SSHing into EC2 server via gives error Please login as the ec2-user user rather than root user

You log in as ec2-user as Klaus suggested:

ssh -i key.pem ec2-user@host

... and then you use sudo to run commands. E.g., to edit the /etc/hosts file which is owned by root and requires root privileges: sudo nano /etc/hosts.

Or you run sudo su to become the root user.

By default root user is not allowed to login but you can use ec2-user as indicated by others.

Once you login with ec2-user you switch to root and change the SSH configuration.

To become the root user you run:

sudo su -

Edit the SSH daemon configuration file /etc/ssh/sshd_config, e.g. by using vi, and replace the PermitRootLogin entry with the following:

PermitRootLogin without-password

Reload the SSH daemon configuration by running:

/etc/init.d/sshd reload

The message Please login as the ec2-user user rather than root user. is displayed because a command is executed when you login with the private key. To remove that command edit ~/.ssh/authorized_keys file and remove the command option. The line should start with the key type (Eg. ssh-rsa).

(*) Do at your own risk. I recommend you to leave always a console open just in case you're not able to login after you make the configuration changes.

For reference you can read the man pages:

man sshd_config
man sshd

I have encountered a similar problem when setting up a hadoop cluster on Amazon ec2.

My head node needs to have root ssh access to each worker/slave nodes. I aliased the connects by adding each slave node's IP address, private address, and alias name to the /etc/hosts/ file. (I get that data by running the command echo -e "`hostname -i`\t`hostname -f`\talias-name" where alias-name is what I call each node (head or n1 for example). Then I put that output for each node in every node's /etc/hosts file.

The problem I have been encountering is that when I type ssh n1 while in my head node to ssh into my first slave node, I get that same error message: Please login as the use "ec2-user" rather than the user "root". So after doing some research, I figured out how to fix it.

First:

• ssh into your server. non-root (ec2-user) access is fine here.
• Then su - your way into root. Now vi /etc/ssh/sshd_config and un-comment the line PermitRootLogin yes.
• Exit vi editor.
• Now restart ssh daemon by typing service sshd stop then service sshd start.

Second:

• Now, here is the part I had to dig for,
• run vi /root/.ssh/authorized_keys
• Comment out everything up to ssh-rsa. Just put a # at the beginning of the file's content, before no-port-forwarding... and hit enter on ssh-rsa to move it to the next line (this way you dont have to delete anything in case you want to backtrack).
• exit vi editor

Now you should be able to login to root without that error message popping up.

Also, if you are using aliases for a cluster setup; Repeat the same steps on each node. First ssh in using ec2-user then follow the steps. After adding the IP address, private address, and alias name info to your /etc/hosts file you should be able to ssh into each node's root using the alias name for example ssh n1.

The tutorial I followed is here: https://www.youtube.com/watch?v=xrxQXfE7t9A

But it didnt discuss the problem with root login.

Hope that helps! It worked for me.

*Keep in mind that I havnt taken any security into concern. This is simply a practice/dev setup.