Simple PXE clients do not support DNS, they'd need the external IP
But if you use ipxe on the clients then you can use DNS
but if I make the router find the main (company) network via DDNS, is it possible?
Rodrigo Alves: which router models are you using at the client locations?
For example, it would certainly be possible to set up a VPN between the central site and the branches; then you could use IPs even if the central site is reached via DDNS
It's also possible to host the ltsp image locally at the branches, so that the clients boot from that, and AFTER they boot, they'd mount /home from the central ltsp server
That would also be the easier and fastest solution
|05:01||vagrantc has joined IRC (vagrantc!~vagrant@2600:3c01:e000:21:7:77:0:20)|
|05:24||vagrantc has left IRC (vagrantc!~vagrant@2600:3c01:e000:21:7:77:0:20, Quit: leaving)|
Finally, ipxe does support DNS (and DDNS), you could load an ipxe script in a variety of ways
|08:18||ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)|
|13:01||fiesh has left IRC (fiesh!~fiesh@2003:fb:1018::21, Quit: system reboot)|
<alkisg> "Rodrigo Alves: which router..." <- Roteador Wireless MU-MIMO AC1900 Tp-Link Archer C80
<alkisg> "For example, it would certainly..." <- hmmm got it
<alkisg> "It's also possible to host the..." <- in that case would I have to have a redirect server in the clinics?
sorry for the question, but this subject was never discussed (like the RDP I asked above? 😑)
|13:43||fiesh has joined IRC (fiesh!~fiesh@2003:fb:1018::21)|
Yes, in that case you'd need a local tftp and nfs server
You don't need to call it a "redirect server", it's a normal nfs server, it serves things locally without redirecting anywhere
How many local rdp clients do you have per clinic?
|15:40||vagrantc has joined IRC (vagrantc!~vagrant@2600:3c01:e000:21:7:77:0:20)|
<alkisg> "Yes, in that case you'd need a..." <- Ah understood
<alkisg> "You don't need to call it a "..." <- I'm calling it a redirect, because it's going to be the intermediary between the clients (at the clinic), and the server (at the company) isn't it??
<alkisg> "How many local rdp clients do..." <- there are 10 computers
Rodrigo Alves: the local clients, will they need a linux account and /home dir, or are they just going to be used for rdp?
If it's just for rdp, then you don't need a central ltsp server at all
no, the idea would be that the clients (from the clinic) do pxe on Ubuntu (which is in the company), and the pxe (through autologin) goes to rdp windows
Then just put a raspberry pi or another small PC at each clinic, configure it as an ltsp server, and make it serve the ubuntu rdp image to the clinic ltsp clients
You don't need a central ltsp server at all
You can clone that ltsp rdp image between clinics, e.g. you can maintain it in a VM in your own PC
alkisg: hmmm interesting
alkisg: I understood
so the suggestion is the following: In the clinics I must have an RDP computer, which connects to the central server (which is in the company, and which is the LTSP). So the clients boot this RDP computer and from it they go to the central server, is that the idea??
|18:02||vagrantc has left IRC (vagrantc!~vagrant@2600:3c01:e000:21:7:77:0:20, Quit: leaving)|
|18:03||vagrantc has joined IRC (vagrantc!~vagrant@2600:3c01:e000:21:7:77:0:20)|
You have 10 PCs in the clinic
You put a disk to one of them and call that "a local ltsp server"
You install ubuntu ltsp on it
The other 9 clients boot from it as ltsp clients
Then all the clients, including the local ltsp server, run xfreerdp, and connect directly to your central windows server via rdp
You don't need a central LTSP server for that
Did I misunderstand something? You want to RDP to a Windows server, right? (otherwise you don't need rdp at all...)
|18:30||woernie_ has left IRC (firstname.lastname@example.org, Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)|
|18:32||woernie has joined IRC (email@example.com)|
|18:32||woernie has left IRC (firstname.lastname@example.org, Read error: Connection reset by peer)|
you understood very well, and I'm considering using your suggestion, it's very interesting
document the networks and DHCP to yourself well before you start implementing. And test first one network at a time
it can leave :)
|20:00||ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)|
|23:26||jgee636 has joined IRC (email@example.com)|
|23:27||jgee636 has left IRC (firstname.lastname@example.org, Client Quit)|
|23:28||jgee has joined IRC (email@example.com)|
|23:30||jgee has left IRC (firstname.lastname@example.org, Client Quit)|
|23:37||jgee has joined IRC (email@example.com)|