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


Channel log from 29 January 2020   (all times are UTC)

01:56sutula has left IRC (sutula!~sutula@184.100.153.5, Ping timeout: 268 seconds)
02:02sutula has joined IRC (sutula!~sutula@184.100.153.5)
02:08sutula has left IRC (sutula!~sutula@184.100.153.5, Read error: Connection reset by peer)
02:08sutula_ has joined IRC (sutula_!~sutula@184.100.153.5)
02:11sutula_ has left IRC (sutula_!~sutula@184.100.153.5, Read error: Connection reset by peer)
02:13sutula has joined IRC (sutula!~sutula@184.97.60.93)
05:09quinox has left IRC (quinox!~quinox@ghost.qtea.nl, Quit: WeeChat 2.6)
05:12quinox has joined IRC (quinox!~quinox@ghost.qtea.nl)
06:01statler has joined IRC (statler!~Georg@p5B30E4C3.dip0.t-ipconnect.de)
06:05kjackal has joined IRC (kjackal!~quassel@2a02:587:3115:3000:11c3:b703:205a:7ff4)
06:50aleppik has joined IRC (aleppik!b99d6c92@185-157-108-146.cust.elevi.ee)
06:52
<alkisg>
Hi aleppik, I'll be back in 30 mins, if you can come back then we can see the issue then
06:53
<aleppik>
OK
07:19
<alkisg>
aleppik: back
07:19
!vnc-edide
07:19
<ltspbot>
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
07:21
<aleppik>
clients with names starting ws are guests
07:24
<alkisg>
aleppik: so, the problem is on ltsp1232 and ltsp1495, right?
07:24
<aleppik>
No, I can broadcast to these computers
07:25
<alkisg>
Where is the problem?
07:25
On wc*?
07:25
*ws?
07:25
<aleppik>
Yes, on WS
07:25
logged in as guests
07:26
Please, try on ws04
07:26
<alkisg>
ok sorry
07:26
I asked in the terminal where to test but you didn't see it
07:32
aleppik: the problem is that the client DISPLAY/XAUTHORITY isn't working
07:32
Looking why...
07:43
aleppik: can you put the guest password in ws04?
07:47
aleppik: so... it seems that setting the correct DISPLAY and XAUTHORITY for user root isn't working when home is nfs
07:47
Which is strange, because it's working fine here on 18.04 and 20.04
07:47
<aleppik>
Strange indeed
07:48
<alkisg>
If you kill the root process of the epoptes-client, you'll be able to broadcast to user screens
07:48
But then you'll lose some other features that the root epoptes-client provides
07:48
I'll note it in the bug report, and see if I can find time to troubleshoot it, but it'll probably be much later (months...)
07:49
<aleppik>
Now I have somethink to think about myself
07:49
I can try the same setup on 20.04
07:50
or 18.04
07:51
<alkisg>
I wonder if snap/apparmor is involved
07:51
We uninstall it here immediately
07:51
(it imposes NFS restrictions etc)
07:51
<aleppik>
I'll try that
07:52
<alkisg>
OK, good luck. I'm available for hire if needed, but it's not an immediate concern for the "free support" of the "open source development" part :)
07:52
I.e. most people don't have that issue
07:53
<aleppik>
Thank you
07:53
<alkisg>
You're welcome
07:54statler has left IRC (statler!~Georg@p5B30E4C3.dip0.t-ipconnect.de, Remote host closed the connection)
08:12woernie has joined IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de)
08:13
<alkisg>
Heh, that eu^ user yesterday was right; pam unlocking works fine on ubuntu mate 16.04, but not on ubuntu gnome 16.04
08:13
*screensaver unlocking
08:21
Ah OK it's a known issue, happens with LDAP too: https://bugs.launchpad.net/ubuntu/+source/nss-pam-ldapd/+bug/1314095
08:39statler has joined IRC (statler!~Georg@mail.lohn24.de)
09:43
<alkisg>
It only affects unity though, not 18.04+, so ...let's not care about it :)
10:16eu^gw-2s00infcsd has joined IRC (eu^gw-2s00infcsd!c25f4201@gw-2s00.infcs.de)
10:17
<eu^gw-2s00infcsd>
Hi
10:17
<alkisg>
mhwasil is that you?
10:17
<eu^gw-2s00infcsd>
yes
10:17
<alkisg>
Hi, type this to set your name here:
10:17
/nick mhwasil
10:17eu^gw-2s00infcsd is now known as mhwasil
10:17
<mhwasil>
cool
10:17
<alkisg>
Nice, now do the following command: x11vnc -connect srv1-dide.ioa.sch.gr
10:17
If you don't have x11vnc, apt install it first
10:19
<mhwasil>
it seems it failed to open xorg
10:19
29/01/2020 11:18:28 x11vnc version: 0.9.13 lastmod: 2011-08-10 pid: 307229/01/2020 11:18:28 XOpenDisplay("") failed.29/01/2020 11:18:28 Trying again with XAUTHLOCALHOSTNAME=localhost ...29/01/2020 11:18:28 29/01/2020 11:18:28 *** XOpenDisplay failed. No -display or DISPLAY.29/01/2020 11:18:28 *** Trying ":0" in 4 seconds. Press Ctrl-C to
10:19
abort.29/01/2020 11:18:28 *** 1 2 3 4 No protocol specified29/01/2020 11:18:32 XOpenDisplay(":0") failed.29/01/2020 11:18:32 Trying again with XAUTHLOCALHOSTNAME=localhost ...No protocol specified29/01/2020 11:18:32 XOpenDisplay(":0") failed.29/01/2020 11:18:32 Trying again with unset XAUTHLOCALHOSTNAME ...No protocol specified29/01/2020 11:18:32
10:19
29/01/2020 11:18:32 ***************************************29/01/2020 11:18:32 *** XOpenDisplay failed (:0)
10:19
<alkisg>
!paste
10:19
<ltspbot>
paste: To avoid channel flooding, please upload text longer than 3 lines to http://paste.debian.net. Don't forget to paste the resulting URL here.
10:19
<alkisg>
Is xorg running/
10:19
?
10:20
Don't run that command as root, run it as user, inside a xorg session
10:20
<mhwasil>
owh, it was root
10:20
<alkisg>
Also, it won't work if you're using wayland
10:21
<mhwasil>
are you asking for password?
10:21
<alkisg>
Don't tell it to me, just write it there and press enter
10:21
So that we're able to run ltsp image etc
10:25
mhwasil: it works for me, how are you usually using root?
10:25
sudo su -l ?
10:25
<mhwasil>
yes
10:28kvaps has joined IRC (kvaps!2e1c6842@wedos.wedos.net)
10:33
<alkisg>
mhwasil: you have a weird /usr/sbin/partprobe which isn't part of any debian package
10:33
And your /sbin/partprobe was marked non-executable
10:34
Do you know anything about these?
10:34
<mhwasil>
No
10:36
<alkisg>
mhwasil: OK so the problem is that losetup -P works, partprobe only works after 1 second (!), and losetup -P isn't available in busybox
10:37
<mhwasil>
so it works with su right?
10:37
<alkisg>
No, su was unrelated
10:37
The problem is a "race condition"
10:37
<mhwasil>
I really have no idea about this
10:37
<alkisg>
it sometimes work, sometimes not, depending on if partprobe returns quickly (=wrong) or 0.0001 sec later
10:37
<mhwasil>
thatÅ› why you put delay in there
10:38
<alkisg>
Yes, but now it's not needed, I found a better way with losetup -P
10:38
But
10:38
<mhwasil>
is it because of the debian distribution?
10:38
<alkisg>
I think it's because you have a very fast drive
10:38
Somehow partprobe returns before the result is ready
10:39
<mhwasil>
does it relate to intel VT?
10:39
I actually activate intel vt in the bios
10:39
<alkisg>
No I don't think so
10:39
Now, I need this from you:
10:39
I will need some time to think and test locally in my pc, busybox etc
10:39
After about half an hour, I'll have a solution and I'll want to test it in your pc
10:39
So, can we continue in half an hour?
10:39
<mhwasil>
okay, no problem
10:40
i will keep the chat open
10:40
<alkisg>
OK talk to you then
10:42kvaps has left IRC (kvaps!2e1c6842@wedos.wedos.net, Remote host closed the connection)
10:42GodFather has left IRC (GodFather!~rcc@wsip-66-210-242-210.ph.ph.cox.net, Read error: Connection reset by peer)
10:43GodFather has joined IRC (GodFather!~rcc@wsip-66-210-242-210.ph.ph.cox.net)
11:03
<alkisg>
mhwasil: here is the explanation: https://github.com/ltsp/ltsp/issues/112
11:03
Now I will start coding the solution...
11:04
https://github.com/ltsp/ltsp/issues/112#issuecomment-579704835
11:06
mhwasil: it will take me more time than I was planning. Do not wait for me. I will upload a new version in the proposed PPA, and I will ask you in the bug report to test it
11:06
<mhwasil>
(y)
11:06
<alkisg>
:)
11:06
mwalters: ah, is it working for you now there?
11:06
So that you can proceed for now?
11:07
Run x11vnc -connect srv1-dide.ioa.sch.gr
11:07
So that I see what I left in the code there
11:07
<mhwasil>
yes, it seems the arg -P works
11:07
<alkisg>
OK, we don't need a sleep nor an "echo" there
11:08
mhwasil: nice, use that one until I have the new version
11:13
<mhwasil>
thank you very much for the support
11:14mhwasil has left IRC (mhwasil!c25f4201@gw-2s00.infcs.de)
11:14mhwasil has joined IRC (mhwasil!c25f4201@gw-2s00.infcs.de)
11:14mhwasil has left IRC (mhwasil!c25f4201@gw-2s00.infcs.de, Remote host closed the connection)
11:20aleppik has left IRC (aleppik!b99d6c92@185-157-108-146.cust.elevi.ee, Remote host closed the connection)
11:38section1 has joined IRC (section1!~section1@178.33.109.106)
11:48Faith has joined IRC (Faith!~Paty_@unaffiliated/faith)
12:33shored has left IRC (shored!~shored@87-92-77-163.bb.dnainternet.fi, Read error: Connection reset by peer)
12:35mgariepy has joined IRC (mgariepy!~mgariepy@ubuntu/member/mgariepy)
12:36shored has joined IRC (shored!~shored@87-92-77-163.bb.dnainternet.fi)
13:25
<mwalters>
18.04/ltsp5 is still working just fine. Thanks for asking ;)
13:33theedude has left IRC (theedude!d8c205e5@as5300-1.216-194-5-229.nyc.ny.metconnect.net, Remote host closed the connection)
14:15theedude has joined IRC (theedude!~theedude@as5300-1.216-194-5-229.nyc.ny.metconnect.net)
14:38
<theedude>
I'm new to this and honestly might be over my head, but I'm going to try and setup a lab. Anyway i was reading documentation and saw that thin client is no longer supported and would have to use vnc or other remote solution. I was hoping to use some Diskless Workstations 1720 that I got from work. These are the type I will not be able to use, correct?
14:39
<quinox>
diskless is still fine
14:39
thin clients = the GUI runs on the server, this is going the way of the dodo
14:40
you'd want thin clients when your workstations are reallly low-spec
14:41
(I have no idea what kind of spec a 1720 has)
14:41
<theedude>
oh ok. thats a relief. I got 4 of these diskless and really want to experiment with them. Thank you so much.
14:43
Intel Atom, Linux Terminal Server Project (LTSP) - Version 5.x Ready, Dual Monitor Capability (VGA + DVI), 2GB DDRII Memory
14:43
thats from their site
14:46
<quinox>
2GB ram is cutting it close nowadays, websites eat up a lot of memory
14:49
http://store.disklessworkstations.com/200123.html
14:49
I didn't even know they branded machines specifically for LTSP
14:51
<alkisg>
quinox: the guys at disklessworkstations.com were part of the initial ltsp team back in 2000, they formed that company for ltsp
14:52
They still sell diskless workstations, but they haven't written ltsp code in more than a decade...
14:52
<elias_a>
Nice :)
15:34GodFather has left IRC (GodFather!~rcc@wsip-66-210-242-210.ph.ph.cox.net, Ping timeout: 268 seconds)
15:38Ark74 has left IRC (Ark74!~Luis@177.238.145.140, Read error: Connection reset by peer)
15:39Ark74 has joined IRC (Ark74!~Luis@177.238.145.140)
15:42gdi2k has joined IRC (gdi2k!~gdi2k@host86-185-148-176.range86-185.btcentralplus.com)
18:42section1 has left IRC (section1!~section1@178.33.109.106, Quit: Leaving)
18:47kjackal has left IRC (kjackal!~quassel@2a02:587:3115:3000:11c3:b703:205a:7ff4, Remote host closed the connection)
18:49kjackal has joined IRC (kjackal!~quassel@2a02:587:3115:3000:e5a1:8ff8:992d:2899)
19:04Ark74 has left IRC (Ark74!~Luis@177.238.145.140, Read error: Connection reset by peer)
19:04kjackal has left IRC (kjackal!~quassel@2a02:587:3115:3000:e5a1:8ff8:992d:2899, Ping timeout: 246 seconds)
19:14vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
19:18kjackal has joined IRC (kjackal!~quassel@2a02:587:3115:3000:e5a1:8ff8:992d:2899)
19:36statler has left IRC (statler!~Georg@mail.lohn24.de, Remote host closed the connection)
20:30
<vagrantc>
alkisg: with the breakage for https://github.com/ltsp/ltsp/issues/43 ... should i hold off on uploading to Debian?
20:31
<alkisg>
vagrantc: yes please there's one more thing I need to push tomorrow
20:31
Eeeh actually it's not related to #43 anymore
20:31
<vagrantc>
new issue?
20:32
<alkisg>
https://github.com/ltsp/ltsp/issues/112#issuecomment-579704835
20:32
<vagrantc>
i can give the current code some testing regardless and see if i don't find anything else :)
20:32
<alkisg>
A partprobe bug in nvme drives
20:32
I want to find a workaround there
20:32
<vagrantc>
right, saw that too :)
20:33
<alkisg>
That's all that's left in my todo list for the release, but I'm ok without it as well
20:33
It can go in a future release, I don't mind
20:33
So if you have time today, go ahead and release, I tested a lot today
20:33
Booted all my .isos and VMs etc
20:33
<vagrantc>
the .iso/VM breakage was the most alarming to me
20:33
or at least the biggest regression
20:34
<alkisg>
The iso breakage was related, there wasn't any vm breakage though
20:34
<vagrantc>
wait, nvme ... on a network booted machine? :)
20:34
<alkisg>
This was a new issue
20:34
No, partprobe on an ltsp server returns before the probe finishes
20:34
So /dev/loop0p1 isn't ready yet
20:34
<vagrantc>
ah!
20:35
<alkisg>
I'll work around that using losetup -P instead, but it's not always available, so I'll need a big case
20:35
I'll have that ready tomorrow, if you want to test today and release tomorrow
20:35
<vagrantc>
will see how it goes :)
20:35
<alkisg>
;)
20:37Faith has left IRC (Faith!~Paty_@unaffiliated/faith, Quit: Leaving)
20:44
<vagrantc>
alkisg: would it break your workflow if you used UNRELEASED in debian/changelog instead of unstable ?
20:44
alkisg: until, of course, a release is made?
20:45
<alkisg>
Not at all; I only put unstable there for a smaller diff, e.g. in case we just release without changes in changelog
20:45
I'll do that next times then
20:45
<vagrantc>
the tooling around package building and various qa pages at debian work a little better that way
20:46
<alkisg>
Sure, np at all
20:46
<vagrantc>
thanks :)
20:46
none of which is a deal-breaker, but if it's no trouble to you :)
20:47
<alkisg>
So, I update the changelog and do ppa builds with UNRELEASED, and when you're ready to upload to debian, you edit changelog and change that to "unstable", right?
20:47
<vagrantc>
basically
20:48
<alkisg>
Nice, will do
20:48
<vagrantc>
i've often enough had various beurocratic changes anyways that i need to touch debian/changelog before a release anyways so it shouldn't add needless commits too often
20:52
<alkisg>
In many cases I don't even update the changelog when sending PPA builds, now I wanted LTSP 20.x to come as early as possible to stop people from calling it "ltsp19" like the version is part of the name... :D
20:52
Anyways, time to head to bed. /me waves :)
20:54* vagrantc waves
20:54
<vagrantc>
yeah, i generally only update the changelog as part of the release, but figured that was at least part of what you were doing :)
21:07kjackal has left IRC (kjackal!~quassel@2a02:587:3115:3000:e5a1:8ff8:992d:2899, Ping timeout: 272 seconds)
21:48woernie has left IRC (woernie!~werner@p57A0E783.dip0.t-ipconnect.de, Remote host closed the connection)
22:05
<vagrantc>
alkisg: didn't stress it too difficult, but my previous installation seems to continue to work just fine.
22:36
!init
22:36
<ltspbot>
I do not know about 'init', but I do know about these similar topics: 'INIT_COMMAND_EPOPTES', 'INIT_COMMAND_GUEST', 'INIT_COMMAND_SSH', 'lowram-initramfs', 'ltsp-initrd'
22:36
<vagrantc>
!ltsp-initrd
22:36
<ltspbot>
ltsp-initrd: Create the ltsp.img initrd add-on: https://github.com/ltsp/ltsp/blob/master/docs/ltsp-initrd.8.md
23:11* vagrantc struggles to figure out how to select display manager with ltsp.conf values
23:11
<vagrantc>
i though simply adjusting /etc/X11/default-display-manager would work, but apparently that's not enough with systemd
23:14
looks like i also have to adjust /etc/systemd/system/display-manager.service
23:40
POST_INIT_DISPLAY_MANAGER="echo /usr/sbin/gdm3 > /etc/X11/default-display-manager"
23:40
POST_INIT_DISPLAY_MANAGER_SERVICE="ln -svf /lib/systemd/system/gdm3.service /etc/systemd/system/display-manager\
23:40
.service"
23:40
that did the trick for me
23:40* vagrantc should probably learn how to update the wiki
23:40
<vagrantc>
it's a pretty niche use ... but it demonstrates an example of how to do use some ltsp features