I can ssh with password ok but after doing the whole key generation procedure at [URL].. section 7, it still prompts for a password. Here is the output:
[root@apvdbs03 log]# ssh -v root@aapxen01
OpenSSH_4.3p2, OpenSSL 0.9.8e-fips-rhel5 01 Jul 2008
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: Applying options for *
debug1: Connecting to aapxen01 [10.0.1.20] port 22.
debug1: Connection established.
[Code]..
I have 3 servers with Oracle 10g installed. In order for rsync to work for log shipping I need to have passwordless SSH configured.*Note: I did not configure these servers, 2 DBA's got canned and I'm left to fix this and I am not a DBA.I have done passwordless SSH in the past and it is easy.Here is my problem: (-> = passwordless ssh works)ServerA -> ServerCServerB -> ServerCBut I need all 3 to be able to do passwordless SSH to each other.I deleted all the files in /home/oracle/.ssh on all 3 servers and this is what I did for all 3 servers:
ssh-keygen -t rsa
ssh-keygen -t dsa
cat id_rsa.pub >> authorized_keys
[code]....
There are a few threads about this but none seem to me.I have 3 machines setup with a nfs home directory on machine 1.I have created keys using ssh-keygen -t rsa (and left everything blank), then as user on machine 1 run
Code:
cat id_rsa.pub >> authorized_keys2
chmod 644 ~/.ssh/*
this used to work until I upgraded to ubuntu 11.04 (from 9.04), and now it requires a password to ssh. I have tried several times deleting ~/.ssh/* and recreating, to no luck Below is the results when I try to ssh into -2 from -1:
[Code]...
(dumb question, but shouldn't the offered RSA public key be id_rsa.pub rather than id_rsa?)
we are running a small cluster (CentOS 5.5) for research in fluid dynamics, using OpenFOAM and Fluent togehter with Sun Grid Engine. For these programs, ssh without password prompt in console is required. So far, I did the method described here: [URL]... with keys distributed at each node. However, imagine a new users needs to be created:
1. Creating user at every node.
2. Create key at every node
3. Distribute key to every node
Makes password prompt threetimes, for 50 nodes - quite a lot. The students dont like that really much, me neither. What I am asking for is any help to make this more comfortable. I am not really a system admin, but more a researcher with some knowledge in admin tasks. Actually, I would like to use a ldap authetication, were passwords are stored centrally. When the user first logs in to a machine, the home directory should be created. I think I can manage it to work.
[Code]...
I have an rsync server and am now setting up cwrsync on my windows machine. I want to be able to run cwrsync over ssh with public private keys. I followed a tutorial over here to set this up. It, however, still prompts me for server password and works only if I provide the password. For some reason the public/private key process is not working.
- I generated a key using the command: ssh-keygen -t rsa -N '' (I verified the key gets generated on my Windows machine)
- I uploaded the generated file id_rsa.pub to server /root/.ssh/authorized_keys
I am also prompted for a password if from command prompt I run this command to log into server: ssh -i c:docume~1user.sshid_rsa root@<server_ip_address> On server I have changed the configuration file (/etc/ssh/ssh_config) to say:
RSAAuthentication yes
PubkeyAuthentication yes
I then restarted the sshd service, however, to no avail.
How can I configure ssh to do the passwordless logins between the users in a Linux server?I am using RHEL4.6 version. I have created some users in this server and I would like to login from one user to the other users in the same server without a password prompt by using SSH. I was using rsh for this purpose before, but since it is against our security policy , I have to switch to SSH.I have done the ssh configuration by creating the keys and copying the public keys to the file .ssh/authorized_keys of the other users in the same server.But unfortunately still I am not able to login to other users in the same system WITHOUT a Password.I did the following settings.From user1:
mkdir ~/.ssh
chmod 755 ~/.ssh
/usr/bin/ssh-keygen -t rsa
[code]...
I'm in my terminal and do git pull, git push alot in a work day. On the server I'm pulling and pushing I can't do password less ssh the usual way by generating rsa keys.Is it still possible perhaps for me to enter my password once for a session and do all the git pulls/pushes I want?
View 1 Replies View RelatedI am trying to run OpenNebula on a set of 2 machines. One of the machines(called host1) must be able to access the others(including itself) with passwordless ssh.Both machines have the same user and use mounted folder from host1 for home directory(exported with NFS). So single key generation should be enough for this configuration - all the machines have access to the same .ssh folder.What I am getting at the end is passwordless ssh from host1 to host2 which is great, but no success from host1 to host1. Host2 to host2 is also OK, host2 to host1 wants password like host1 to host1.Both machines have the same sshd_config and ssh_config files.File permissions required by ssh should be correct since it works for host1 to host2.The only difference(I see) between the machines is that host1 runs nfs service, so host2 has access to the exported folders.
The ssh-keygen is executed on host1authorized_keys has the same public key two times(to authorize access for host1 on host1 and host2).The distro is Scientific Linux 6.0 aand I also use rsa.Something else - no pass ssh with another user from host1 to host1 works, so I suppose the problem comes from this mounted folder.
I have key-based authentication working for one user on a cluster--
Code:
user@thisnode > ssh user@othernode
last login blah blah blah..
user@othernode >
but if I switch to that user and try it, I get an error--
[Code]...
I'm working on some code from my computer at home and would like to do CVS updates to a remote server. When I'm at work I don't need to type in a password, but at home each time I need to type in my password each time I do a CVS update. I've followed the instructions here on how to set up a passwordless public/private ssh key pair: [URL].. However, after that I still need to type in my ssh password each time I do CVS update (or anything CVS related) from my local computer.
View 1 Replies View RelatedI followed the steps here: [URL]
It works fine, except every time I try to do something that requires authentication of the key, I have to type in the password in a prompt that looks like this:
Code:
$ git pull
Enter passphrase for key '/home/<user>/.ssh/id_rsa':
remote: Counting objects: 16, done.
remote: Compressing objects: 100% (9/9), done.
remote: Total 9 (delta 7), reused 0 (delta 0)
...
This is very annoying. How can I have it so I don't have to enter my password each time?
On my local machine I followed the same instructions and don't have to enter the password every time...
I have for a while been trying to setup ssh so hadoop can ssh without a password to all it's nodes. (Currenly 2 nodes)But i can not get the ssh to login without either asking for a password or just getting access denied (public key)I have been following two guidesFirst:Setting up SingleNode ClusterThen after:Setting up Multinode ClusterI have also done various ssh tutorials on setting up ssh.I got to the point where i do the command bin/start-dfs.shThen it begins asking for passwords. Which it should'nt.The problem is that ssh is not setup correcly.And whatever i try i get problems with logging in.the computers are called.
hadoop1.utoft.be (Namenode and JobTracker)
hadoop2.utoft.be
hadoop3.utoft.be
[code]....
I am running Ubuntu 10.04 (OpenSSH_5.3p1 Debian-3ubuntu3, OpenSSL 0.9.8k 25 Mar 2009) I am trying to setup passwordless ssh/scp for some servers on a LAN I am running the following commands:
Code:
ssh-keygen
ssh-copy-id user@<ipaddress>
ssh-add
I added the last step as the following error prevented logins:
Code:
agent admitted to failure to sign using the key After running ssh-add, i still get the error(agent admitted to failure to sign using the key) but passwordless ssh/scp works. The following is the output of user@localmachine:~$ ssh -vvv remote@10.0.7.112
[Code]...
I want use two computer to run OpenFOAM or Fluent software.These two computers are connected via a cross cable. I have access to another by "ssh maysam@192.168.1.1" OR "ssh 192.168.1.1" The problem is i should have access to another PC by "ssh maysam-desktop" command which can run parallel. maysam-desktop is name of anothe PC and maysam is name of its user.
View 1 Replies View RelatedI have two load balancing boxes... One primary (box1) and one backup/failover (box2). I want to make sure the heartbeat is running on both boxes via a script. While I'm on box1, I want to run as any user: ssh box2 service pulse status and not have box2 ask for a password.
View 1 Replies View RelatedWhat's the easiest way to have a system script (running as root) execute remote commands over ssh? I've written some scripts that execute commands remotely via ssh, and they work great when I run them as myself, as I've set up ssh-agent and keys for passwordless login. I'd like to call these when my laptop docks and undocks. I've been successful at running arbitrary scripts when docking/undocking, but since the ACPI event scripts run as root, trying to run my ssh script fails during authentication.
I tried using sudo with the -u and -i flags to simulate running the script as my user, e.g.: sudo -u redmoskito -i /home/redmoskito/bin/remote_command which successfully finds my private key and tries to use it, but the ssh-agent credentials are still missing, so it still needs my passphrase.
I am trying to set up passworldess login - I am trying to automate ssh to use send commands from a Red Hat el5.2 server to a solaris 9 server (I did get it working from solaris 10-9).I have ran ssh-keygen -t rsa and then the ssh-copy-id and i can login passwordless via ssh. N HURRAH
Unfortunately when I use my scripts to automate the commands I ALWAYS receive an "Error Code - 255" - the command is successfully sent.
I have to establish password SFTP between one Unix box and Windows in which certain files will be pushed from unix to windows server. Our client is insisting on using TurboFTP on Windows for ssh daemon to run. There are two components in TurboFTO- client and server.
1. I need help if I need to install both. As far as my knowledge with this software is considered, I dont think TurboFTP client to be running as the server componoent will itself run sshd.
2. Also, when I try to add server in TurboFTP server console, it provides me two options - local and remote server types. Local uses 3830 port by default and uses ip 127.0.0.1. If I use Remote server type which uses 3839 and I dont know what ip address to provide. I need help whether I shud go ahead with local server or remote server.
My doubt is if I consider local server type then whether my Unix box will be able to recognize if thie windows machine (which is also naving network specific ip address) has sshd running because windows has local server type enabled which uses 127.0.0.1.
I have been trying to establish a connection between two pcs via the ssh channel. I successfully made the connection as a root user, but when i tried as a non root user i had to type in the password the make contact with the other pc. how exactly does this passwordless login actually work?
View 3 Replies View RelatedI can't get this to work on my machines.
So far I have:
1. created a key with ssh-keygen on the server to be logged in to
2. copied the .pub key to my local machine
3. chmod 700 ~/.ssh on both machines
4. chomd 600 ~/.ssh/ic_rsa on the server, and on known_hosts on my local machine
5. added the .pub key to ~/known_hosts on my local machine
my local machine doesn't have an "authorized_keys" file which is what everything is telling me I should append my .pub key to. The only thing that was in my .ssh folder was known_hosts, so I tried that. I also tried making an authorized_hosts file to no avail, changing permissions appropriatly on all files.
Should I/Can I reset ssh in some way? Is there are reason I don't have an authorized_keys file or is my known_hosts file my authorized_keys file?
Would it be better just to uninstall/reinstall ssh?
I was asked to list passwordless or locked accounts in a fedora system. As far as I know, fedora or RH dont allow passwordless accounts (the system locks the account Edit: ... unless you set passwd -d :-P ), so I might show info about locked accounts.
I am totally aware that I could do a grep or an awk searching for "!!" or "", but my people is asking me to check against passwd -S (which prints info of accounts). The thing, is that in Solaris you can check with passwd -a -s all the users, but in Linux I couldnt find anything similar. what approach do you reccomend for passing, say , the list of users ( like the output of awk searching only user names) to the command passwd -S one by one?
I'm trying to configure a process triggered by an SVN post-commit hook which will log into a different host and carry out an SVN update on a file path on that host before exiting. An earlier attempt mounted the remote filepath on the SVN host using sshfs and performed the update locally. This worked but it was incredibly slow (minutes to complete an SVN update).
So, Plan B was to set-up a passwordless login for the user the script runs as and then use pam-script to script a checkout from a repository using the same credentials. The problem is, passwordless SSH login using private/public keys appears to bypass the PAM authentication system or at least interact with it in a way that no environment variables (including the SSH user's name and pass) are resolved by the authentication script being used by pam-script.
I've tested the pam-script behaviour for normal log-ins and it exposes these variables fine. This leaves me in a Catch-22 with trying to script access on one host to perform actions on another while avoiding user/pass prompts or the need to store plaintext passwords on the remote host.
Anyone know if there's a way to resolve a user account password via PAM when using passwordless SSH or, another approach I could take to perform scripted tasks on the remote system requiring authentication? Ideally without storing the passwords on the remote system (at least in unencrypted form).
I've installed a new fresh Centos 5.5 ..and after a lot of work..something starts to work (X now works with the new ATI HD5145 using ATI drivers) but i've a serious problemThe ethernet controller JMC250 JMicron PCI EXPR. doesent work at all so my notebook is out of the world !!!Here is some infos:rpm -q centos-release
centos-release-5-5.el5.centos
rpm -qa kernel* | sort
kernel-devel-2.6.18-194.el5
[code]....
i am trying to get centos 5.4 installed and bootable on my 16gb flash drive, with persistent overlay using and ext3 formatted drive.i want to be able to boot into centos and be able to have all updates from yum, etc, saved when shutting down for my situation i cannot use vfat.
View 9 Replies View RelatedI installed CentOS. NetworkManager successfully installed and launched. I can ping any IP from terminal but I can't open any site in browser except 127.0.0.1 I can't open IP of the modem 192.168.1.1 in browser also.
The results of some commands:
ifconfig
eth0Link encap: EthernetHWaddr 7D:60:00:00:00:00
inet addr: 192.168.1.5Bcast: 192.168.1.255Mask:255.255.255.0
inet6 addr: fe80::7f60:ff:fe00:0/64Scope:Link
[Code]....
first of all, ill provide you with the package name i got [URL]...i went to /usr/src/redhat/SOURCES/bwbar-1.2.2 did ./configure ; make
then i get this problem : [root@server bwbar-1.2.2]# ./bwbar ./bwbar: /usr/local/lib/libpng12.so.0: no version information available (required by ./bwbar) i have installed libpng & libpng-devel
[Code]...
after looking around a bit i thought it would be better to get the sources from kernel.org directly and try to compile that one, but then i found it to have the exact same error. would be very nice if someone could direct me to a solution for this. i have the bwbar already installed on another server which runs CENTOS 4 , i installed it there AGES ago , so i dont really remember what was my method back then.
the CENTOS 5 server dose not like me installing bwbar on it for some reason. maybe its bwbar's fault?, im no coder so i hope anyone that got a clue could check this out for me.
i have installed latest skype-2.0.0.72-centos.i586.rpm from the skype website. i have installed it at a datacentre which runs in headless mode the procedure i followed to start skype is
i)Xvfb :1 -screen 0 800x600x24 &
ii)export DISPLAY=localhost:1.0
iii)echo skypeusername skypepw | skype --pipelogin &
it seems the skype not working as i was unable to see this user's status as logged in from another skype account on a different system the architecture is: CentOS release 5.3 (Final) uname -r : kernel- .6.18-128.1.6.el5 uname -a: Linux localhost.localdomain 2.6.18-128.1.6.el5 #1 SMP Wed Apr 1 09:19:18 EDT 2009 i686 i686 i386 GNU/Linux
[Code]...
I installed the latest version of Centos 5.4.It came with Xen version 3.0.2 I think.I am running it on a Dell R510. It supports hardware virtualization.I was able to virtualize Windows XP and Windows 2003 server on it by selecting the install from DVD option.I was having problems trying to get a PCI DekTek card seen by the virtualized Windows domains so I tried to upgrade XEN to 3.4.2.I used the repos available from Gitmo and it went easy.
The only problem is now that when I try to create a "New virtual host" it only allows me to use HTTP, FTP option and not the iso or DVD option. They are greyed out. I was wondering if anyone has had this problem and might know a work around for the Xen 3.4.2 problem. Perhaps a better location for the kernel or something.
Agere got purchased by LSI and now LSI tech support claims there is no Linux driver for this card even though the literature clearly states the availability of a Linux driver.I have installed the firewire driver mentioned in other forum posts from ELRepo.
View 4 Replies View RelatedI installed CentOS 5.3 on my Shuttle SS-31T with SiS 190 network adapter. While booting the message "Bringing up eth 0 [ OK ]" appears. After booting it isn't possible to surf to an internet location (p.a. www.google.com). It's also impossible to ping other machines in the network. After typing ifconfig everything seems to be ok.
ipv6 is turned off
IP : 192.168.0.3
Subnet : 255.255.255.0
Gateway : 192.168.0.1