hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Keren Ouaknine <ker...@gmail.com>
Subject Hadoop 0.23.1 - cluster startup and job test
Date Sat, 12 May 2012 23:23:52 GMT
Hello,

I configured 0.23 thanks to cloudavenue's
<http://www.thecloudavenue.com/2012/01/getting-started-with-nextgen-mapreduce.html>post.
My UI seems ok, but reports only one node out of the ten.

My master is running:
18130 DataNode
18264 ResourceManager
18507 NodeManager
19881 NameNode

And my slaves are running:
DataNode
*Since start-all.sh is deprecated, I had to launch it manually, using a for
loop!*
*Q1: Do I have to write my own script to start the entire cluster??!*

Q2: I think I am missing a tasktracker (or its equivalent in 0.23)?
Is that the reason why the other nodes dont figure in the UI?

Q3: I tried, to submit a simple PI calculation job, and got:
> ./bin/hadoop jar
share/hadoop/mapreduce/hadoop-mapreduce-examples-0.23.1.jar pi 10 10000

12/05/12 19:21:40 INFO mapreduce.Job:  map 0% reduce 0%
*12/05/12 19:21:40 INFO mapreduce.Job: Job job_1336859612524_0005 failed
with state FAILED due to: Application application_1336859612524_0005 failed
1 times due to AM Container for appattempt_1336859612524_0005_000001 exited
with  exitCode: 1 due to: *
*.Failing this attempt.. Failing the application.*
12/05/12 19:21:40 INFO mapreduce.Job: Counters: 0
Job Finished in 2.847 seconds

Thanks,
Keren

Mime
View raw message