hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adam Phelps <...@opendns.com>
Subject Starting hadoop services at boot
Date Fri, 12 Nov 2010 23:49:09 GMT
I'm sure there is some detail I'm missing, but I've been testing node 
failures and noticed that the various services (datanode, tasktracker, 
regionserver in my case) don't automatically restart on boot (I'm 
running on Ubuntu on EC2).  Its currently hiding from me, however I 
assume there is a configuration setting somewhere that enables this?

I noticed that there various start scripts in /etc/init.d:

root@ip-10-123-1-151:~/control# ls /etc/init.d/hadoop*
/etc/init.d/hadoop-0.20-datanode    /etc/init.d/hadoop-0.20-namenode 
        /etc/init.d/hadoop-0.20-tasktracker 
/etc/init.d/hadoop-hbase-regionserver
/etc/init.d/hadoop-0.20-jobtracker 
/etc/init.d/hadoop-0.20-secondarynamenode 
/etc/init.d/hadoop-hbase-master      /etc/init.d/hadoop-hbase-thrift

However these don't actually appear to work in this setup:

root@ip-10-123-1-151:~/control# /etc/init.d/hadoop-0.20-namenode stop
Stopping Hadoop namenode daemon: no namenode to stop
ERROR
root@ip-10-123-1-151:~/control# /etc/init.d/hadoop-0.20-namenode start
Starting Hadoop namenode daemon: starting namenode, logging to 
/usr/lib/hadoop-0.20/bin/../logs/hadoop-root-namenode-ip-10-123-1-151.out
/usr/lib/hadoop-0.20/bin/hadoop-daemon.sh: line 114: 
/var/run/hadoop/hadoop-root-namenode.pid: Permission denied
nice: cannot set niceness: Permission denied
ERROR.

I'm currently starting services using "hadoop-daemon.sh start XXX".

Thanks
- Adam

Mime
View raw message