hi,all:
on my datanode ,hava 12 disk ,but the consume was not blances ：
node1:
[hadoop@master2 logs]$ df -h | grep hdfs
/dev/sdc1 3.6T 3.0T 466G 87% /hdfs/data3
/dev/sdb1 3.6T 3.0T 464G 87% /hdfs/data2
/dev/sdl1 3.6T 3.0T 464G 87% /hdfs/data12
/dev/sdd1 3.6T 2.4T 1.1T 69% /hdfs/data4
/dev/sdg1 3.6T 3.0T 412G 89% /hdfs/data7
/dev/sdj1 3.6T 3.0T 467G 87% /hdfs/data10
/dev/sdf1 3.6T 2.1T 1.4T 61% /hdfs/data6
/dev/sdk1 3.6T 3.0T 412G 89% /hdfs/data11
/dev/sde1 3.6T 399G 3.1T 12% /hdfs/data5
/dev/sdi1 3.6T 3.0T 412G 89% /hdfs/data9
/dev/sdh1 3.6T 3.0T 464G 87% /hdfs/data8
/dev/sda1 3.6T 3.0T 460G 87% /hdfs/data1
node2:
[hadoop@slave1 ~]$ df -h | grep hdfs
/dev/sdd1 3.6T 3.0T 454G 87% /hdfs/data4
/dev/sdj1 3.6T 2.8T 665G 81% /hdfs/data10
/dev/sdf1 3.6T 2.3T 1.2T 67% /hdfs/data6
/dev/sda1 3.6T 3.0T 458G 87% /hdfs/data1
/dev/sdl1 3.6T 3.0T 433G 88% /hdfs/data12
/dev/sdh1 3.6T 1.9T 1.6T 55% /hdfs/data8
/dev/sdb1 3.6T 3.0T 454G 87% /hdfs/data2
/dev/sdk1 3.6T 3.0T 455G 87% /hdfs/data11
/dev/sdg1 3.6T 1.7T 1.8T 49% /hdfs/data7
/dev/sde1 3.6T 2.6T 876G 75% /hdfs/data5
/dev/sdi1 3.6T 2.6T 849G 76% /hdfs/data9
/dev/sdc1 3.6T 3.0T 455G 87% /hdfs/data3
and maybe some disk Use% about 90% , when I run hdfs dfs -ls ， was very slow.
1、 how can I blances the usage of disks
2、 what to do to find any clue about why the hdfs become slow
2017-02-14
lk_hadoop
