"rack awareness" is a bit of a misnomer. Hadoop is not aware of racks in the sense that it can figure out your network topology. Instead, you need to write a script that tells Hadoop which machines belong to which racks. For example: http://wiki.apache.org/hadoop/topology_rack_awareness_scripts. You can tell if your script is working when "default-rack" is replaced by one of your rack names.
Jeff From: Jain, Prem [mailto:premanshu.j...@netapp.com] Sent: Wednesday, February 01, 2012 1:55 PM To: hdfs-user@hadoop.apache.org Subject: Testing rack awareness Is there a way to check if the rack awareness is working ? I seem to configure it right, but when I run a fsck on my files written, it still shows "Number of racks: 1" What I am looking for is a simple command which can tell me if rack awareness is working or not ? Thanks in advance ------------------------------------------------------------------------------------------------------------------ root@namenode]# hadoop fsck /user/file -files -blocks -locations -racks FSCK started by root from /10.61.97.224 for path /user/file at Wed Feb 01 14:36:42 EST 2012 /user/file bytes, 1 block(s): OK 0. blk_-2693119624313554881_407726 len=18483674 repl=2 [/default-rack/10.61.97.214:50010, /default-rack/10.61.97.217:50010] Status: HEALTHY Total size: 18483674 B Total dirs: 0 Total files: 1 Total blocks (validated): 1 (avg. block size 18483674 B) Minimally replicated blocks: 1 (100.0 %) Over-replicated blocks: 0 (0.0 %) Under-replicated blocks: 0 (0.0 %) Mis-replicated blocks: 0 (0.0 %) Default replication factor: 2 Average block replication: 2.0 Corrupt blocks: 0 Missing replicas: 0 (0.0 %) Number of data-nodes: 8 Number of racks: 1 FSCK ended at Wed Feb 01 14:36:42 EST 2012 in 0 milliseconds The filesystem under path '/user/file ' is HEALTHY
