|00:48||dgroos has joined IRC (firstname.lastname@example.org)|
Hello again! I haven't got screen broadcasting on epoptes to work on ltsp19. I can log out, shutdown and WOL the clients w/epoptes, though.
|02:42||dgroos has left IRC (email@example.com, Quit: dgroos)|
dgroos, in case you see irclogs, this means that the *system* epoptes-client isn't running, while the *session* epoptes-client is running
|05:55||kjackal has joined IRC (kjackal!~quassel@2a02:587:3107:2e00:9d60:90ea:afa0:294f)|
|06:01||statler has joined IRC (statler!~Georg@p5B30EE2F.dip0.t-ipconnect.de)|
Aaaah sorry, or it means you're using wayland :)
VNC won't work in Wayland
|06:08||kjackal has left IRC (kjackal!~quassel@2a02:587:3107:2e00:9d60:90ea:afa0:294f, Remote host closed the connection)|
|06:16||ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)|
|06:26||vsuojanen has left IRC (firstname.lastname@example.org, Ping timeout: 246 seconds)|
|06:28||vsuojanen has joined IRC (email@example.com)|
|06:44||woernie has joined IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de)|
|07:00||woernie has left IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de, Remote host closed the connection)|
|07:09||georgeneophytou has joined IRC (firstname.lastname@example.org)|
|07:17||statler has left IRC (statler!~Georg@p5B30EE2F.dip0.t-ipconnect.de, Remote host closed the connection)|
|07:36||kjackal has joined IRC (kjackal!~quassel@2a02:587:3107:2e00:f9c5:a75:9a6a:688d)|
|07:38||woernie has joined IRC (email@example.com)|
|07:39||georgeneophytou has left IRC (firstname.lastname@example.org, Quit: My MacBook has gone to sleep. ZZZzzz…)|
|07:46||Leolo_2 has left IRC (Leolo_2email@example.com, Ping timeout: 264 seconds)|
|07:59||gdi2k has joined IRC (firstname.lastname@example.org)|
|08:32||georgeneophytou has joined IRC (email@example.com)|
I am still undecided if wayland is a good idea
I think it's necessary, to reflect all the hardware changes
It went from "draw an arc and a rectangle to display a window", to bitmaps, to opengl, to 3d only rendering and compositors... not sure that xorg codebase can be kept clean after all those changes
no I understand the necessity to replace xorg, I am just not sure wayland is an adequate solution for that
|08:49||statler has joined IRC (statler!~Georg@gwrz3.lohn24.de)|
|08:50||georgeneophytou has left IRC (firstname.lastname@example.org, Quit: My MacBook has gone to sleep. ZZZzzz…)|
|08:51||meo has left IRC (meo!~systemdju@unaffiliated/mikeseth, Ping timeout: 276 seconds)|
|08:52||georgeneophytou has joined IRC (email@example.com)|
|08:54||georgeneophytou has joined IRC (firstname.lastname@example.org)|
|08:55||statler has left IRC (statler!~Georg@gwrz3.lohn24.de, Ping timeout: 276 seconds)|
|08:56||sas34 has joined IRC (email@example.com)|
|08:57||georgeneophytou has left IRC (firstname.lastname@example.org, Client Quit)|
hmm a possibly dumb question: ltsp-chroot doesn't mount the distfiles for me (that's where gentoo saves the package sources in)
where can I add that so -c / -m mount them automatically?
Ouch LTSP5 :D
|09:01||* alkisg refuses to re-read the ltsp5 code base :D|
I haven't seen any updated version in the gentoo portage tree... :-P
(not that there's any version in there to begin with)
If I recall correctly, you should provide a -vendor-functions override, and from there, override the mount_package_cache function (or a similar name)
|09:08||statler has joined IRC (statler!~Georg@gwrz.lohn24.de)|
|09:08||georgeneophytou has joined IRC (email@example.com)|
|09:09||meo has joined IRC (meo!~systemdju@unaffiliated/mikeseth)|
ah thanks, there's a /usr/share/ltsp/ltsp-chroot-functions that was altered
I mean adapted to gentoo
|09:37||georgeneophytou has left IRC (firstname.lastname@example.org, Quit: My MacBook has gone to sleep. ZZZzzz…)|
|09:44||statler has left IRC (statler!~Georg@gwrz.lohn24.de, Remote host closed the connection)|
|09:45||georgeneophytou has joined IRC (email@example.com)|
|09:52||sas34 has left IRC (firstname.lastname@example.org, Ping timeout: 260 seconds)|
|10:18||kjackal has left IRC (kjackal!~quassel@2a02:587:3107:2e00:f9c5:a75:9a6a:688d, Remote host closed the connection)|
|10:22||kjackal has joined IRC (kjackal!~quassel@2a02:587:3107:2e00:2562:6e6:6bf1:6b28)|
|10:42||uumas has left IRC (uumas!uumaslyseo@gateway/shell/matrix.org/x-jjvgnqtozcmkxxye, Read error: Connection reset by peer)|
|10:42||enaut[m] has left IRC (enaut[m]!enautmatri@gateway/shell/matrix.org/x-iordagzqehqbzrin, Remote host closed the connection)|
|10:42||enaut[m]1 has left IRC (enaut[m]1!enauttchnc@gateway/shell/matrix.org/x-clesrfgkhehqtacu, Read error: Connection reset by peer)|
|10:42||uumas_ has left IRC (uumas_!uumasmatri@gateway/shell/matrix.org/x-aabxneizgieyvfhu, Write error: Connection reset by peer)|
|10:42||alexxtasi[m] has left IRC (alexxtasi[m]!alexxtasim@gateway/shell/matrix.org/x-rbadciahvdppcrpb, Write error: Connection reset by peer)|
|11:01||woernie has left IRC (email@example.com, Remote host closed the connection)|
|11:06||alexxtasi[m] has joined IRC (alexxtasi[m]!alexxtasim@gateway/shell/matrix.org/x-sqbkpvkyhbgojnap)|
|11:39||uumas_ has joined IRC (uumas_!uumasmatri@gateway/shell/matrix.org/x-vwpozhiohwlfxver)|
|11:39||Guest20711 has joined IRC (Guest20711!enautmatri@gateway/shell/matrix.org/x-pyrikjwwbfaetcdc)|
|11:39||enick_952 has joined IRC (enick_952!enauttchnc@gateway/shell/matrix.org/x-wfxqbzojbahueppc)|
|11:39||uumas has joined IRC (uumas!uumaslyseo@gateway/shell/matrix.org/x-cuufsuzwicmyqcgt)|
|11:42||dgroos has joined IRC (firstname.lastname@example.org)|
|11:45||Hyperbyte has left IRC (Hyperbyteemail@example.com, Ping timeout: 240 seconds)|
|11:47||Hyperbyte has joined IRC (Hyperbytefirstname.lastname@example.org)|
alkisg: Thanks for the reply. I'm not using wayland so I guess it's the system vs session epoptes issue. I don't really understand what that means so not able to generate possible troubleshooting procedures. What do you suggest?
|11:54||Faith has joined IRC (Faith!~Paty_@unaffiliated/faith)|
dgroos: I'll be back in 10-15 minutes if you're around then we can vnc
alkisg: I'm not at school now but will be there in about 20-30 minutes, thanks!
|12:10||section1 has joined IRC (email@example.com)|
|12:11||dgroos has left IRC (firstname.lastname@example.org, Quit: dgroos)|
|12:44||eu^205215175117 has joined IRC (email@example.com)|
eu^205215175117: in 10 mins now :)
vnc-edide: To share your screen with me, open Epoptes → Help menu → Remote support → Host: srv1-dide.ioa.sch.gr, and click the Connect button
eu^205215175117: anyway leave it open there and I'll have a look when I come back again
I'm not at the office atm so I'm only available intermittently
|12:55||dgroos has joined IRC (firstname.lastname@example.org)|
dgroos: login to a client
|12:59||gp has joined IRC (email@example.com)|
and oops, just another moment
What is the best way to use network home folders? Network mount on LTSP server and leave client home drive as is?
gp: I'm currently doing that and it works fine. I'll be switching to mounting directly on clients though because I have no reason really to have the ltsp server in the middle. Just a bit more work to setup.
|13:19||dgroos has left IRC (firstname.lastname@example.org, Quit: dgroos)|
|13:24||dgroos has joined IRC (email@example.com)|
gp,should be just one FSTAB line in lts.conf
dgroos: vnc disconnected for some reason
I think your issue is related to tigervnc
Try installing xvnc4viewer instead, and uninstalling tigervnc
will do, thanks!
|13:39||douglas_br has joined IRC (firstname.lastname@example.org)|
Is it possible to do ltsp client with Raspberry PI?
|13:51||gp has left IRC (email@example.com, Ping timeout: 240 seconds)|
douglas_br: yes, but it's hard
It requires terminal commands
E.g. the ability to maintain chroots or netboot over nfs rw
It needs better documentation before it becomes very easy
right I am seeing PINET, but I do not know
raspberrypi: (#1) Ubuntu/LTSP on Pi 2: https://help.ubuntu.com/community/UbuntuLTSP/RaspberryPi, or (#2) Debian/LTSP (with raspbian chroot) on Pi: http://cascadia.debian.net/trenza/Documentation/raspberrypi-ltsp-howto/, or (#3) unofficial Ubuntu/LTSP (with raspbian chroot) on Pi: http://pinet.org.uk/
Those methods used to work
Then they became unmaintained
|13:58||woernie has joined IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de)|
I tested rpi with ltsp19 and it works fine, but it needs documentation
what do you recommend for schools for low price and confortable experience
I booted rpi2, rpi3 and rpi4 with it
Stay away from raspberries :)
(unless you only want to do python classes without surfing at all)
|14:11||woernie has left IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de, Remote host closed the connection)|
|14:12||dgroos has left IRC (firstname.lastname@example.org, Read error: Connection reset by peer)|
|14:12||dgroos_ has joined IRC (email@example.com)|
|14:12||douglas_br has left IRC (firstname.lastname@example.org, Ping timeout: 260 seconds)|
|14:13||woernie has joined IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de)|
|14:40||gvy has left IRC (gvy!~mike@altlinux/developer/mike, Ping timeout: 264 seconds)|
|14:41||gvy has joined IRC (gvy!~mike@altlinux/developer/mike)|
|15:53||vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)|
|15:57||vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Read error: Connection reset by peer)|
|16:01||statler has joined IRC (statler!~Georg@gwrz3.lohn24.de)|
|16:01||vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)|
|16:05||kjackal has left IRC (kjackal!~quassel@2a02:587:3107:2e00:2562:6e6:6bf1:6b28, Ping timeout: 246 seconds)|
|16:11||kjackal has joined IRC (kjackal!~quassel@2a02:587:3107:2e00:c585:f21c:70ac:1f71)|
vagrantc or any other native English speaker, could you comment on this? https://github.com/ltsp/ltsp/issues/57#issuecomment-547989903
I.e. HOSTNAME_BASE/EXTRA, or HOSTNAME_PREFIX/SUFFIX ?
|16:23||gdi2k has left IRC (email@example.com, Read error: Connection reset by peer)|
(Not native, but I'll give my opinion anyway)
Prefix and suffix are definitely clearer than base and extra. You could probably keep the old ones working for backwards compatibility, but document the new ones.
no backwards compatibility!
alkisg: yeah, prefix/suffix makes more sense
Thank you uumas; but I didn't like the fact that there's no "middle" part
prefix/suffix is supposed to have a "body"
presuming prefix is always at the beginning and suffix is always at the end
Yes, but there's no parameter for "middle", it's only prefix and suffix
Well, the body is the automatically generated part
prefix is the "ltsp" part
and suffix is the ip or the mac
There are only 2 parts, not 3
it's a little linguistically odd to have a prefix+suffix without a middle ... but not tragically :)
OK so I keep the documentation and change the code, to use prefix/suffix :)
(the documentation says prefix/suffix while the code says base/extra :P)
even better, it's fixing a bug in the code!
|16:33||Leolo_2 has joined IRC (Leolo_2firstname.lastname@example.org)|
|16:38||gdi2k has joined IRC (email@example.com)|
alkisg: fwiw, i really prefer longer descriptive variable names to shorter ones that need to be deciphered ... seems like you've been leaning the other way
|16:39||gdi2k has left IRC (firstname.lastname@example.org, Client Quit)|
vagrantc: for example?
alkisg: kind of distracted right now, but i'll follow up
|16:41||douglas_br has joined IRC (email@example.com)|
whoah, you support SDDM :)
ah, this is what made me think of it: PWMERGE_SUR=, PWMERGE_SGR=, PWMERGE_DGR=, PWMERGE_DUR=
granted, those would get really long to unpack ... but it's pretty opaque at first glance
|16:46||Klimm has joined IRC (Klimm!~Georg@p5B30EE2F.dip0.t-ipconnect.de)|
alkisg: there might have been some other internal variables, though those aren't quite as big a deal
vagrantc: those are the parameters of pwmerge, which is a tool that might be useful outside of ltsp too
It's pwmerge --sur etc
And at that context, when you only have sur/sur/sgr/dgr etc, expanding them doesn't make much sense... too long
a tool external to ltsp?
E.g. pwmerge can be used to migrate the users of an old installation to a new one
I tried to make pwmerge and pamltsp as standalone as possible
So that people that need them can use them without ltsp
alkisg: so then my question about variable names translates into commandline options :)
True; I'd be glad to have better names for pwmerge parameters, but I couldn't find any
And --source-user-regex was just too large
(considering that all 4 of them were regexes)
not with tab completion :P
If you're willing to implement tab completion, sure
I don't have the time for it
I'm about to stop working in ltsp until the summer, to finish my phd :)
So I want to wrap things up; and continue in august to prepare 20.04 for schools
Large parameters also make code calls unreadable; it would need 3 lines instead of 1, to call pwmerge
OK ok like stan lee used to say, 'nuff said :)
So far I have 3 bug fixes to do tomorrow, and hopefully that'll be all for now
looks like i'll need to fix the download unverified binaries from the internet bug before pushing to unstable
That'd be most welcome
Personally I feel that bug reports to have them included in debian packages might be better; and I"m not worried about https security so much, as it's not directly runnable code, it goes to the clients via tftp and then the kernel is loaded and removes everything
there are more people between the attack on the internet than on the local network, even though the local network attack is easier
It's hard to make an "ipxe malware"
|17:06||Klimm has left IRC (Klimm!~Georg@p5B30EE2F.dip0.t-ipconnect.de, Remote host closed the connection)|
i don't follow; ipxe loads your kernel, initrd, etc. ... trivial to implement a very low-level exploit
"here, have this not-at-all-shady-kernel instead"
I think people would notice if they started downloading kernels from the internet; it needs 5 minutes to load them
And if they already have a man in the middle of https traffic, I'm not sure the other connections can be trusted either
It's already "s", secure; we have other areas much more in need of security
But if you wanna do it, of course I'd welcome it
yeah, i don't feel comfortable moving it into an actual release without something more secure
given that, well, i've seen enough script kiddy attacks on https
debian release, that is ...
fine for experimental
|17:12||dgroos_ has left IRC (firstname.lastname@example.org, Ping timeout: 240 seconds)|
re command line options: the general practice is to have long names be however long they need to be, then have short names that are a single letter when possible
if they are all regex, then having that in the param name is redundant. --source-user would be plenty, and not that long to have a var name
|17:15||dgroos has joined IRC (email@example.com)|
for https, as long as you're verifying the certs, that's plenty secure enough. you have to take over the client's cert authority first to spoof that
there are plenty of shady certificate authorities that are "trusted"
I assume it's a lot easier to spoof tftp
agreed, but the number of potential attackers on the local network is presumably much smaller than on the internet
the issue I see with it is that ltsp installs tend to be adhoc initially, for testing, and you'd likely only have self-signed, so either you need a cert authority on the server too that the clients can use, or you're left to checking them
people! please: celeron J 1800 can be good choice to ltsp client? My first experience with ltsp is being in this year. We have some schools that have and old info lab. I am trying to use the old pcs, some of them have a poor MB and too old celeron processor that we did change for pentium. so, now stay good experience. for new labs, we are searching
low price configuration.
||cw: or you implement independent verification, and just use https for transport
definitely getting the packages fixed in debian (and all the other distros) would be preferable ... but that coudl take quite some time
douglas_br: depends what you want to do on it. in most cases, it's the ram that's the bottleneck these days
|17:23||* vagrantc mind boggles at thinking of a pentium as anything other than an i586|
pentium 1 was i586. pentium 2 was i686. from then on it became a brand only, and not an architecture. intel just invested too much into it as a brand.
yes, i know the history :)
5ium i686 was silly then too
||cw at the first time, only for labs schools - education - games - youtube
huh. somehow I never made that connections.
--source-user would be plenty, and not that long to have a var name => but it means a different thing, now it's a user, not a regex
douglas_br: if you want a quick test, boot up a libe-usb and try those things. ltsp should do a little better than that if you have gigabit lan, or a really slow USB drive
it's a user specification, which can be a regex.
(07:18:39 PM) vagrantc: there are plenty of shady certificate authorities that are "trusted" => I'm not sure what an exploit would involve. Someone being your internet provider? Having the ltsp server connect via wifi? Even so, can someone pretent to be github.com?
common example: find's -name. it can take a name literal or a pattern. the parameter's name doesn't need to reflect this.
||cw: sure, but the context matters too. pwmerge is about merging users and groups; shouldn't names be more precise there?
a single specific name is also a valid regex...
but yes, distinguishing users and group would be important
douglas_br: for my schools here, I bought i3's, so that they can last for years
so --source would be too short.
A celeron would browse the web today, but what about tomorrow?
Usage: pwmerge [--sur=] [--sgr=] [--dur=] [--dgr=] [-l] [-q] sdir ddir mdir
Short names for all these wouldn't really make sense
-l --ltsp, and -q --quiet, sure, make sense
mdir is "merge dir"
-s -g -S -G ? not as clear, but this is a common practice
that's all I mean, there's common practice for this
alkisg you right
heh, there's even an IEEE document for it https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap12.html
||cw: oh sure, all the ltsp applets support both short and long options; it just was hard for this specific tool
|17:40||eu^205215175117 has left IRC (firstname.lastname@example.org, Remote host closed the connection)|
|17:52||statler has left IRC (statler!~Georg@gwrz3.lohn24.de, Ping timeout: 240 seconds)|
Hi @alkisg do I need to --purge or just remove tigervnc-viewer? I have x11vnc already installed, what was the other to install?
--purge just removes any system config files
dgroos: install xvnc4viewer first, purge tigervnc after that
Otherwise removing the package would also remove epoptes
|18:06||statler has joined IRC (statler!~Georg@gwrz.lohn24.de)|
OK, and rebuilding the client image now...
Just test booted a client, broadcast screen not working.
[be back in 55 min, another class starts in 12 seconds...
If you want to leave it open
I can have a look in 10'... or then
|18:33||statler has left IRC (statler!~Georg@gwrz.lohn24.de, Remote host closed the connection)|
hello! I have 2 NIC in server. Sometimes it "lost" DNS on dnsmasq, if I reset server, come back normal
douglas_br: are you sure it loses dns and not e.g. nat?
What are the symptoms?
sometimes status dnsmasq does not show ip dns, only internal. After reboot status dnsmasq show 20x.xxx.xxx
douglas_br: do you mean this command? systemd-resolve --status
Are you using network manager? Is your network connection "For all users"?
Do you mean this? systemctl status dnsmasq
alkisg you using network manager? Is your network connection "For all users"?
alkisgDo you mean this? systemctl status dnsmasq
douglas_br: dnsmasq isn't supposed to be updating its dns; it only happens once, when it starts
alkisg: just signed on to server, be back in a bit. Thanks for offering...
|19:36||dgroos has left IRC (email@example.com, Quit: dgroos)|
|19:39||kjackal_v2 has joined IRC (firstname.lastname@example.org)|
|19:39||kjackal has left IRC (kjackal!~quassel@2a02:587:3107:2e00:c585:f21c:70ac:1f71, Ping timeout: 276 seconds)|
dgroos, if you see irclogs; your clients are using wayland
This in ltsp.conf disables it
|19:44||dgroos has joined IRC (email@example.com)|
dgroos, if you see irclogs; your clients are using wayland (09:42:49 PM) alkisg: GDM3_CONF="WaylandEnable=false" (09:42:53 PM) alkisg: This in ltsp.conf disables it
Oh! didn't know that they were.
And your server too
Let me just fix your server as well
dgroos: so: log out, select "ubuntu" in login screen
Then log back in, and everything should now work
Will do, thought I had.
Awesome and Thanks! I'll check it out
|19:54||* alkisg thought that ubuntu 18.04 still defaulted to xorg, no idea what caused your wayland sessions...|
Rebooted and when I selected the gear icon, "Ubuntu" was already selected. I re-selected it and booted.
Ran commands: env | grep XDG and
cat /etc/X11/default-display-manager and they both indicate gdm (GNOME)
booted clients and same result as before.
|20:03||section1 has left IRC (firstname.lastname@example.org, Quit: Leaving)|
could it be that my computer is hooked up to display both a monitor and a promethean board?
reconnected w/sudo -i ready.
|20:42||woernie has left IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de, Remote host closed the connection)|
|20:44||Faith has left IRC (Faith!~Paty_@unaffiliated/faith, Quit: Leaving)|
Ah! when I tested broadcast on an un-logged-in client it didn't work, but it does work on a client that's logged in!
And see that while broadcast to an client not logged in doesn't work, I can view/assist them via vnc.
|21:01||douglas_br has left IRC (email@example.com, Remote host closed the connection)|
Do you still have any issues?
To see if wayland is running, I do `ps aux|grep -i wayland`
If there's Xwayland etc, then it's a wayland session
Both xorg and wayland use gdm, so checking /etc/X11/default-display-manager isn't appropriate
Thanks for that info!
The issue I'm trying to solve is that the clients can't ping 126.96.36.199 yet.
I'm about to head for bed; if you still have issues let's solve them quickly
That sounds related to NAT
Do you have NAT=1 in ltsp.conf?
it's commented out...
OK enable it and run `sudo systemctl restart ltsp`
:-) will do!
And thanks again :-D
And it works :-) G'night!
np; good night all
|22:16||kjackal_v2 has left IRC (firstname.lastname@example.org, Ping timeout: 264 seconds)|
|23:03||dgroos has left IRC (email@example.com, Quit: dgroos)|
|23:04||ricotz_ has joined IRC (ricotz_!~ricotz@ubuntu/member/ricotz)|
|23:08||ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Ping timeout: 265 seconds)|