We have a server were we no longer seem to be able to gain access, neither locally or remotely.When we try to connect via SSH we get a "Connection Refused". We have checked that it's not our standard firewall blocking the request, so it can either be the local firewall (which we disabled when installing) or that the process isn't running.However, when we try to login locally on the machine (as root) all we get is "Login failed". We've tried restarting the machine in single user mode to change the root password, and while we can do this succesfully, it still doesn't give us access when rebooting the machine afterwards.No other users (besides the standard users created when installing) are available on the server, so we can't say if this is a problem with the root account, or with logging on in general.
Does anyone know how to do a 'shadow' session with freenx client, qtNX?I don't see any options in that particular client to 'attach' or 'shadow' a current X11 session. Does anyone have a slackbuild for nomachines NX client?
I installed freenx-0.7.3-i486-1alien.tgz on the server (with Slack 12.1), all according to instructions. I installed the no machine nxclient 3.4.0-7 on the client (with Slack 12.2). I ran setup and configured. I used the default no-machine keys, not custom ones. Since authentication failed at the beginning, I enabled DBauthentication, and added my user and password, which seemed to allow authentication to occur.
Here is my node.conf: # node.conf # # This file is provided by FreeNX. It should be placed either into # /etc/nxserver/node.conf (FreeNX style) or /usr/NX/etc/node.conf # (NoMachine NX style).....
The following message comes up when I boot up: Logging in user Warning: Cannot open ConsoleKit session: Unable to open session: Launch helper exited with unknown return code 0. When I press OK, the system completes the start up and everything looks normal. But when I try to connect to internet, I get the following message:
KNetworkManager cannot start because the installation is misconfigured. System DBUS policy does not allow it to provide user settings; contact your system administrator or distribution. KNetworkManager will not start automatically in future. If I reboot the system, I logg in successfully. So far the problem has appeared approximately upon every second time I boot up. Rebooting the system seems to take care of it.
Don't know what info is of interest. I'm using
Opensuse 11.2 KDE 4.4.2 (Factory) After upgrade from 4.4.1 to 4.4.2 it worked fine for a week or so.
I do a clean install of slackware64 13.1 beta1 with KDE and switch default runlevel to 4 in /etc/inittab.
I try to login in kdm, I always come back to the login....
I try this with default runlevel 3 and an .xinitrc with "ck-launch-session startkde" .. works without problems, so I switch back to default runlevel 4, now i can login and only get the error "Cannot open ConsoleKit session: Unable to open session..."
I've spent days trying to setup access properly from a public address to a monitoring server that works fine locally. Everything works from public access until I try to link to a CVS repository. The rancid CVS repository is set up as a separate server (virtualhost). It appears the referring link causes a DNS error (105: Server Not Found) when the CVS repository server is accessed from the public address. Things work fine when accessing via localhost.
Public link: (this results in 105 error caused by redirection (bold portion of link))
I am trying (desperately) to setup freeNX so that I can access my Mandriva box from my Macbook over my network. I have great need to be able to do this. Here's the necessary info for you: Mandriva PC: running Mandriva 2010.0 with freeNX installed from the Mandriva repos Macbook: running OS 10.4 with the Nomachine NX Client installed Now, I have followed the setup from the Mandriva wiki as well as the freenx wiki to no avail. I simply cannot connect the Macbook's NX client to the server. I get an authentication failed message. However, I can ssh from the Macbook to the server just fine (that's how I'm getting most of this work done).
I've been running a Samba server under RedHat 8 for five years without a hiccup. I want to cut over to a F10 box but cannot get shares accessible. smbclient attempts fail over NT password error. SELinux is disabled. Server is visible on the network. Users require no password access to shared data.
# Samba config file created using SWAT # from UNKNOWN (>) # Date: 2009/06/12 14:15:15
Is there an open source application that can download e-mails from a generic POP mail server, keep a copy of the mails locally and then deliver copies to multiple accounts on a LAN? I have a friend who uses a group e-mail account, and more than one person is always hooked onto that account. The problem is they cannot leave e-mail on the server, as it eats out their webhosting space, but if someone downloads a message in Outlook, the other users do not see that message.
I've got an SSH server setup on my box which is running locally fine. I can connect to it through my local laptop but when I try to connect through the internet either via my IP or using the DynamicDNS I have setup both get an operation timed out message.I've got firestarter running with an inbound policy to allow connections from any host to port 22. I've also gone into my Netgear router and selected to forward any inbound connections to port 22 to the IP of the box that's got the SSH server on it.
Even with all this the connection just isn't being made. I've also tried temporarily setting the DMZ option on the router to forward all inbound connections to the IP of the Ubuntu box but still get the same Operation time out. Is there any other options that wouldn't allow remote connections via SSH? Or am I missing something hugely obvious here? Maybe it's possible to connect to SSH via a different port or something?I'm running Ubuntu 10.4 and have installed openssh-server via Synaptic. Would be great to get this working so I can use it as a tunnel from work and access the VNC, which I've been told should be done via SSH anyway.
I'm using wmctrl on an Ubuntu machine to manage windows from a script, that I run inside a (gnu) screen.
If I start the screen session from the local machine, wmctrl works fine, including if I completely close the terminal window and issue the wmctrl commands when connecting to the screen remotely via ssh. Conversely, if I connect remotely with ssh and start a screen, wmctrl doesn't work (returns "Cannot open display") even if I attach that screen session locally from the Ubuntu Terminal.
I guess there is some hidden screen parameter that doesn't get set in a way that allows accessing the display when it's launched remotely -- any ideas what it is and how to modify it from within a remote ssh-screen session so that that the script can access the windows?
Also, I would like users to authenticate against using LDAP rather having created manually in the NX server. I installed using yum install freenx nx and rename the node.conf.sample to node.conf and installed the windows client. but when I am trying to conenct the nx server it says it is connected but it gives me NX> 202 Authenticating user: nxNX> 208 Using auth method: publickeyNX> 204 Authentication failed.I have change this nx user in sshd_config file to user that i created using nx server command which exists in local machine.
Currently I have successfully setup a server as remote freenx-server terminal and I may login to this system using local users.
As I want a central ldap based user management, I have setup a working ldaps configuration. I am able to login as a ldap user via ssh.
Now the strange issue I am facing is that the freenx-server won't let me login to my server anymore when I enable the following setting in /etc/ssh/sshd_config:
Code: UsePAM yes Naturally if I disable UsePAM, ldap based logins will fail because they are passed through PAM. uname -a: Code: Linux srv01 2.6.26-2-openvz-amd64 #1 SMP Thu Feb 11 01:40:09 UTC 2010 i686 i686 i386 GNU/Linux my sshd_config:
I've been working on getting NoMachine and FreeNX working together ... and they're both working well, listening on different ports (guess I should write a howto). At some point I decided to write an init script to control FreeNX behaviour (start, stop, status, setup), since the Ubuntu version has one (called freenx-server) and openSUSE doesn't, but I noticed that it was not necessary, as FreeNX got started automatically... but how ? I could not find any init.d script or command which would start the FreeNX server. Indeed it wasn't actually running :
Code: neelix:~ # /usr/bin/nxserver --status NX> 100 NXSERVER - Version 3.2.0-73 OS (GPL, using backend: 3.2.0)
I have been messing around with freenx to see if I can support my customer via remote desktop... I have installed the server on Ubuntu Karmic and the client on Windows 7. I love linux but need the windows computer for customer work and that is why I chose to install the client on windows. I setup the server the best I could and created custom keys which I then copied to windows and imported them to the freenx client. I am looking for security for my customer and that is why I chose custom keys... There are several issues to deal with but I want to just get started to see where it takes me. I added the test server computers ip address to the freenx client configuration and tried to connect to Ubuntu... I have left the port as 22 at the moment
Desktop setting on the client...I chose vnc which seems to be a good choice according to all the material I have been reading....Other settings I have left alone at the moment to see how things work.... All that said - I have the login window asking for information... I have the password asking for information....and I have the session wanting information....? Obviously I understand that login info can be set on the server and the password...but I haven't seen how to set up that? Of course I am testing the server from a client on my home network at the moment - but the customer has his own router as do I so those settings will need to be looked at as well.
I'm having issues with my SSH server.The server is running on a Slackware 13.37 machine, using the default SSH server software, which I believe is OpenSSH.On my Windows machine, I've tried connecting to my Slackware machine but can only do so using the external IP, not the LAN IP, instead I get a "Connection Refused" error using PuTTy. I've forwarded port 22 for both UDP and TCP packets to my Slackware machine.I don't really know what information you need so just ask me and I'll respond to the best of my abilities.
I just finished setting up my small server, however I realized that no one can connect to it from the outside. The reason being is that my router is assigning internal ip's and dyndns is linked to that. I need port 80 for the Apache server, but if I have the router forward that port to my particular ip, won't it screw with everyone else's internet too?
I have a home server that I was able to up and till recently able to connect to externally and internally but something has changed. I can't connect through vnc, putty or ftp. if I ping the local address it can't be found. If I try to get to it using its external IP address I can get to the server and put in my user name but it wont accept my password.If I go to the server and use the password it works fine.
We've set up a Red Hat Enterprise Linux 5 server and can log into it remotely just fine (with root and other users), but if I try to log in locally with a keyboard and monitor connected to the machine, I keep being returned to the login prompt. If a login is incorrect, I get an error message, but if it is correct then the screen just blanks and shows the login prompt again.While booting the only error message that comes up has to do with SNMPD, so I don't think it's related.
I read this post on the forum which seemed like a similar problem but it didn't help (I wasn't getting the "Module is unknown" message he was). I also read another post somewhere that said he was having the problem because the server was trying to use AD to authenticate, but I checked that and it's not the case with our server.From what I'm told (I didn't set up the server myself) it used to work fine, but then "something" happened one day that made it do this.
Sorry if this is in the wrong group but its a bit of an odd one but I think comes under Server Support.. I've set up both x11vnc and freenx in an attempt to eliminate either as a cause for this issue. The problem is that when I log in with either, I can get into the desktop and run things from the menus but when I make the System menu active - the session hangs and you have to killall for the freenx or vnc processes.
I was wondering if anyone might be able to help me investigate this and hopefully fix it as I'm not sure where to look given that both types of remote connection seem to have the same issue.
I've created a brand new CentOS 5.4 (Final) 64bit machine AMI on Amazon EC2. This was based off an existing image. I was able to follow the wiki to add NX server. I am using WIN XP desktop for NX client.
I can connect to the EC2 machine and get the GNOME desktop fine. I see the usual CentOS desktop and poke around.
I expected to open the Terminal window and get a shell prompt to su into root user (I need to be root to install some software that needs GUI). I do not want to install this from my plain SSH connection to EC2 (hence the NX server etc.).
When I open the Terminal window, all I get is the NX>105 prompt. I need to get to a shell prompt so I can su into root. For life of me, I cannot get around this prompt (I looked at NX documentation too). Note that this is a prompt NOT on client but on the remote machine. I do not need this as I'm already authenticated and logged in to remote GNOME desktop.
Obviously the TERMINAL is running some NX start up script (I've no idea which one). If there is some other way to sudo into root?
I'm running vsftpd-2.2.2 on Slackware 13.1 and I'm behind a NAT with a dynamic IP. I'm using pasv_address, which makes it necessary to have a crontab that checks whether my IP has changed, and if it has, edits the conf and restarts vsfptd. It's hackish but it works. I have also enabled FTPES on the same port that normal FTP runs. I have allowed only the 5001:5003 ports and I am aware that the recommented is 50 ports. However I never had an issue with the transfers since I'm the only user and have at maximum two simultaneous transfers.
I can connect remotely with FTP and FTPES with every client I tried but I'm having issues locally. FileZilla refuses to connect on FTP and FTPES. WinSCP can connect on FTP but not FTPES. Issues are gone if I enable pasv_promiscuous, which the manpage strongly recommends not to use unless "you know what you're doing". I don't. Why is this happening? And, can I fix the local connections without resorting to pasv_promiscuous?
Here is my config and log files from a failed local attempt.