it looks like hadoop on ubuntu did not get a home directory: /home/hadoop
You can log into instances using the following ssh commands: [hadoop-datanode+hadoop-tasktracker]: ssh -i /home/ab/whirr/ab-ubuntu-12-64.pem -o "UserKnownHostsFile /dev/null" -o StrictHostKeyChecking=no [email protected] [hadoop-namenode+hadoop-jobtracker]: ssh -i /home/ab/whirr/ab-ubuntu-12-64.pem -o "UserKnownHostsFile /dev/null" -o StrictHostKeyChecking=no [email protected] To destroy cluster, run 'whirr destroy-cluster' with the same options used to launch it. ab@ubuntu12-64:~$ ssh -i /home/ab/whirr/ab-ubuntu-12-64.pem -o "UserKnownHostsFile /dev/null" -o StrictHostKeyChecking=no [email protected] Warning: Permanently added '54.225.34.149' (ECDSA) to the list of known hosts. Welcome to Ubuntu 12.04.2 LTS (GNU/Linux 3.2.0-48-virtual x86_64) * Documentation: https://help.ubuntu.com/ System information as of Wed Aug 7 17:55:20 UTC 2013 System load: 0.27 Processes: 58 Usage of /: 18.0% of 7.87GB Users logged in: 0 Memory usage: 27% IP address for eth0: 10.164.72.201 Swap usage: 0% Graph this data and manage this system at https://landscape.canonical.com/ Get cloud support with Ubuntu Advantage Cloud Guest: http://www.ubuntu.com/business/services/cloud Use Juju to deploy your cloud instances and workloads: https://juju.ubuntu.com/#cloud-precise 42 packages can be updated. 24 updates are security updates. Last login: Wed Aug 7 17:54:04 2013 from 50.0.160.200 ab@ip-10-164-72-201:~$ java -version java version "1.7.0_25" Java(TM) SE Runtime Environment (build 1.7.0_25-b15) Java HotSpot(TM) 64-Bit Server VM (build 23.25-b01, mixed mode) ab@ip-10-164-72-201:~$ ps -ef | grep hadoop ab 2312 2252 0 17:55 pts/0 00:00:00 grep hadoop ab@ip-10-164-72-201:~$ sudo su - hadoop No directory, logging in with HOME=/ $ which start-all.sh /usr/local/hadoop-1.2.1/bin/start-all.sh $ start-all.sh Warning: $HADOOP_HOME is deprecated. starting namenode, logging to /var/log/hadoop/logs/hadoop-hadoop-namenode-ip-10-164-72-201.out Error: JAVA_HOME is not set. localhost: Could not create directory '/home/hadoop/.ssh'. localhost: Failed to add the host to the list of known hosts (/home/hadoop/.ssh/known_hosts). localhost: Permission denied (publickey). localhost: Could not create directory '/home/hadoop/.ssh'. localhost: Failed to add the host to the list of known hosts (/home/hadoop/.ssh/known_hosts). localhost: Permission denied (publickey). starting jobtracker, logging to /var/log/hadoop/logs/hadoop-hadoop-jobtracker-ip-10-164-72-201.out Error: JAVA_HOME is not set. localhost: Could not create directory '/home/hadoop/.ssh'. localhost: Failed to add the host to the list of known hosts (/home/hadoop/.ssh/known_hosts). localhost: Permission denied (publickey). $ i create a home directory to see if that was enough - it wasn't: $ exit ab@ip-10-164-72-201:~$ sudo mkdir /home/hadoop ab@ip-10-164-72-201:~$ chown -R hadoop:hadoop /home/hadoop chown: changing ownership of `/home/hadoop': Operation not permitted ab@ip-10-164-72-201:~$ sudo chown -R hadoop:hadoop /home/hadoop ab@ip-10-164-72-201:~$ sudo su - hadoop $ start-all.sh Warning: $HADOOP_HOME is deprecated. starting namenode, logging to /var/log/hadoop/logs/hadoop-hadoop-namenode-ip-10-164-72-201.out Error: JAVA_HOME is not set. localhost: Warning: Permanently added 'localhost' (ECDSA) to the list of known hosts. localhost: Permission denied (publickey). localhost: Permission denied (publickey). starting jobtracker, logging to /var/log/hadoop/logs/hadoop-hadoop-jobtracker-ip-10-164-72-201.out Error: JAVA_HOME is not set. localhost: Permission denied (publickey). $ ps -ef |grep hadoop root 2582 2252 0 18:03 pts/0 00:00:00 sudo su - hadoop hadoop 2583 2582 0 18:03 pts/0 00:00:00 su - hadoop hadoop 2584 2583 0 18:03 pts/0 00:00:00 -su hadoop 2857 2584 0 18:04 pts/0 00:00:00 ps -ef hadoop 2858 2584 0 18:04 pts/0 00:00:00 grep hadoop it looks like i need the ssh keys - i don't know which ones - but, anyway, i'm wasting time. any idea why it might had skipped this provisioning step? any idea what i should try next? ________________________________ From: Andrew Bayer <[email protected]> To: [email protected]; a b <[email protected]> Sent: Wednesday, August 7, 2013 10:39 AM Subject: Re: how to login as hadoop after cluster is launched Login as ubuntu, sudo su - hadoop. A. On Wed, Aug 7, 2013 at 10:30 AM, a b <[email protected]> wrote: i ran "launch-cluster" on ec2, but hadoop did not start. i would like to login as hadoop and run start-all.sh and see what error i get. > >i tried ssh -i bootstrap-private-key.pem hadoop@ec2-public-ip >and that failed > >i tried ssh -i bootstrap-private-key.pem ubuntu@ec2-public-ip >which worked then i tried to switch users to hadoop: >su - hadoop >but it asked me for a password i do not have. > >i could use a suggestion. > >
