Tag Info

New answers tagged

0

I think I found the problem. In the process of loading and initially configuring Debian Wheezy on to the LEGO EV3, I had to establish an ethernet link over USB. That necessitated altering the IPv4 config which I did in Debian 7.5 desktop by System Settings > Network> choose the wired connection to the LEGO brick (the other wired connection is the ethernet ...


1

I'm going to chime in because the given answer only covers the use case where you want your VM to also be exposed to the internet, and is also more work than necessary. If you want to get this working on a virtual internal network: Set up the virtual switch as Internal network and apply it to the VM In the Hyper-V manager, under the networking tab, you ...


0

This started for us after we installed syslog-ng on a RHEL 6.5 box for some improved functionality. To fix the issue we disabled rsyslogd with chkconfig. If you are already logged into the server, it should be an easy fix. We are performing root cause on this now. bash$ chkconfig rsyslog off bash$ chkconfig |grep rsyslog ~~~ rsyslog 0:off 1:off ...


0

Use PuTTYgen. Start PuTTYgen. Load your private key in .ppk format. The go to Menu > Conversions > Export > OpenSSH. This creates a key in .pem format. Though I'm not sure what is this good for. WinSCP uses .ppk format.


0

The server is deciding to drop the connection, so you'd have to debug the problem from the server side. If you have "root" access on the server, you can run sshd interactively: /path/to/sshd -ddd -p 42 This will launch a copy of sshd listening on port 42--you could specify a different number--in debug mode. It will run in the foreground, accepting a ...


0

We got it working. There was an issue with the way it was being configured on the remote host, specific to the depressingly old hardware we're working with.


0

Try using either gcutil pull: gcutil [--global_flags] pull [--command_flags] <instance-name> <file-1> ... <file-n> <destination> e.g., $ gcutil pull --zone=us-central1-a \ my_instance \ /remote/file1 \ /remote/file2 \ ~/local/path or gcloud compute copy-files: gcloud compute copy-files ...


0

The problem lies in your user name. If you login locally Windows is not case sensitive. But if you login "remotely" you have to take care of it. You must write the user name exactly how it was defined in Windows.


0

The color settings are located in ~/.mc/ini. Add a color settings, for example [Colors] ...


0

Well, you can start another ssh session right from your .ssh/authorized_keys file: command="/usr/bin/ssh -i ~/.ssh/id_ecdsa_<user> <server>",no-port-forwarding,no-agent-forwarding,no-X11-forwarding,no-pty ecdsa-sha2-nistp256 <key-data> You will still need to create a local key file and add it to the target user/server ...


0

You probably have to change or need to set the IdentityFile in your ssh config. Edit or Insert the line, into the File /etc/ssh/ssh_conf: IdentityFile ~/.ssh/myserver Normaly you only have one private key. If you have multiple private key's it's probably the best way to create "config" file in your ".ssh directory" host myserver Hostname myserver user ...


0

Possible solution, add something like this to your .ssh/config file: host myserver hostname myserver.com identityfile ~/.ssh/myserver user myuser and you'll be fine with 'ssh myserver'.


1

On your client machine, create ~/.ssh/config host box1 hostname 55.55.55.55 user bob host box2 hostname 55.55.55.56 user bob Then "ssh box1", "ssh box2" do the expected things. You can also set other options like port 2222, ForwardX11 yes, ForwardAgent yes etc.


2

The solution, as the Server Fault post describes, is to mount your EBS to another (new) instance which you can connect to. The EBS will be just a drive there, and you will be able to remote connect to the instance since it is new. You can then sudo chown the directories you broke in the first place. Stop (NOT terminate) your first instance from AWS console ...


1

If all variables (apart from $REMOTE_HOST) are available in the environment on your remote server you can try: ssh $REMOTE_HOST 'echo sed "s|export ${KEY}=${VAL}|export ${KEY}=${NEW_VAL}|" "$ENV_FILE"' And if everything looks fine, remove echo. Update With local variables: ssh $REMOTE_HOST "echo sed \"s|export ${KEY}=${VAL}|export ${KEY}=${NEW_VAL}|\" ...


1

You can't recover your private key. If there is another account that allows ssh access to that VM, and that account has sudo privileges, you can ssh with that account and sudo into your account's files. From there you can change the key to a pair you already have. Alternatively, if your information is stored on a Cinder volume, you can shut down the machine ...


0

Chech if you did install openssh-server server . sudo apt-get install openssh-server then check you have proper port enabled /etc/ssh/sshd_config if you go in it with different port number from outside or your router has a rule of forwarding it to another port you may also need to uncomment in there if you want to be authorized by password. ...


0

I’ve been working on Geofront project. I think it could help you. Geofront is a simple SSH key management server. It helps to maintain servers to SSH, and authorized_keys list for them. Read the docs for more details. Situations If the team maintains authorized_keys list of all servers owned by the team: When someone joins ...


0

Go in to network settings in Virutalbox and change the network adapter to either bridged or host only. Make sure the VM's IP address is on the same subnet as the host's.


0

I could not create a passwordless user, but I created one with username and password programname and in the /etc/ssh/sshd_config file I added Match User testuserand ForceCommand python /path/file.py That way anyone can log in (the login info is in the web site at that server) see the program and then log out automatically when the program closes.


0

You can allow passwordless logins through SSH authorization keys running ssh-keygen, which generated a public and private key. The user needs the private key to login and server needs the public key in ~/.ssh/authorized_keys You can get more info on How to set up ssh so you aren't asked for a password About the restriction, you can create a chrooted ...


1

There are 2 ways to setup sshd: GUI and (the real way) command line. Assuming 10.6 or later... GUI: System Preferences > Sharing > Check "Remote Login" Make sure you set "Allow Access for" to whomever you want to restrict this to Command Line /usr/bin/systemsetup -setremotelogin on Modify /etc/sshd_config to the settings optimal for your environment. ...


0

the file should be 600, the directory 700. chmod go-w ~ chmod 700 ~/.ssh chmod 600 ~/.ssh/authorized_keys That the permissions formula I always use. I'm running debian, but should work the same for ubuntu


3

Most likely you have a stateful firewall/nat between your system and the remote system that is timing out the state data for your connection after 15/20 minutes. Modify the keepalive setting in your client. A value of 120 seconds is what I usually use. Modifying the firewall/nat device causing the problem would also be possible, but may have other side ...


1

not really. One can just download the authorized_keys file, edit it removing the command="internal-sftp" part and upload it on top of yours and then ssh into the box.


1

In Putty Settings > Window > Behavior, you can check one of the boxes to open the system menu on a certain keypress (I personally use ALT-Space). With this setting in place, you can hit ALT-Space, then type the d key to Duplicate Session. This will allow you to open a new putty window without needing to touch the mouse. It's not a command line tool, but I ...


0

Easy way: go to "network settings", "port forwarding" and add a rule to map VM port 22 to, for instance, port 9022 on the host. Then run ssh localhost -p 9022. Better way: install Ubuntu lxc package, then, on VirtualBox, go to the VM Network settings, set Attached to to Bridged Adapter, set Name to lxcbr0, reboot the virtual machine and it would get a new ...


0

I find it best not to store identity files (keys) in the ~/.ssh directory, since it is known to the SSH client, which (as you have noticed) has an annoying tendency to try all identities it finds in this directory, even when you explicitly specify a sole identity file for it to use. I store all of my identity files in another directory (~/.ssh2) that is not ...


2

When you run putty or ssh--I don't think there's any difference in this context--with a command to run on the remote system, the remote ssh server runs the command as a shell command: /bin/bash -c '~/scripts/test; $SHELL -l' So you have a bash instance on the remote system executing this pipeline. You also have another bash instance, launched by the first ...


0

ctrl+c for the putty remote command is actually interrupting the ssh session, so this should be expected that it doesn't execute the remaining remote commands. If your aim is just to have the session continue to remain active, your infinite loop only should be enough.


1

No you can't. But you can use screen for that purpose, it creates a virtual terminal where you can run the process you want, then detach from it and reattach whenever you want. Create a new screen: screen -S [session_name] Detach from that session <Ctrl> + <A> then <D> Reattach to a session: screen -r [session_name]


0

I added to ~/.ssh/config this: Host * IdentitiesOnly yes It enables option IdentitiesOnly=yes by default. If you'll need to connect with private key, you should specify it with option -i


4

$OPENBLAS_NUM_THREADS is being evaluated locally by bash before it is sent to your remote machine. You need to either escape the $: ssh worker-2 "export OPENBLAS_NUM_THREADS=1; echo \$OPENBLAS_NUM_THREADS" or use single quotes, which inhibit the bash variable evaluation: ssh worker-2 'export OPENBLAS_NUM_THREADS=1; echo $OPENBLAS_NUM_THREADS'


0

You can use ANT as build task runner with a FTP task. With the property depends setted to yes. By guide: transfers only new or changed files if set to "yes". Defaults to "no".


0

Create a key pair with ssh-keygen.To set up trust between the two users on the two machines, you need add the contents of ~/.ssh/id_rsa.pub on your machine to ~/.ssh/authorized_keys on the remotemachine and chmod it user only. You should be able to login without password now. Then to use sudo remotely you need to configure something else in the sudoers file ...


0

Try apt-get install openssh-server openssh-client. I had the same issue when I updated my system to Debian Testing. Even though SSH was already installed, in my case it wasn't pulled in for updating when I ran apt-get update && apt-get upgrade and so the system still had the same SSH binaries with the now-outdated libraries linked against it, hence ...


1

Add this to the top of your file so the shell knows which interpreter to use. #! /bin/sh Also make sure your file is executable: chmod +x yourScript There is also no need for sudo with ssh. Only use ssh ssh user@machine If you need to pass the Password I recommend sshpass apt-get install sshpass sshpass -p<password> ssh user@machine


2

I forgot how many levels of auth can throw MOTDs. It was PAM's fault. pam_motd, to be exact, specified in /etc/pam.d/sshd


0

I think that a very stripped down version of busybox (just a basic shell and a few network configuration tools you may need for setting up the network) with the addition of dropbear as ssh client (statically linked) will be way under the 11mb you mention. I'd say around 1-2mb would suffice for sure, but you could trim it down even more. Plus add a minimal ...


0

Make sure you remove host key from username/.ssh/known_hosts prior to connecting. I had a problem using sshfs and sysrescuecd (every boot gets new keys etc).


0

It appears changing the font to indicate bold text make it look fuzzy/blurry, from memory and testing on multiple PC's. For now I changed it to indicate bold te


2

Try leaving off the "root@" portion of the -R argument: ssh -v -R hostname:19999:localhost:22 root@hostname With the four-parameter version of -R, the four parameters are the address and port to listen for connections on the remote host, and the address and port to forward to from the local host. You're giving a value of "root@hostname" as the remote ...


0

twm autoplacement can be set in your ~/.twmrc file with "RandomPlacement". e.g., $ cat ~/.twmrc RandomPlacement $ twm is rough though, there are a lot of things you cannot configure. "man twm" has a lot of good information.


1

The easiest way to do this is that you create an additional Profile for using Firefox with the socks proxy. To create an additional Firefox profile. Close all Instances of Firefox Start firefox with the -p flag This will show up the Userprofile dialog Here you see your default Firefox profile named default. Click on Create new Profile and give it a name ...


0

Turned out that the problem had to do with the Connection->Proxy settings were set for the Default Settings and my particular saved session. Once I removed them from the Default it stopped spawning multiple PLINK processes.


1

Use the sshpass utility. sudo apt-get install sshpass Then send commands via: sshpass -p '<password>' <ssh/scp command>


0

If have two options: just use the command line you already typed, you will be asked for the password of user "root" setup password-less authentication, like this: http://wiki.centos.org/HowTos/Network/SecuringSSH#head-9c5717fe7f9bb26332c9d67571200f8c1e4324bc , your command will be executed using the SSH key and you won't be asked to insert the password


0

Use sshpass: ssh uses direct TTY access to make sure that the password is indeed issued by an interactive keyboard user. sshpass runs ssh in a dedicated tty, fooling it into thinking it is getting the password from an interactive user. sshpass -p 'password' scp -r user@example.com:/some/remote/path /some/local/path You can also use sshpass with ssh, rsync ...



Top 50 recent answers are included