app/webservice
# 模拟web日志目录 /app/webservice/logs
# 模拟待上传文件存放的目录 /app/webservice/logs/up2hdfs # 在这个目录上传到HDFS
[yun@mini01 webservice]$ pwd
/app/webservice
[yun@mini01 webservice]$ ll
total
drwxrwxr-x yun yun Jun : logs
-rw-r--r-- yun yun Jun : testlog.jar
# 运行jar包，打印日志
[yun@mini01 webservice]$ java -jar testlog.jar &
……………………
# 生成web日志查看
[yun@mini01 logs]$ pwd
/app/webservice/logs
[yun@mini01 logs]$ ll -hrt
total 460K
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun 11K Jun : access.log.
-rw-rw-r-- yun yun .6K Jun : access.log
3.4. Shell脚本执行
# 脚本测试
[yun@mini01 hadoop]$ pwd
/app/yunwei/hadoop
[yun@mini01 hadoop]$ ll
-rwxrwxr-x yun yun Jun : uploadFile2Hdfs.sh
[yun@mini01 hadoop]$ ./uploadFile2Hdfs.sh
………………
3.5. 查看结果
3.5.1. 脚本执行结果查看
Web日志转移与上传查看
# web日志转移查看
[yun@mini01 logs]$ pwd
/app/webservice/logs
[yun@mini01 logs]$ ll -hrt
total 28K
-rw-rw-r-- yun yun .3K Jun : access.log ### 表明日志已经转移
drwxrwxr-x yun yun 16K Jun : up2hdfs # 待上传文件存放的目录
[yun@mini01 up2hdfs]$ pwd
/app/webservice/logs/up2hdfs
[yun@mini01 up2hdfs]$ ll -hrt
-rw-rw-r-- yun yun 11K Jun : access.log.7_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.6_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.5_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.4_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.3_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.2_20180614174001_DONE
-rw-rw-r-- yun yun 11K Jun : access.log.1_20180614174001_DONE
## 文件以 DONE 结尾，表明已上传到HDFS
脚本日志查看
# shell脚本日志查看
[yun@mini01 log]$ pwd
/app/yunwei/hadoop/log
[yun@mini01 log]$ ll
total
-rw-rw-r-- yun yun Jun : uploadFile2Hdfs.sh.log
[yun@mini01 log]$ cat uploadFile2Hdfs.sh.log
-- :: uploadFile2Hdfs.sh access.log.1_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.2_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.3_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.4_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.5_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.6_20180614174001 ok
-- :: uploadFile2Hdfs.sh access.log.7_20180614174001 ok
3.5.2. HDFS命令行上传文件查看
[yun@mini01 ~]$ hadoop fs -ls /data/webservice/
Found items
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.1_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.2_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.3_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.4_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.5_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.6_20180614174001
-rw-r--r-- yun supergroup -- : /data/webservice//access.log.7_20180614174001
3.5.3. 浏览器查看看
http://10.0.0.11:50070/explorer.html#/data/webservice/20180614 路径 ： /data/webservice/20180614
3.6. 脚本加入定时任务
[yun@mini01 hadoop]$ crontab -l # WEB日志上传HDFS 每小时执行一次
*/ * * * /app/yunwei/hadoop/uploadFile2Hdfs.sh >/dev/null >&
3.7. 相关的脚本和jar包
在Git上，路径如下：
https://github.com/zhanglianghhh/bigdata/tree/master/hadoop/hdfs