hdfs datanode

A file may be splitted to many chunks and replications stored on many datanodes in HDFS. Now, the question is how to find the DataNodes that actually store a file in HDFS?

一个文件可能被分割成许多块，并且复制存储在HDFS的许多数据节点上。 现在，问题是如何找到实际在HDFS中存储文件的DataNode？

You may use the dfsadmin -fsck tool from the hdfs util. Here is an example:

您可以使用 hdfs util中的dfsadmin -fsck工具。 这是一个例子：

$ hadoop fsck /user/aaa/file.name - -locations -blocksConnecting to namenode via http://dstore-170:50070FSCK started by hadoop (auth:SIMPLE) from /10.0.3.170 for path /user/path/to/file.gz at Fri Oct 17 12:25:55 HKT 2014/user/path/to/file.gz 12448905476 bytes, 93 block(s):  OK0. BP-1960069741-10.0.3.170-1410430543652:blk_1074365040_625145 len=134217728 repl=2 [10.0.3.173:50010, 10.0.3.174:50010]1. BP-1960069741-10.0.3.170-1410430543652:blk_1074365041_625146 len=134217728 repl=2 [10.0.3.175:50010, 10.0.3.174:50010]2. BP-1960069741-10.0.3.170-1410430543652:blk_1074365042_625147 len=134217728 repl=2 [10.0.3.175:50010, 10.0.3.174:50010]3. BP-1960069741-10.0.3.170-1410430543652:blk_1074365043_625148 len=134217728 repl=2 [10.0.3.175:50010, 10.0.3.174:50010]4. BP-1960069741-10.0.3.170-1410430543652:blk_1074365044_625149 len=134217728 repl=2 [10.0.3.181:50010, 10.0.3.174:50010]...91. BP-1960069741-10.0.3.170-1410430543652:blk_1074365131_625236 len=134217728 repl=2 [10.0.3.175:50010, 10.0.3.174:50010]92. BP-1960069741-10.0.3.170-1410430543652:blk_1074365132_625237 len=100874500 repl=2 [10.0.3.181:50010, 10.0.3.174:50010]Status: HEALTHY Total size:	12448905476 B Total dirs:	0 Total files:	1 Total symlinks:		0 Total blocks (validated):	93 (avg. block size 133859198 B) Minimally replicated blocks:	93 (100.0 %) Over-replicated blocks:	0 (0.0 %) Under-replicated blocks:	0 (0.0 %) Mis-replicated blocks:		0 (0.0 %)  replication factor:	2 Average block replication:	2.0 Corrupt blocks:		0 Missing replicas:		0 (0.0 %) Number of data-nodes:		10 Number of racks:		1FSCK ended at Fri Oct 17 12:25:55 HKT 2014 in 1 millisecondsThe filesystem under path '/user/aaa/file.name' is HEALTHY

翻译自:

hdfs datanode