03:44 | muvlon has left IRC (muvlon!~muvlon@x2f7f0e2.dyn.telefonica.de, Ping timeout: 240 seconds) | |
03:52 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
04:06 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Ping timeout: 240 seconds) | |
05:00 | Statler has joined IRC (Statler!~Georg@p579FEA10.dip0.t-ipconnect.de) | |
05:18 | alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Ping timeout: 252 seconds) | |
05:20 | alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg) | |
06:09 | ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz) | |
06:11 | mikkel has joined IRC (mikkel!~mikkel@mail.dlvs.dk) | |
06:25 | Eric5 has joined IRC (Eric5!~eric@sdi.iut-valence.fr) | |
07:36 | Statler has left IRC (Statler!~Georg@p579FEA10.dip0.t-ipconnect.de, Remote host closed the connection) | |
07:54 | HTTP_____GK1wmSU has joined IRC (HTTP_____GK1wmSU!~DEEP-BOOK@103.27.125.238) | |
07:57 | bcg has left IRC (bcg!b@dsl-tkubng11-54f942-246.dhcp.inet.fi, Quit: leaving) | |
08:00 | bcg has joined IRC (bcg!b@dsl-tkubng11-54f942-246.dhcp.inet.fi) | |
08:02 | Statler has joined IRC (Statler!~Georg@p579FE0B2.dip0.t-ipconnect.de) | |
08:54 | vlt has left IRC (vlt!~vlt@lvps87-230-93-209.dedicated.hosteurope.de, Ping timeout: 260 seconds) | |
08:54 | vlt has joined IRC (vlt!~vlt@2a01:488:66:1000:57e6:5dd1:0:1) | |
09:07 | Eric5 has left IRC (Eric5!~eric@sdi.iut-valence.fr, Ping timeout: 248 seconds) | |
09:11 | Eric5 has joined IRC (Eric5!~eric@sdi.iut-valence.fr) | |
09:20 | eu^107254163 has joined IRC (eu^107254163!53d420e4@gateway/web/freenode/ip.83.212.32.228) | |
09:30 | kjackal__ has joined IRC (kjackal__!~quassel@2a02:587:3125:7e00:9ce2:fc81:c406:c457) | |
09:30 | kjackal_ has left IRC (kjackal_!~quassel@2a02:587:310f:3a00:44e8:9c3a:e4c0:4c29, Ping timeout: 264 seconds) | |
09:46 | kjackal__ has left IRC (kjackal__!~quassel@2a02:587:3125:7e00:9ce2:fc81:c406:c457, Ping timeout: 252 seconds) | |
09:47 | kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:3114:bf00:9ce2:fc81:c406:c457) | |
10:01 | kjackal_ has left IRC (kjackal_!~quassel@2a02:587:3114:bf00:9ce2:fc81:c406:c457, Ping timeout: 255 seconds) | |
10:01 | kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:3111:be00:9ce2:fc81:c406:c457) | |
10:47 | kjackal__ has joined IRC (kjackal__!~quassel@2a02:587:3114:6a00:9ce2:fc81:c406:c457) | |
10:47 | kjackal_ has left IRC (kjackal_!~quassel@2a02:587:3111:be00:9ce2:fc81:c406:c457, Ping timeout: 264 seconds) | |
10:54 | kjackal__ has left IRC (kjackal__!~quassel@2a02:587:3114:6a00:9ce2:fc81:c406:c457, Ping timeout: 255 seconds) | |
10:54 | kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:3110:9a00:9ce2:fc81:c406:c457) | |
11:00 | kjackal_ has left IRC (kjackal_!~quassel@2a02:587:3110:9a00:9ce2:fc81:c406:c457, Ping timeout: 255 seconds) | |
11:01 | kjackal_ has joined IRC (kjackal_!~quassel@2a02:587:3118:2b00:9ce2:fc81:c406:c457) | |
11:25 | lucascastro has joined IRC (lucascastro!~lucas@170.231.131.95.lunarnet.com.br) | |
11:53 | lucascastro has left IRC (lucascastro!~lucas@170.231.131.95.lunarnet.com.br, Remote host closed the connection) | |
12:23 | <Hyperbyte> Hello. :-) Is there some easy way in LTSP to make sure no programs stay running in the background after a user logs out? Currently I have a terminal server that's bogged down with LibreOffice processes consuming 100% CPU that are started by users that have since logged out
| |
12:45 | <vlt> Hyperbyte: I know that problem well but we never found an LTSP built-in solution.
| |
12:53 | muvlon has joined IRC (muvlon!~muvlon@2001-4dd7-2465-0-ccd2-72bd-aa2a-cb2f.ipv6dyn.netcologne.de) | |
12:55 | lucascastro has joined IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br) | |
13:00 | <alkisg> Hyperbyte: thin or fat clients?
| |
13:00 | Also, which distro/version? ubuntu 16.04?
| |
13:12 | lucascastro has left IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br, Remote host closed the connection) | |
13:17 | GodFather has joined IRC (GodFather!~rcc@96-92-43-9-static.hfc.comcastbusiness.net) | |
13:21 | <Hyperbyte> alkisg, thin. 14.04, migrating to 16.04.
| |
13:28 | Eric5 has left IRC (Eric5!~eric@sdi.iut-valence.fr, Ping timeout: 260 seconds) | |
13:33 | <alkisg> Hyperbyte: 16.04 uses systemd and manages processes better, so it should work fine for thin clients too
| |
13:33 | (14.04 already worked fine for fat clients)
| |
13:34 | (systemd-logind, for user sessions)
| |
13:34 | Well, at least MATE, not sure about Unity
| |
13:42 | muvlon has left IRC (muvlon!~muvlon@2001-4dd7-2465-0-ccd2-72bd-aa2a-cb2f.ipv6dyn.netcologne.de, Ping timeout: 246 seconds) | |
13:45 | mikkel has left IRC (mikkel!~mikkel@mail.dlvs.dk, Quit: Leaving) | |
13:51 | ben_roose has joined IRC (ben_roose!~roose@roose.cs.wichita.edu) | |
13:55 | lucascastro has joined IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br) | |
13:59 | <Hyperbyte> So you're saying this problem will disappear with Ubuntu 16.04? :-)
| |
14:03 | <alkisg> Yup, although I don't really test thin clients a lot
| |
14:03 | Hyperbyte: which desktop environment?
| |
14:34 | muvlon has joined IRC (muvlon!~muvlon@2001-4dd7-2465-0-ccd2-72bd-aa2a-cb2f.ipv6dyn.netcologne.de) | |
14:44 | <Hyperbyte> Mate.
| |
14:45 | marsairforce has joined IRC (marsairforce!~marsairfo@198.41.29.45) | |
14:56 | <alkisg> Should be fine then
| |
15:18 | Eric5 has joined IRC (Eric5!~eric@sdi.iut-valence.fr) | |
15:27 | GodFather has left IRC (GodFather!~rcc@96-92-43-9-static.hfc.comcastbusiness.net, Ping timeout: 255 seconds) | |
15:29 | marsairforce has left IRC (marsairforce!~marsairfo@198.41.29.45, Quit: Leaving) | |
15:32 | <bennabiy> alkisg: Hyperbyte : I have noticed this issue still with 16.04 and thin clients, but it is only with certain apps. It is not uncommon to have at least a small handful of processes run by logged out users
| |
15:32 | MATE
| |
15:41 | marsairforce has joined IRC (marsairforce!~marsairfo@198.41.29.45) | |
15:44 | marsairforce has left IRC (marsairforce!~marsairfo@198.41.29.45, Remote host closed the connection) | |
15:51 | GodFather has joined IRC (GodFather!~rcc@75-145-237-204-Michigan.hfc.comcastbusiness.net) | |
16:16 | vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc) | |
16:33 | muvlon has left IRC (muvlon!~muvlon@2001-4dd7-2465-0-ccd2-72bd-aa2a-cb2f.ipv6dyn.netcologne.de, Ping timeout: 255 seconds) | |
16:43 | Statler has left IRC (Statler!~Georg@p579FE0B2.dip0.t-ipconnect.de, Remote host closed the connection) | |
17:24 | lucascastro has left IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br, Remote host closed the connection) | |
17:38 | muvlon has joined IRC (muvlon!~muvlon@2001-4dd0-af15-5b65-60dc-a224-9f16-9c92.ipv6dyn.netcologne.de) | |
17:40 | Statler has joined IRC (Statler!~Georg@p579FEA10.dip0.t-ipconnect.de) | |
17:42 | <vagrantc> w8
| |
17:44 | Eric5 has left IRC (Eric5!~eric@sdi.iut-valence.fr, Ping timeout: 248 seconds) | |
17:57 | <alkisg> bennabiy: I thought systemd-logind was using cgroups to track the processes... it'll be worth checking if those processes do indeed belong to the same cgroup, and if there's some systemd option to "kill all processes on logout", which for some reason isn't on by default...
| |
17:58 | In my experience, after logout, I see processes for 5-10 seconds, and then they're cleaned up automatically
| |
18:01 | kjackal_ has left IRC (kjackal_!~quassel@2a02:587:3118:2b00:9ce2:fc81:c406:c457, Remote host closed the connection) | |
18:10 | muvlon has left IRC (muvlon!~muvlon@2001-4dd0-af15-5b65-60dc-a224-9f16-9c92.ipv6dyn.netcologne.de, Ping timeout: 252 seconds) | |
18:23 | lucascastro has joined IRC (lucascastro!~lucas@170.78.53.20) | |
18:53 | muvlon has joined IRC (muvlon!~muvlon@2001-4dd0-af15-5b65-38f2-7212-872c-ee95.ipv6dyn.netcologne.de) | |
18:57 | GodFather has left IRC (GodFather!~rcc@75-145-237-204-Michigan.hfc.comcastbusiness.net, Ping timeout: 248 seconds) | |
18:58 | lucascastro has left IRC (lucascastro!~lucas@170.78.53.20, Remote host closed the connection) | |
19:13 | <bennabiy> I will check on it later... not much lab use right now
| |
19:13 | lucascastro has joined IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br) | |
19:21 | Statler has left IRC (Statler!~Georg@p579FEA10.dip0.t-ipconnect.de, Remote host closed the connection) | |
19:35 | lucascastro has left IRC (lucascastro!~lucas@189.90.38.210.jupiter.com.br, Remote host closed the connection) | |
20:37 | GodFather has joined IRC (GodFather!~rcc@47.33.250.142) | |
20:54 | angel_ has left IRC (angel_!bdd55dab@gateway/web/freenode/ip.189.213.93.171, Ping timeout: 260 seconds) | |
21:28 | ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving) | |
21:34 | dgroos has joined IRC (dgroos!~dagro001@205.215.175.117) | |
21:45 | GodFather has left IRC (GodFather!~rcc@47.33.250.142, Ping timeout: 248 seconds) | |
21:48 | dgroos has left IRC (dgroos!~dagro001@205.215.175.117, Ping timeout: 240 seconds) | |
22:05 | ben_roose has left IRC (ben_roose!~roose@roose.cs.wichita.edu, Remote host closed the connection) | |
22:05 | dgroos has joined IRC (dgroos!~dagro001@205.215.175.117) | |
22:07 | <dgroos> Hi All, Still working on this. My ltsp-pnp fat clients work great (with local users, and with internet access) until I join my server to the district’s AD authentication system. Then I can’t log in as a local user (like I could before) nor as an AD user.
| |
22:11 | I did a ‘before and after’ test to see how joining the AD affected: getent passwd LOCAL USER & getent group LOCAL USER. I tried those commands both before and after joining AD, and got the same results in syslog.
| |
22:12 | GodFather has joined IRC (GodFather!~rcc@47.33.250.142) | |
22:29 | <dgroos> I plugged my laptop into the LAN along with the rest of the clients.
| |
22:32 | Now on my laptop I’m able to ssh into the ltsp server as a local users… and after joining the server to the AD like before, I’m able to ssh into server both as local and AD users!
| |
22:33 | i.e. ssh into server as AD user works, but login as AD user from a fat client doesn’t work. Any ideas things to check?
| |
22:50 | <vagrantc> dgroos: are you using ldm or some other display manager?
| |
22:50 | <dgroos> I assume ldm. Using Unity
| |
22:51 | <vagrantc> you've been fighting with this long enough i would recommend not making assumptions anymore :P
| |
22:51 | <dgroos> :-)
| |
22:52 | vagrantc: not always sure what i’m guessing and what I’m knowing and what I’m blindly assuming!
| |
22:52 | I’ll check…
| |
22:55 | I typed: echo $DESKTOP_SESSION and it says ubuntu
| |
22:56 | but oops… let me keep digging…
| |
22:59 | vagrantc: cat /etc/X11/default-display-manager returned:
| |
23:03 | <vagrantc> it really returned nothing?
| |
23:03 | <dgroos> It’s not wanting to display what it returned. Let me try again: /usr/sbin/lightdm
| |
23:03 | <vagrantc> sounds like you're not using LDM
| |
23:03 | <dgroos> wouldn’t print on it’s own line, sorry
| |
23:03 | <vagrantc> fat client?
| |
23:03 | <dgroos> yes
| |
23:04 | <vagrantc> ok, that explains a lot
| |
23:04 | pgrep -l -a -f lightdm
| |
23:04 | or systemctl status lightdm
| |
23:04 | does it show processes running?
| |
23:05 | in any case, what's probably happening is lightdm is configured to use LDAP, AD, whatever, etc. and partially working
| |
23:05 | <dgroos> the first command showed 3 processes: /usr/sbin/lightdm
| |
23:05 | and 2 others
| |
23:05 | <vagrantc> and pgrep -l -a -f ldm
| |
23:06 | <dgroos> didn’t return anything
| |
23:06 | <vagrantc> ok, we've now identified the issue; you're not running a display manager that knows how to LTSP :)
| |
23:06 | you're kind of half-way into our future plans, though ... :)
| |
23:07 | <dgroos> ah… the ltsp 6 holy grail!
| |
23:07 | <vagrantc> so, something isn't triggering on boot that should stop lightdm from running
| |
23:08 | <dgroos> but why does everthing work well with fat clients logging in when not joined to the AD?
| |
23:08 | <vagrantc> e.g. /usr/share/ltsp/init-ltsp.d/*
| |
23:08 | <muvlon> hey, quick question
| |
23:08 | what is ltsp-pnp and do I want that?
| |
23:08 | <vagrantc> dgroos: did you use --cleanup when building the image?
| |
23:08 | <dgroos> yes
| |
23:08 | <vagrantc> dgroos: sounds suspiciously like you haven't
| |
23:09 | muvlon: it allows you to build a network bootable LTSP environment out of the installed server's OS
| |
23:09 | muvlon: and yes, it's the recommended way to try out LTSP these days
| |
23:09 | <dgroos> I just checked and the last command was sudo ltsp-update-image —cleanup /
| |
23:10 | <vagrantc> sudo mount -o loop /opt/ltsp/images/i386.img /mnt ; grep SOMEUSER /mnt/etc/passwd
| |
23:10 | (or amd64.img)
| |
23:10 | where SOMEUSER is some user on the server.
| |
23:11 | or wc -l /etc/passwd /mnt/etc/passwd
| |
23:12 | <dgroos> I’ve got a test user (standard) so did: sudo mount -o loop /opt/ltsp/images/amd64.img /mnt ; grep addit /mnt/etc/passwd
| |
23:12 | and it returned nothing.
| |
23:12 | <vagrantc> and grep addit /etc/passwd returns a line for that user?
| |
23:13 | <dgroos> yes
| |
23:13 | <muvlon> vagrantc, any downsides to using ltsp-pnp?
| |
23:13 | <vagrantc> muvlon: server and client architecture must match
| |
23:13 | <muvlon> ah
| |
23:13 | <vagrantc> muvlon: users have login access to the server, effectively
| |
23:14 | dgroos: well, i don't know how lightdm could possibly authenticate as a user that isn't part of the system
| |
23:14 | <muvlon> so the server must be safe against privilege escalation?
| |
23:14 | <vagrantc> dgroos: unless you're LDAP/AD configuration somehow cached the local users in some alternate location
| |
23:14 | muvlon: sure
| |
23:15 | dgroos: you must have some scary magic, or found a very scary bug in lightdm
| |
23:15 | <muvlon> up-to-date Debian should be reasonably secure
| |
23:16 | <vagrantc> muvlon: although, the default setup for classic LTSP had the same issue
| |
23:17 | <dgroos> vagrantc: must be a local cache then since I can see the AD users that I’ve authenticated to while sitting at the server, and they have home folders and all, but right, they aren’t part of /etc/passwd.
| |
23:18 | <muvlon> vagrantc: can I do ltsp-pnp without a separate DHCP server?
| |
23:18 | <vagrantc> dgroos: i know this may sound like a lot of work, but have you tried setting this up without AD just to see that it all works, and then incrementally add AD until it breaks?
| |
23:18 | muvlon: sure
| |
23:18 | <muvlon> i.e. can I tell dnsmasq to be a full DHCP server instead of DHCP-proxy
| |
23:18 | <vagrantc> muvlon: yes.
| |
23:19 | <dgroos> Well I think I’ve not been detailed enough in my description.
| |
23:19 | <muvlon> ah, found it
| |
23:20 | <dgroos> If I’m joined to the AD, I can’t log into fat clients either as local or AD user. BUT…
| |
23:20 | <vagrantc> dgroos: your first problem is that LDM isn't running at all
| |
23:21 | <dgroos> If I unjoin from AD, then without reboot or anything at the fat clients, I can log in at the clients! Internet works and my system is just great.
| |
23:21 | So the break point is joining my server to the AD.
| |
23:22 | <vagrantc> well, the problem is you're running the wrong display manager for LTSP
| |
23:23 | dgroos: it might have something to do with the fact that your clients all have cached AD credentials of the server, and once the server joins the AD domain, AD refuses connections from other machines with the same credentials
| |
23:23 | dgroos: but you're not using LTSP at this point, effectively
| |
23:23 | dgroos: as lightdm has no ltsp-specific hooks or code
| |
23:23 | <dgroos> OK, I followed the instructions on Alkis’ ltsp-pnp install page (which is now outdated) and I also tried adding ltsp-manager.
| |
23:24 | <vagrantc> dgroos: just to make sure, you ran the commands to check for a running lightdm on a fat client, not on the server?
| |
23:24 | <dgroos> no… on the server :-(
| |
23:24 | let me re-try…
| |
23:24 | <vagrantc> ah, so we've been chasing the wrong issue, possibly
| |
23:25 | well, i gotta head out soon
| |
23:25 | <dgroos> vagrantc For sure, thanks for giving it a go again.
| |
23:27 | I’m going to try a re-install and see what happens. Also, Simon B and I are going to create detailed docs on using AD in ltsp. He has set up a VM environment. But that will take a while.
| |
23:28 | In the mean while I’m trying to solve this here so may bother you again tomorrow after the students leave. Have a good eve!
| |
23:29 | <vagrantc> good luck!
| |
23:29 | * vagrantc waves | |
23:29 | <dgroos> :-)
| |
23:29 | <vagrantc> dgroos: also, do run the lightdm checks on the fat client; that would be interesting to know
| |
23:29 | <dgroos> vagrantc: will do!
| |
23:31 | vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving) | |
23:31 | <dgroos> vagrantc: yes, it returned /usr/sbin/ldm
| |
23:36 | dgroos has left IRC (dgroos!~dagro001@205.215.175.117, Ping timeout: 240 seconds) | |