Forum Replies Created
-
Posts
-
Hi,
the window that pops up is the NoMachine panel showing what desktops are available. When you connect to a Terminal Server product, Terminal servers always show this panel because it asks you to select either the physical or a virtual desktops. I’m assuming you have a TS product installed, so you could try the free edition or the Enterprise Desktop product, both of which are for connecting to the physical display only. In that case, your command will connect you directly to the desktop.If you have a vnc session running, killing it will then allow NoMachine to create the virtual xframebuffer session. After that restart the nxserver with
nxserver --restart. Check what is configured in the DefaultDesktopCommand key in the node.cfg file. You could try using, for example, the same command you use on the vnc server.If none of the above help, then we require the NoMachine logs with debug enabled from the server host. You should follow the instructions here https://kb.nomachine.com/DT07S00243, reproduce the problem, and then submit here or send to forum[at]nomachine[dot]com.
Moreover, I’m surprised at not seeing these config files
Those others you mention are related to the installation and not the configuration and the installation dir is not the scope of the
--configsavecommand. Backing up the install path would create problems when installing on a different host if you are not using exactly the same path. What about if you want to move your configuration to a different OS? As mentioned in the article I linked, the configsave command is to 1) “preserve custom configurations, when it’s necessary to uninstall the current server for whatever reason and install it again”. Or 2) to “migrate the current server installation to a different machine, even running a different Operating System. (E.g. migrate from Windows to Linux)” or to 3) “in the case of multi-node setup, to preserve the current list of nodes and avoid re-adding and configuring them manually after a reinstall or a host migration.”About the second question you posted, what do you mean with “some files seem to have been forgotten”? Did you try executing the configrestore command as per the article. What was missing?
Hi,
nothing has changed between NoMachine 8.4 and 8.16 in how NoMachine Free Edition works. If the server is not headless, you will connect directly to the physical desktop. On Linux, when connecting to the physical desktop which doesn’t have an X server running (e.g. it’s headless), NoMachine Free Edition is able to use its own display service (which is an embedded X server) to let a user connect seamlessly to a physical display running in the background on the remote machine. If you install it on a headless host, the following message appears:
“Cannot detect any display running, Do you want NoMachine to create a new display and proceed to connect to the desktop?”
Selecting OK will create a virtual display (NoMachine’s embedded X server). Is your Debian headless? Are you not seeing this message?
It’s worth pointing out that the free edition does not create virtual desktops like the terminal server products in the Enterprise suite. Virtual desktops are individual desktops, independent from each other that can run on the same server simultaneously and can be left running while you are disconnected. To be able to get multiple virtual desktops, you need a Workstation installed (or higher end product).
Hi, in version 9 an Enterprise Cloud Server Cluster (ECSC) is a pair of Cloud Servers that work together to maintain high available access to the centralized infrastructure. Let me sum up how it will work. This is an active-active cluster where the cluster master and slave roles can be switched between the primary and secondary servers. The master is in charge of distributing client connections and the load of session negotiation and management between itself and the slave. The primary and secondary server constantly monitor each other, hence active/active. When one of the servers loses contact with the other one, it replaces the failed server. Sessions already managed by this working server remain connected and the failover is transparent for them. The management of sessions handled by the failed server is passed from the failed server to the working server. In that case, the sessions running on the remote nodes will be automatically reconnected to the working server.
Hi, we need additional logs and further information.
1. Edit ‘/usr/NX/etc/server.cfg’ file, remove the # from the key name and set the following key:
SessionLogLevel 82. Restart the NoMachine server:
$ sudo /etc/NX/nxserver --restart3. reproduce the problem and upload log archive:
$ sudo /etc/NX/nxserver --debug --collectIt would also be useful to show some information about the physical session:
4. Show the X server process:
$ ps aux | grep X5. Show the X server process environment variables:
$ cat /proc/<X server pid>/environ6. List the user sessions:
$ loginctl list-sessions7. Show information about the physical session (with Type=wayland):
$ loginctl show-session <session-id>8. Show information about the session owner (parameter “User=” from the previous request):
$ loginctl show-user <user-id>We got the logs thank you, and we’ll come back to you soon about those. Do you confirm you are always getting black screen regardless of the client OS you are connecting from?
Did you try the workaround in the other topic? I.e did you disable eglcapture with
/usr/NX/bin/nxserver --eglcapture noon the Kubuntu host?It seems you disabled debug on the server this time, possibly you executed only
--collectwithout the--debug. We got only the debug agent logs (from the special package). You wrote that the black screen problem only shows when connecting from iOS but when you tried the special package we sent you, you are now getting black screen no matter what client you are connecting from? Is that correct?1) Enable debug level on the problematic server/node machine:
nxserver --debug --enable all
2) Connect to the NoMachine server and reproduce the problem
3) Create the log archive on the server/node machine:
nxserver --debug --collectThanks very much for your collaboration.
When you mention NoMachine’s interoperability with other hypervisors, can you confirm whether you are just referring to the fact that products like NoMachine Desktop can be installed onto VM’s managed by those hypervisors?
That’s right.
I am currently thinking I might be able to use NoMachine Terminal Server paired with a custom session to accomplish what I am trying to do.
Take a look at the article here: https://kb.nomachine.com/AR10K00720. NoMachine supports VNC sessions. You need to enable VNC on the TS/ETS host first and then you will be able to run a VNC session to the VM. In the ‘VNC remote desktop’ panel, provide the IP or hostname of the VNC server and port to continue. In the next panel you will be requested to provide username and password to login to the VNC server.
Hi, take a look at this article here: https://kb.nomachine.com/AR05T01160. You can use the
--configsavecommand.Hi, that scenario is not supported, I’m afraid. We had planned to develop our own Virtualization Server, but with so many options on the market at the time, we decided to invest efforts in developing our current product suite (for example, Cloud Server which besides functioning as the gateway to your infrastructure is also a powerful remote access management tool). We also invested efforts in improving NoMachine’s interoperability with the multitude of hypervisors already available (not just physical workstations) as well cloud-based platforms. Indeed all NoMachine products integrate seamlessly with VMware, Virtualbox, Proxmox and other hypervisor solutions, as well as cloud providers such as AWS EC2, Amazon WorkSpaces Core, Azure, and Google Cloud Platform for example.
Hi, at the moment there is no setting to force the windows to stay minimized. The NoMachine window pops up to report a connection status change which could be useful for the user, and to apply some window configuration requests when needed. Your request is certainly a reasonable one though, so we’ll see how we can improve this aspect 🙂