IRC chat logs for #ltsp on irc.freenode.net (webchat)


Channel log from 7 June 2020   (all times are UTC)

00:08oh207 has joined IRC (oh207!~oh207@h114.191.132.40.static.ip.windstream.net)
01:35vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
02:28woernie has joined IRC (woernie!~werner@p5ddedb95.dip0.t-ipconnect.de)
02:29woernie_ has left IRC (woernie_!~werner@p5b2965c8.dip0.t-ipconnect.de, Ping timeout: 265 seconds)
03:08oh207 has left IRC (oh207!~oh207@h114.191.132.40.static.ip.windstream.net, Ping timeout: 256 seconds)
06:55statler has joined IRC (statler!~Georg@p54897b60.dip0.t-ipconnect.de)
09:48statler has left IRC (statler!~Georg@p54897b60.dip0.t-ipconnect.de, Quit: cu @ll)
10:14
<fiesh>
wow didn't know about x2go though, so thanks for that information, markit-who-aint-here-no-more
11:22
<alkisg>
fiesh: xrdp might be a better option than x2go
11:22
And it might even work on wayland later on...
11:24
<fiesh>
oh I see
11:24
thanks for the hint
11:24
<alkisg>
np
12:25ltspbot has joined IRC (ltspbot!~supybot@devs.ts.sch.gr)
12:52GodFather has left IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com, Ping timeout: 258 seconds)
13:35GodFather has joined IRC (GodFather!~rcc@64.53.141.7)
13:41GodFather has left IRC (GodFather!~rcc@64.53.141.7, Ping timeout: 260 seconds)
13:41GodFather has joined IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com)
14:05GodFather has left IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com, Remote host closed the connection)
14:10GodFather has joined IRC (GodFather!~rcc@d53-64-7-141.nap.wideopenwest.com)
14:52markit has joined IRC (markit!~marco@mail.ammdomus.it)
15:02markit has left IRC (markit!~marco@mail.ammdomus.it, Read error: Connection timed out)
15:03markit has joined IRC (markit!~marco@mail.ammdomus.it)
15:04
<markit>
alkisg: hi, I've a new lab with recent Intel Nuc pc, probably they are able to do http boot, I don't understand how to try and if https://github.com/ltsp/ltsp/issues/75 has the solution with +dhcp-boot=tag:iPXE,"http://${next-server}/ltsp/ltsp.ipxe"
15:32RaphGro has joined IRC (RaphGro!~raphgro@fedora/raphgro)
15:37mmarconm has joined IRC (mmarconm!~mmarconm@191.223.17.42)
16:01mmarconm has left IRC (mmarconm!~mmarconm@unaffiliated/mmarconm, Quit: Leaving)
16:08
<alkisg>
markit: why do you want to do http boot, is tftp slow?
16:22
<markit>
alkisg: boot seems fast enough, but if I can have "cheap" faster one, it's welcome :)
16:23
btw, do I really need "NFS_HOME=1" in ltsp.conf? I don't have it and was sure it's using nfs for home (fstab_home=... nfs stuff)
16:26
(if I use mount | grep home in a client, I see is nfs4 mounted)
16:28
<alkisg>
markit: NFS_HOME=1 is for `ltsp nfs` to automatically generate the correct exports, without passing a --nfs-home=1 parameter
16:28
Try to add http://... to the boot filename in ltsp-dnsmasq.conf
16:28
And set up an http server in /srv/tftp
16:28
<markit>
ah I see, that explains why --nfs-home=1 is not specified in the example
16:29
btw, I find a bit confusing ltsp.conf with a mix and match of parameters for ltsp command and parameters for ltsp "work"
16:31
do you mean a web server that has /srv/tftp as "root dir", right? I'll have a look, and do some performance test, thanks a lot
16:32
<alkisg>
Right,
16:32
regarding the confusion, what do you mean?
16:32
All parameters in ltsp.conf get into the environment; existing environment variables also work; and parameters also work
16:32
It's the normal way for all programs, not just for ltsp
16:35
<markit>
from the "developer" prospecting, you are right, they are the same. From the "user" point of view, I've parameters that set the "client environment" (like FSTAB_, NAT etc) and other that are NOT related to the "runtime" of ltsp, like "MKSQUASHFS_PARAMS"
16:36
<alkisg>
And that's what the sections [server] and [client] are for...
16:36
<markit>
in the "[server]" section I think there are parameters for the "ltsp runtime server side"
16:36
<alkisg>
Would you prefer separate files instead? I think it would just be more confugsing
16:36
<markit>
I will do
16:36
would
16:37
whatever :)
16:37
<alkisg>
ltsp.conf already has a lot of sections, for [OLD_MONITOR] or for [mac:address]
16:37
And the administrator is expected to understand how they're all combined etc
16:37
Having a common place for TFTP_DIR or DNS_SERVER, that affects both clients and server, makes sense
16:38
<markit>
like "NFS_HOME=1" I don't understand at first sight if is a "runtime" (enables NFS stuff) of "default config of ltsp command" parameter
16:38
<alkisg>
Now you can even have the same ltsp.conf in different servers
16:38
And use [server-mac1] and [server-mac2]
16:38
To execute different commands and parameters on each different ltsp server
16:39
NFS_HOME isn't in ltsp.conf, so it's not for the clients
16:39
All command line parameters can be defined as ltsp.conf options
16:39
So, it's --nfs-home that's documented
16:39
(in `man ltsp nfs`)
16:40
<markit>
I understand, but don't agree that command line parameters "defaults" and "runtime parameters" be in the same file
16:40
I will get used to it of course, but is not the first time I get confused
16:41
<alkisg>
Another example. You may use this:
16:41
[server]
16:41
POST_SERVICE_DO_SOMETHING="blah"
16:41
This runs blah when the *server* boots
16:42
<markit>
really?
16:42* markit takes note
16:42
<alkisg>
It's extremely flexible, but more on that later, gotta go!
16:42
See you!
16:42
<markit>
thanks a lot as usual :)
17:10vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
19:08
<alkisg>
So, another example: [server1-mac] NFS_HOME=1, and [server2-mac] NFS_HOME=0, this would allow having the same ltsp.conf on two servers, where one of them has nfs home enabled, and the other one not
19:09
I.e. this scheme allows [mac address] sections to affect servers as well
19:10
<vagrantc>
/8/8
19:10
hah
19:10* vagrantc waves :)
19:10
<alkisg>
Hey vagrantc
19:10
How are things? No news from you in a while...
19:11
<vagrantc>
not too much goings on
19:11
you're still busy improving LTSP i see :)
19:11
<alkisg>
Nah, it's been mostly paused until August when I'll refresh it to prepare it for schools
19:12
But I've used it in ~20 schools so far, so I don't expect many things to change
19:12
It even supported multiseat with just 1 additional file... it's flexible
19:12
You? Mostly working on reproducible builds?
19:13
<vagrantc>
yeah, it seems nicely flexible
19:13
mostly reproducible builds and my solar project ... some arm laptops ... arm boards ... a bit of guix ... the usual :)
19:14
<alkisg>
Heh, I need to try guix at some point
19:15
<vagrantc>
instead of getting thrown around in aikido, swinging sticks around in the park
19:15
alkisg: part of me thinks you'll love it ... and but i also suspect you'll find it a mess :)
19:17
<alkisg>
Currently I'm doing a sprint to finish my phd... it's going well, the questionnaire is already uploaded in surveymonkey... hopefully I'll finish it until xmas and have enough time for experimenting with all these new tools...
19:22
<vagrantc>
alkisg: what's the nature of your phd?
19:24
<alkisg>
In 3rd grade of greek high schools, a theoretical programming language was used, similar to pseudocode. I developed an IDE+intepreter for it, and supported it in a forum for around 2 decades. It's a mix of all that, plus educational issues regarding the lesson
19:25
<vagrantc>
seems more than theoretical then :)
19:26
<alkisg>
I teached that course in a school in 2000, and I saw the need for it; when I published it, thousands of teachers started using it because they didn't want to keep teaching in the blackboard...
19:26
Then they never put me in a high school again :P
19:28
markit: another example is DEFAULT_IMAGE, KERNEL_PARAMETERS and MENU_TIMEOUT. These are per client, so they go under [mac-address]. But, it's `ltsp ipxe` that uses them, to generate the appropriate ltsp.ipxe
19:29
So if I was to follow the idea of a separate server.conf, then you'd need to put client mac addresses there as well
20:03
<markit>
alkisg: is not the idea of separate server.conf, but separate "command default" from ltsp configuration
20:04
"multiseat"? very interesting
20:09RaphGro1 has joined IRC (RaphGro1!~raphgro@fedora/raphgro)
20:11RaphGro has left IRC (RaphGro!~raphgro@fedora/raphgro, Ping timeout: 256 seconds)
20:12RaphGro1 has left IRC (RaphGro1!~raphgro@fedora/raphgro, Client Quit)
20:52markit has left IRC (markit!~marco@mail.ammdomus.it, )
21:53Talikka has left IRC (Talikka!uid18862@gateway/web/irccloud.com/x-ilrhoragxwxrhdye, *.net *.split)
21:54Talikka has joined IRC (Talikka!uid18862@gateway/web/irccloud.com/x-ilrhoragxwxrhdye)
21:55bengoa has left IRC (bengoa!~alberto@194.50.55.200, *.net *.split)
21:55Hyperbyte has left IRC (Hyperbyte!~jan@middelkoop.cc, *.net *.split)
21:55bengoa has joined IRC (bengoa!~alberto@194.50.55.200)
21:55Hyperbyte has joined IRC (Hyperbyte!~jan@middelkoop.cc)
21:55bcg has left IRC (bcg!~b@df-vm8yyyyyyyyyyyyyyt-3.rev.dnainternet.fi, *.net *.split)
21:56bcg has joined IRC (bcg!~b@df-vm8yyyyyyyyyyyyyyt-3.rev.dnainternet.fi)
21:56alexxtasi[m] has left IRC (alexxtasi[m]!alexxtasim@gateway/shell/matrix.org/x-kuphsbwzqetgdqhr, Ping timeout: 244 seconds)
21:57shored has left IRC (shored!~shored@87-92-92-55.bb.dnainternet.fi, *.net *.split)
21:57mgariepy has left IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy, *.net *.split)
21:58shored has joined IRC (shored!~shored@87-92-92-55.bb.dnainternet.fi)
21:58mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy)
22:17alexxtasi[m] has joined IRC (alexxtasi[m]!alexxtasim@gateway/shell/matrix.org/x-ivslooitmahrjrvt)