We have a setup whereby we cannot login to ServerB directly by using putty.
We have to first login to ServerA by providing username and password and it somehow takes us to serverB and prompts for the password for my login id for ServerB. Providing the password again logs me onto ServerB where I have a script test.sh which I need to execute.
I wanted to automate the execution of script test.sh on ServerB from my local windows machine hence I wrote the below command.
C:\Program Files\PuTTY>plink -agent user1#ServerA "chmod +x /export/home
/user1/test.sh; sh /export/home/user1/test.sh" -v
This command logs me onto ServerA but does not take me to ServerB hence test.sh script is not found.
I cannot logon to serverB due to this hopping concept in place.
Kindly help me with some work around.
Note: ServerA and ServerB both are unix servers
Try using the -nc parameter something likeL
plink -agent user1#ServerA -nc user#serverB "chmod +x /export/home /user1/test.sh; sh /export/home/user1/test.sh" -v
Related
I'm trying to create a function that connects me to a ssh server then become su, and then ssh into another server so I did the following:
function test
ssh -t testuser#server1 'sudo ssh -t testuser#server2'
end
When I execute it I get the following error ssh: command not found
But when I execute it straight to the terminal it works with no problems.
This sounds like a path issue on server1.
From the command line, what do you see if you type the following?
ssh -t testuser#server1 sudo which ssh
If SSH is not in the path for root, you might need to specify full paths, such as something like:
ssh -t testuser#server1 sudo /usr/bin/ssh testuser#server2
You may need to adjust the paths to match your environment, of course.
Also, if you're trying to connect from root at server1 to server2, can you just ssh directly to root#server1? If so, you could perhaps use the "ProxyJump" functionality that was added with OpenSSH 7.3. This depends upon the ability to remotely login as root, which may not be an option, depending on your environment.
ssh -J root#server1 testuser#server2
My problem was solved when I added each one of them to the fish functions folder:
~/.config/fish/functions
I just created a file called myfunction.fish and inside of that file I pasted the function definition:
function myfunction
ssh -t testuser#server1 'sudo ssh -t testuser#server2'
end
saved it, exited fish and now that function is permanent.
I have ServerA which serves a central backup server for all linux machines in the field. They sync with rsync. I need keyless entry for these machines - say ServerX, ServerY and ServerZ. The idea is to give each client server a separate username and home folder to backup to - thus isolating each server's data and risk.
ServerA has UserX, UserY, UserZ.
On ServerX:
[root#ServerX ~]ssh-copy-id -i ~/.ssh/id_rsa.pub root#ServerA
[root#ServerX ~]ssh root#ServerA
[root#ServerA ~]
I can login as root without a password prompt
Now if I try using one of the other users on ServerA
[root#ServerX ~]ssh-copy-id -i ~/.ssh/id_rsa.pub UserX#ServerA
[root#ServerX ~]ssh UserX*#ServerA
UserX#ServerA's password:
[UserX#ServerA ~]
The ssh-copy-id does not report any erros and appears to work- but yet I cannot log in to ServerA "key-less". I have tried a couple combinations of the commands, but I cannot get passwordless login as UserX into ServerA
I am sure I am missing something obviouse here. :) Any feedback or advice would be appreciate to get passwordless access for UserX,UserY and UserZ.
Thank you for your help and time,
Regards,
Rudolf
From my laptop, I often ssh into another machine in my university department. I have to put in a password every time currently.
Could someone give me an idiot's guide to having the password be automatically entered each time I log in please.
Thank you in advance.
You can override by enabling Password less authentication. But you should install keys (pub, priv) before going for that.
Execute the following commands at local server.
Local $> ssh-keygen -t rsa
Press ENTER for all options prompetd. No values need to be typed.
Local $> cd .ssh
Local $> scp .ssh/id_rsa.pub user#targetmachine:
Prompts for pwd$> ENTERPASSWORD
Connect to remote server using the following command
Local $> ssh user#targetmachine
Prompts for pwd$> ENTERPASSWORD
Execute the following commands at remote server
Remote $> mkdir .ssh
Remote $> chmod 700 .ssh
Remote $> cat id_rsa.pub >> .ssh/authorized_keys
Remote $> chmod 600 .ssh/authorized_keys
Remote $> exit
Execute the following command at local server to test password-less authentication.
It should be connected without password.
$> ssh user#targetmachine
I assume you are using Linux. Lot of places in the internet where it is already documented.
For example(s):
http://www.rebol.com/docs/ssh-auto-login.html
http://www.linuxproblem.org/art_9.html
You can log in without providing password if PKI (public key infrastructure) is set up.
Otherwise you'll have to look for ssh client that can store passwords and supports your operating system.
Use a tool (such as AutoHotkey, assuming you are using Windows) to record and replay key sequences: http://www.autohotkey.com/
I am trying to connect to a remote host from my local host through the below command.But there was a setting in the remote host that soon after we login it will prompt to enter a badge ID,password and reason for logging in, because it was coded like that in profile file on remote-host How can I overcome those steps and login directly non-interactively, without disturbing the code in profile.
jsmith#local-host$ ssh -t -t generic_userID#remote-host
Enter your badgeID, < exit > to abort:
Enter your password for <badgeID> :
Enter a one line justification for your interactive login to generic_userID
Small amendment: to overcome remote server expect approach is required, but in case local script connects to bunch of remote servers, which configuration may be broken, just use SSH options:
ssh -f -q -o BatchMode=yes -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null USER#TARGETSYSTEM
This will omit ask for password in case there is no ssh_key setup, exit silently and continue with script/other hosts.
Puts ssh to background with -f, which is required when calling ssh command from sh (batch) file to remove local console redirect to remote input (implies -n).
Look into setting up a wrapper script around expect. This should do exactly what you're looking for.
Here are a few examples you can work from.
I have upvoted Marvin Pinto's answer because there is every reason to script this, in case there are other features in the profile that you need, such as Message of the Day motd.
However, there is a quick and dirty alternative if you don't want to make a script and you don't want other features from the profile. Depending on your preferred shell on the remote host, you can insist that the shell bypasses the profile files. For example, if bash is available on the remote host, you can invoke it with:
ssh -t -t generic_userID#remote-host bash --noprofile
I tested the above on the macOS 10.13 version of OpenSSH. Normally the command at the end of the ssh invocation is run non-interactively, but the -t flag allows bash to start an interactive shell.
Details are in the Start-up files section of the Bash Reference Manual.
I'm using cygrunsrv and autossh on windows XP to create a service building a tunnel to a remote server but i also want to create another tunnel from the remote server to another server.
I can achieve it with this command line :
autossh -M 5432 serverA -t 'autossh -M 4321 serverB -N'
but when I want to set it up in cygwin through cygrunsrv to make it works as a service :
cygrunsrv -I TUNNEL -p /usr/bin/autossh -a "-M 5432 serverA -t 'autossh -M 4321 serverB -N'" -e AUTOSSH_NTSERVICE=yes -e AUTOSSH_POLL=20 -e AUTOSSH_GATETIME=30
It's not fully working. The service is creating the tunnel correctly to ServerA but it's not sending the autossh command "autossh -M 4321 serverB -N" to ServerA.
I tried to escape the quote but all my efforts didn't make any difference and I'm not seeing any command sent in the autossh logs.
I think the problem is related to pseudo terminal that is not created through the cygrunsrv.
I'd like to know if there's a way to fix my cygrunsrv command line to make it work or should I consider a different approach ?
Lionel, try removing the AUTOSSH_NTSERVICE=yes from the cygrunsrv invocation. As /usr/share/doc/autossh/README.Cygwin explains:
Setting AUTOSSH_NTSERVICE=yes in the calling environment ...
change[s] autossh's behavior in three useful
ways:
(1) Add an -N flag to each invocation of ssh, thus disabling shell
access. The idea is that if you're running autossh as a system
service, you're using it to forward ports; it wouldn't make sense to
run a shell session as a system service. (If you think this reasoning
is wrong, please send a bug report to the author or Cygwin maintainer,
and tell us what you're trying to do.)
Despite what the above says, it seems that you may have a good reason for not wanting -N (which suppresses command execution) in your service's ssh invocation. Removing AUTOSSH_NTSERVICE=yes should take care of it. It will have a couple of other minor disadvantages, but you can probably live with it. Read the rest of README.Cygwin for the details.