IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 27 September 2010 (all times are UTC)
| 00:12 | <elias_a> Good morning!
| |
| 00:12 | <muppis> Morning.
| |
| 00:19 | M4gic5t0rM has quit IRC | |
| 00:21 | <Appiah> tessier: ssh -X ?
| |
| 00:27 | vagrantc has joined #ltsp | |
| 00:34 | <elias_a> As at least Europeans seem to begin to wake I repeat my quest I stated last week....
| |
| 00:34 | Any hints about intelligent pointing devices like Averpen that would work with LTSP?
| |
| 00:35 | <muppis> Tried to install drivers to chroot ?
| |
| 00:36 | <elias_a> We are looking for something that would do something like what smartboards do but using a whiteboard, video projector and a pointing device.
| |
| 00:37 | alkisg has joined #ltsp | |
| 00:59 | zamba has joined #ltsp | |
| 01:01 | <Appiah> what dist are you using elias_a ?
| |
| 01:29 | zamba has quit IRC | |
| 01:31 | <elias_a> Appiah: Mostly Ubuntu. I am asking because I do sort of national coordination here....
| |
| 01:37 | Promethean ActiveInspire seems to have support for Ubuntu: http://www.prometheanworld.com/upload/pdf/ActivInspire_SS_04-10-V2.0_EN.pdf
| |
| 01:41 | <Appiah> just try it on ubuntu and it if does install the stuff (if needed any) on in a chroot and test
| |
| 01:41 | vagrantc has quit IRC | |
| 01:42 | zamba has joined #ltsp | |
| 01:51 | try2free has joined #ltsp | |
| 02:07 | try2free has left #ltsp | |
| 02:11 | gnunux has joined #ltsp | |
| 02:12 | <gnunux> hi
| |
| 02:12 | <alkisg> Good morning
| |
| 02:16 | dobber has joined #ltsp | |
| 02:40 | <elias_a> But as said - we are looking for something a bit more like Averpen...
| |
| 02:40 | <Appiah> ...
| |
| 02:45 | <elias_a> The problem is that there's a lot of gadgets out there and I am trying to get hints on toys that would possibly work on LTSP.
| |
| 02:46 | <johnny> so far, the only things that don't work are cd burners as fars as i know
| |
| 02:46 | most anything else can be made to work
| |
| 02:46 | and of course, if you can use fat client, it's mostly irrelevant on th cd burner side too, or anything else
| |
| 02:48 | <Appiah> last time I said that cdburners dont work on thin clients someone told me I was wrong
| |
| 02:48 | and that the issue was resolved long ago
| |
| 02:49 | elias_a well I havent heard of people using gadgets at all , it's just usb storage,cd/dvd readers/burners and printers
| |
| 02:49 | so you will just have to try
| |
| 02:50 | ask the manufacture if they work on linux
| |
| 02:50 | <elias_a> Appiah: OK. Will do. Thank you!
| |
| 02:57 | alkisg has quit IRC | |
| 03:11 | alkisg has joined #ltsp | |
| 03:15 | mikkel has joined #ltsp | |
| 03:29 | Da-Geek has joined #ltsp | |
| 04:20 | sweetpi has quit IRC | |
| 04:53 | cliebow has joined #ltsp | |
| 05:59 | gu has joined #ltsp | |
| 06:27 | pmatulis has joined #ltsp | |
| 06:36 | cliebow has quit IRC | |
| 07:06 | Da-Geek has quit IRC | |
| 07:21 | bobby_C has joined #ltsp | |
| 07:43 | try2free has joined #ltsp | |
| 07:49 | japerry has quit IRC | |
| 08:00 | <pmatulis> any standard LTSP way of restricting users to a single login other than editing /etc/security/limits/conf ?
| |
| 08:00 | <alkisg> LDM_LIMIT_ONE_SESSION or something like that
| |
| 08:02 | !compiz
| |
| 08:02 | <ltspbot> alkisg: "compiz" :: if compiz is giving you problems, one way to disable it for all users is: sudo gconftool-2 --direct --config-source xml:readwrite:/etc/gconf/gconf.xml.mandatory --type string --set /desktop/gnome/session/required_components/windowmanager metacity
| |
| 08:04 | try2free has left #ltsp | |
| 08:05 | <pmatulis> alkisg: can't find that parameter
| |
| 08:05 | alkisg: or anything like it in /usr/share/doc/ltsp-server/lts-parameters.txt.gz
| |
| 08:05 | (ubuntu)
| |
| 08:06 | <alkisg> pmatulis: it's not documented, as 50% of the rest lts.conf parameters ;)
| |
| 08:06 | LDM_LIMIT_ONE_SESSION=True ==> logs out all other sessions of that user
| |
| 08:06 | <pmatulis> alkisg: i see, i wonder why (undocumented)?
| |
| 08:06 | <alkisg> LDM_LIMIT_ONE_SESSION_PROMPT=True ==> gives a dialog before logging him out from the other pcs
| |
| 08:06 | I guess the developers hate documentation :)
| |
| 08:06 | <pmatulis> strange
| |
| 08:07 | <alkisg> Sure, only 99% of the developers hate documentation... :D
| |
| 08:08 | <Appiah> :)
| |
| 08:12 | <pmatulis> alkisg: so a user trying to log in from a second client will only be allowed if he goes back to the first client and accpets to log out?
| |
| 08:13 | <alkisg> No, he'll be allowed, but the processes from his first login would be forcefully killed
| |
| 08:13 | (so the dialog is a good option too)
| |
| 08:14 | <pmatulis> alkisg: oh ok. so my question applies only if the _PROMPT parameter is used?
| |
| 08:14 | <alkisg> Erm....
| |
| 08:14 | no?
| |
| 08:15 | <pmatulis> alkisg: well, both are used
| |
| 08:15 | <alkisg> Suppose a user has logged in in client01, and wants to login to client02 as well
| |
| 08:15 | If the admin has set LDM_LIMIT_ONE_SESSION=True, then all processes from the client01 login will be killed without any prompts when the user logs in to client02
| |
| 08:16 | If the admin has ALSO set LDM_LIMIT_ONE_SESSION_PROMPT=True, then the user will be displayed a dialog
| |
| 08:16 | If he choses "yes, kill my other logins" (I don't remember the exact text), then the same thing as above is done ^^^
| |
| 08:17 | <pmatulis> alkisg: yeah, so both will force an accept, great, thanks a bunch
| |
| 08:17 | <alkisg> OK
| |
| 08:26 | alexqwesa has quit IRC | |
| 08:26 | alexqwesa has joined #ltsp | |
| 08:34 | mikkel has quit IRC | |
| 08:45 | Gadi has joined #ltsp | |
| 09:38 | jhutchins_lt has joined #ltsp | |
| 10:00 | vagrantc has joined #ltsp | |
| 10:30 | gnunux has quit IRC | |
| 10:47 | alkisg has quit IRC | |
| 10:47 | dobber has quit IRC | |
| 11:01 | <pmatulis> should a thin client user session be able to reboot the server if they have the permissions? i've heard complaints of the process stalling
| |
| 11:08 | <vagrantc> technically, they should be able to ... though it's really easy to do so by accident, so it's maybe not a good idea...
| |
| 11:09 | mistik1 has quit IRC | |
| 11:10 | alkisg has joined #ltsp | |
| 11:39 | mistik1 has joined #ltsp | |
| 12:01 | <Kyle__> pmatulis: If they really are thin clients, I mean, no local HDD to worry about, etc... ANY user can reboot...pull the plug and pop it back in.
| |
| 12:01 | Until ubuntu fixes this fscking kernel bug I'm seeing, my users are resorting to that on my fat clients.
| |
| 12:02 | <pmatulis> Kyle__: i don't get it. pull the plug of the server?
| |
| 12:03 | <Kyle__> pmatulis: No, of the thin-client.
| |
| 12:03 | pmatulis: Doh... I just re-read that after you said server. My moronic mis-read :)
| |
| 12:03 | <pmatulis> Kyle__: well, my questions was about the server
| |
| 12:04 | <Kyle__> pmatulis: Peronsally I would keep reboot-ability to those in wheel, sudo, or however you like to divvy it up. Hopefully users in that group will know better than to do that at 4:55PM on a Friday :)
| |
| 12:05 | <pmatulis> Kyle__: i'm asking b/c an admin person is having trouble rebooting. server hangs
| |
| 12:05 | <Kyle__> pmatulis: Humm? Where is it hanging? Or is it halting, but not powering off?
| |
| 12:06 | <pmatulis> Kyle__: i need to ask more details, it's for a client
| |
| 12:06 | of mine
| |
| 12:07 | <Kyle__> pmatulis: My guess is that it's halting, and not rebooting...but without further data :) there's no way to know.
| |
| 12:07 | * Kyle__ wonders why the hell he's using so many emoticons... braincells must still be off from the weekend. | |
| 12:10 | thunsucker has joined #ltsp | |
| 12:10 | <thunsucker> hey guys, on average how much bandwidth do you think each client needs?
| |
| 12:11 | i'm working on an issue in a lab, and strangely enough I saw 10 or so thin clients running at 12-28mbps, which surprised me
| |
| 12:17 | hersonls has joined #ltsp | |
| 12:38 | vagrantc has quit IRC | |
| 12:41 | Sorinan has joined #ltsp | |
| 12:41 | <Sorinan> Hi! It's me again, and, to vary, with problem about printers...
| |
| 12:42 | the printer on client simply don't print what's sent, and no problem is reported on CUPS, altought the printer works flawlessly locally, and I even tested with another printer on another client, same problem
| |
| 12:43 | <alkisg> thunsucker: a client visiting youtube needs about 50mbps :)
| |
| 12:43 | <Sorinan> I tested too with my smartphone connecting directly to the printer socket (wich uses other driver) and exactly same problem
| |
| 12:44 | this leads me to believe that the problem is really on the client, not in the machine, but in the OS running on it
| |
| 12:45 | I've sniffed the connection, no big problems seems to appear (altought i'm not an expert in eavesdropping)
| |
| 12:46 | what should I do now? jetdirect doesn't seem to die too....
| |
| 12:47 | i'm running debian lenny, may upgrading the system to run a newer version of LTSP?
| |
| 12:47 | *maybe
| |
| 12:48 | i'm really lost (and yes, i'm kinda verbose :-)
| |
| 12:48 | hawkI has quit IRC | |
| 13:01 | <Sorinan> it's normal jetdirect only answer with ACK's or window updates?
| |
| 13:03 | Mixologic has left #ltsp | |
| 13:05 | <Kyle__> Sorinan: This sounds more like a CUPS issue than anything else.
| |
| 13:06 | <Sorinan> hmm, but why so even my smartphone can't print?
| |
| 13:06 | <Kyle__> Sorinan: Did you point your smartphone directly at the jetdirect's IP address, or at the CUPS server yoru clients are trying to get to?
| |
| 13:06 | <Sorinan> directly
| |
| 13:06 | <Kyle__> Sorinan: And what type of printer is hooked up to the jet direct?
| |
| 13:07 | <Sorinan> USB, HP P1005 LaserJet
| |
| 13:07 | U mode on lts.conf
| |
| 13:07 | detected in /dev/usb/lp0
| |
| 13:08 | and locally (connected to server's USB) the printer works good, so it isn't a driver issue
| |
| 13:10 | jetpipe doesn't seems to die even
| |
| 13:10 | there's a way to "sniff" the usb stream?
| |
| 13:12 | <Kyle__> Sorinan: Um. Do you have it hooked up to USB and network at the same time? Some printers really don't like that.
| |
| 13:13 | <Sorinan> no, it doesn't have a network port (if had it would solve a lot of problems :D)
| |
| 13:14 | hey! well, that time the jetpipe died
| |
| 13:14 | the first try to print it simply answer with ACK, the second it dies
| |
| 13:14 | <Kyle__> If I recall correctly, a P1005 speaks a funky funky protocol, not really any normal ones natively.
| |
| 13:14 | So printing from your smartphone may do jack...
| |
| 13:15 | <Sorinan> i tried to with an Epson TX125
| |
| 13:15 | <Kyle__> HUmm.
| |
| 13:16 | Sorinan: And what happens when you hook the printer up to a machine via USB, and use that as a CUPS server?
| |
| 13:16 | <johnny> try printing from the server..
| |
| 13:16 | jhutchins_lt has quit IRC | |
| 13:16 | <Sorinan> with the printer connected directly to server by USB, it works normally
| |
| 13:17 | <Kyle__> Sorinan: Then what's the problem leaving on there, using a CUPS server, and then IGNORING lts.conf, and dropping a cupsd.conf in your client image that tells it to show/use all printers shared by other systems?
| |
| 13:18 | <Sorinan> wait, install CUPS on client image?
| |
| 13:19 | <Kyle__> Sorinan: Sure, why not?
| |
| 13:19 | <Sorinan> one CUPS can talk with another CUPS?
| |
| 13:19 | * Kyle__ nods | |
| 13:19 | <Kyle__> Sorinan: CUPS on a desktop is exactly the same as CUPS on the server. It's just how it's configured.
| |
| 13:19 | <Sorinan> and the protocol is the same? jetdirect?
| |
| 13:19 | or IPP?
| |
| 13:20 | <Kyle__> Sorinan: their native protocol is IPP, but server1 can print to server2 via IPP, and server2 can go ahead and toss that printjob to printer1 via jetdirect.
| |
| 13:21 | CUPS to CUPS is genereally IPP. CUPS to printer can be IPP/LDP/JD/FTP/IP/usb/lpt/serial
| |
| 13:21 | Is this making sense?
| |
| 13:22 | <Sorinan> a little :D
| |
| 13:23 | well, but how can I name the printers on a CUPS installation on client? it wouldn't mess up if i use a printer on another client?
| |
| 13:25 | <Kyle__> Sorinan: when a client sees a shared printer, it shows up as the shared name.
| |
| 13:26 | Sorinan: There are some rules inside of cups to handle what happens if it sees two or more servers with the same printer name, but I'm not sure exactly what the rules are.
| |
| 13:26 | <Sorinan> but that way, in cupsd.conf in client image must have a config for every printer?
| |
| 13:27 | <Kyle__> Sorinan: No. You can make a config that includes no printers, but will use ANY printer shared by a cups server.
| |
| 13:28 | <Sorinan> do you have a link or a search term to a howto for this?
| |
| 13:28 | <Kyle__> Sorinan: http://cups.org
| |
| 13:29 | Sorinan: It's actually pretty streightforward.
| |
| 13:29 | <Sorinan> hmm, but what exactly i must search? generic printer config?
| |
| 13:29 | jhutchins_lt has joined #ltsp | |
| 13:29 | <Kyle__> Sorinan: BrowseAddress
| |
| 13:30 | Sorinan: You would have to test it, but more or less something like...
| |
| 13:30 | Browsing On\nBroseAllow all\nBrowseLocalProtocols CUPS\nBrowseAddress <yourseverip>
| |
| 13:31 | If you'll excuse the c carriage returns :), that should show all printers shared by your CUPS print server, and ONLY the printers shared by your CUPS print server
| |
| 13:32 | Head over to #CUPS (I think that's a real channel...) or cups.org for more info on it.
| |
| 13:32 | <Sorinan> hmm, I think I got it, the server1 would have permission to browse and do whatever he likes on client1 CUPS?
| |
| 13:32 | <Kyle__> Server1 would share the printer, client1 would browse server1.
| |
| 13:33 | <Sorinan> but isn't needed printer-specifc config on cupsd.conf in server1, because he supports something like hotplug, isn't?
| |
| 13:34 | <Kyle__> Sorinan: If you've got hotplug, and trust it/etc, then yes, as long as your default is to share you don't need to do any config on the server.
| |
| 13:34 | Sorinan: But generally I would add the printer on the server, and make sure it's shared. Then you're all done.
| |
| 13:35 | <Sorinan> hmm, but that wouldn't require printer-specific config on cupsd.conf on LTSP client image?
| |
| 13:36 | <Kyle__> Sorinan: No, they need the simple config that just shows all shared printers.
| |
| 13:37 | Sorinan: If CUPS as a client is speaking to CUPS as a server, CUPS as a client gets everything it needs from CUPS as a server.
| |
| 13:37 | <Sorinan> ok, so LTSP client (print server) says: "Hey LTSP server, that's all printers that i have, do you wanna print something?" and on LTSP server (print client) that resides the printer-specific config?
| |
| 13:39 | <Kyle__> Is the print-server one of your LTSP-clients?
| |
| 13:39 | <thunsucker> alkisg: what about heavy flash usage (bandwidth wise)
| |
| 13:39 | <Sorinan> Kyle__, yes
| |
| 13:40 | <Kyle__> Sorinan: Ah. _that_ machine, that single solitary machine, needs to know about the printer, and share it. Yes. The other clients just need to know to listen
| |
| 13:41 | <alkisg> thunsucker: well, full screen flash video doesn't use XV acceleration, so it's screen-width * screen-height * (bits-per-pixel/8) * frames-per-second
| |
| 13:42 | If your monitor is 1920x1080, and the flash video is e.g. 30 fps, you have big problems :)
| |
| 13:42 | <Kyle__> alkisg: It doesn't?
| |
| 13:42 | <alkisg> No, flash doesn't. Totem and the rest of the apps do.
| |
| 13:42 | <johnny> does lightspark?
| |
| 13:43 | <alkisg> For XV, it's "original-video-width * original-video-height * 12 * frames-per-second", independed of the actual screen size
| |
| 13:43 | <thunsucker> alkisg: all 1024x768 running the rdesktop screen script
| |
| 13:43 | <alkisg> thunsucker: rdesktop is a different beast, it uses compression
| |
| 13:43 | I don't know about that.
| |
| 13:44 | I'm sure it will use a lot of CPU and drop a lot of frames, though
| |
| 13:44 | <johnny> try lightspark, seems to be developing at a fast clip, and falls back on gnash as necessary..
| |
| 13:44 | din_os has joined #ltsp | |
| 13:45 | <alkisg> XV would be irrelevant if rdesktop is used...
| |
| 13:48 | <din_os> Hello everyone, ltsp on lucid seems to be flying fast!! but after I enter the login credentials on the client it takes a while: 1.to display the panel, 2. then after a while the desktop, 3 and after a while the desktop icons...
| |
| 13:49 | after this process (totals to a bit more than a minute) everything is fast and responsive.. any ideas?
| |
| 13:49 | <alkisg> dns timeouts? Try putting the client name in /etc/hosts, e.g. 10.160.31.131 ltsp131
| |
| 13:50 | Otherwise, install bootchart to watch what processes take time at that point
| |
| 13:50 | <din_os> well, internet sites are responsive... not sure if this is dns related
| |
| 13:51 | <alkisg> When the clients connect to the server, the server does a reverse lookup
| |
| 13:51 | So if you don't have a local dns server, then the external servers are questioned, which may give long delays
| |
| 13:53 | <din_os> aha, interesting... I'll do some more trials based on that... bootchart also sounds good, I'll look into it
| |
| 13:54 | <alkisg> If you watch the server nic at that point with wireshark or tcpdump, you may see "which has IP 10.160.31.131"? That's a reverse lookup
| |
| 13:55 | If you put those entries in /etc/hosts and the delays are gone, then that's the problem, otherwise try bootchart
| |
| 13:55 | (or just install dnsmasq ;))
| |
| 14:04 | <Sorinan> My server is amd64, but the LTSP image is i386, when I install cups, it needs a module, i've mounted the procfs of the server, when I load the image on client it would search for i386 or amd64 modules?
| |
| 14:10 | dlezcano has joined #ltsp | |
| 14:10 | <Sorinan> well, let's try to reformulate the question, how do I mount procfs in a way to fake the architeture? amd64 is the host and i386 is the arch that i want to fake
| |
| 14:14 | <dlezcano> Sorinan: I just connected to the channel, so I don't have the context but I suggest to unshare a mount namespace, write a file with your cpu description and mount --bind mycpuinfo /proc/cpuinfo
| |
| 14:14 | in order to check:
| |
| 14:14 | unshare -m /bin/bash
| |
| 14:15 | echo i386 > mycpuinfo
| |
| 14:15 | mount --bind mycpuinfo /proc/cpuinfo
| |
| 14:15 | cat /proc/cpuinfo
| |
| 14:25 | mgariepy has joined #ltsp | |
| 14:27 | <alkisg> I'm using NFS home for fat clients on many labs. On one of them, the mounting process just hangs. It appears to be some race condition, because if I stop nfs-kernel-server, then boot the client up to ldm, and switch to vt1 and manually try to mount /home, then it works.
| |
| 14:27 | Any ideas for troubleshooting?
| |
| 14:28 | *I start again nfs-kernel-server before manually trying to mount /home
| |
| 14:28 | <Kyle__> alkisg: Umm... What is your NFS server?
| |
| 14:29 | <alkisg> Stock ubuntu lucid
| |
| 14:29 | (clients too)
| |
| 14:29 | <Kyle__> alkisg: Make sure your statd is running on your server.
| |
| 14:30 | <alkisg> I checked that, it is. If I try to boot a vbox client (vbox running on the server), it always succeeds. But the real clients fail...
| |
| 14:30 | <Kyle__> Humm. I can bounce all 30 clients, in the larger lab, simultaniously, and they all grab NFS no problem.
| |
| 14:31 | <alkisg> Right, I have the exact same setup in dozen other labs with no problems
| |
| 14:31 | It must be some upstart or similar race condition, but I have no clue how to search that...
| |
| 14:31 | <Kyle__> alkisg: This is one machien in one lab, or the whole lab?
| |
| 14:32 | <alkisg> The whole lab, 2 kinds of machines
| |
| 14:32 | <Kyle__> Well, you could hackishly include a startup script that checks for a mounted /home, and mounts it manually otheriwse...
| |
| 14:32 | Ugh.
| |
| 14:33 | <alkisg> When the first try for nfs mounting hangs, then all subsequent nfs mounting commands also fail
| |
| 14:33 | <Kyle__> alkisg: Possible that there's a mis-behaving switch or router between that lab and the rest of the world?
| |
| 14:33 | Oh.
| |
| 14:34 | <alkisg> E.g. I tried ssh'ing into the client when it hanged, and tried to restart services etc, to no avail, I needed to reboot the client to get it to mount nfs
| |
| 14:34 | Maybe, is it possible that switches can cause nfs problems?
| |
| 14:34 | <Kyle__> alkisg: But everything else works?
| |
| 14:34 | <alkisg> Yes, e.g. if I manually run the nfs command, then everything works fine
| |
| 14:34 | <Kyle__> alkisg: In some very strange situations... yea...but a firewall/router misbehaving is more likely.
| |
| 14:34 | Are you doing nfs over tcp?
| |
| 14:35 | <alkisg> Yes
| |
| 14:35 | <Kyle__> Damn.
| |
| 14:35 | <alkisg> Can I just change that to udp and retry?
| |
| 14:36 | <Kyle__> May be worth it, but I have _no_ idea why that would go screwie....
| |
| 14:36 | <alkisg> Any way to properly kill the nfs client when I'm ssh'ed into the client?
| |
| 14:36 | (to test without rebooting all the time)
| |
| 14:36 | E.g.: ls /home ==> hangs
| |
| 14:36 | <Kyle__> Does anyone know how to add a firefox plugin to all users in a system? I've only ever downloaded them through the firefox gui...
| |
| 14:36 | alkisg: Ugh. I know the situation...
| |
| 14:36 | <alkisg> service stop nfs-common ==> ls is still hanged
| |
| 14:37 | umount /home => busy, etc etc...
| |
| 14:37 | <Kyle__> alkisg: Sloppy umount
| |
| 14:37 | alkisg: Ok ok, the man page says lazy, I call it sloppy.
| |
| 14:37 | umount -lf /home
| |
| 14:38 | <alkisg> You mean umount -l? ah , ok
| |
| 14:38 | ty
| |
| 14:38 | <Kyle__> -l is lazy, -f is force.
| |
| 14:38 | Welcome.
| |
| 14:38 | I think nfsio is the process BTW...but try that first.
| |
| 14:41 | vagrantc has joined #ltsp | |
| 14:41 | <alkisg> I think nfsio takes one full cpu in the "waiting" stage :)
| |
| 14:46 | jhutchins_lt has quit IRC | |
| 14:52 | pmatulis has quit IRC | |
| 14:53 | M4gic5t0rM has joined #ltsp | |
| 15:06 | Appiah has quit IRC | |
| 15:06 | Gadi has left #ltsp | |
| 15:15 | <alkisg> Hmmm debugging shows this: nfs is mounted and working properly when the user is at ldm. E.g. I can ls/cat anything in /home/user from vt1.
| |
| 15:15 | When he tries to login with a simple xterm session, nfs hangs... :-/
| |
| 15:15 | jhutchins_lt has joined #ltsp | |
| 15:15 | <Kyle__> O_o
| |
| 15:15 | alkisg: That rather makes no sense....
| |
| 15:15 | <alkisg> I KNOW!!! :( :( :(
| |
| 15:16 | I wonder if it has anything to do with gvfs
| |
| 15:16 | <Kyle__> Is the user already logged in elsewhere?
| |
| 15:16 | <alkisg> No
| |
| 15:16 | I've also with a new user, the same
| |
| 15:16 | * Kyle__ frowns | |
| 15:17 | korcan has joined #ltsp | |
| 15:17 | <Kyle__> alkisg: are you using gvfs to mount /home?
| |
| 15:17 | <alkisg> No, gnome mounts .gvfs under /home/user/.gvfs...
| |
| 15:17 | * Kyle__ really hates gvfs.... | |
| 15:17 | <alkisg> (but I suppose that doesn't happen with an xterm session, does it?)
| |
| 15:18 | <Kyle__> Ahh ok. Hummm... Do you have a kde session you could try it with?
| |
| 15:18 | <alkisg> No, only gnome...
| |
| 15:18 | * alkisg is thinking about uninstalling gvfs... :D | |
| 15:18 | <Kyle__> alkisg: Will ubuntu's gnome session work without it?
| |
| 15:19 | <alkisg> No idea, I saw it somewhere on google but never tried it
| |
| 15:21 | <Sorinan> dlezcano, sorry for the delay, the idea is only fake the arch (amd64 to i386) write a whole new cpuinfo would be a pain in the ass, from where the "uname -m" gets the info?
| |
| 15:21 | <dlezcano> Sorinan: oh, ok. You may change the process personality then.
| |
| 15:22 | <alkisg> Kyle__: btw, if I boot that client as a thin client, then it works fine....
| |
| 15:22 | (so no driver etc problems)
| |
| 15:22 | <Sorinan> dlezcano, hmm, how?
| |
| 15:24 | <Kyle__> alkisg: Aaaarg. That's a pain.
| |
| 15:24 | <dlezcano> Sorinan: personality(PER_LINUX32);
| |
| 15:24 | in C, otherwise setarch i686 for a shell script
| |
| 15:24 | eg. setarch i686 uname -m
| |
| 15:27 | <Sorinan> hmm, thanks, the scripts now see the arch as i386, but i keep getting an error on cups install
| |
| 15:27 | FATAL: Could not load /lib/modules/2.6.26-2-amd64/modules.dep: No such file or directory
| |
| 15:27 | this isn't the kernel that LTSP image have
| |
| 15:28 | do you know if when I run it on client, it would work correctly?
| |
| 15:28 | <dlezcano> nope
| |
| 15:28 | what gives you uname -r ?
| |
| 15:28 | <Sorinan> the host kernel
| |
| 15:29 | wich is amd64
| |
| 15:29 | well, nothing better than try
| |
| 15:31 | Appiah has joined #ltsp | |
| 15:34 | <Sorinan> dlezcano, worked!
| |
| 15:34 | thanks for the help
| |
| 15:35 | <dlezcano> you are welcome
| |
| 15:35 | ogra_ac has joined #ltsp | |
| 15:41 | mgariepy has quit IRC | |
| 15:42 | <Kyle__> ltsp-update-kernels is NOT run as part of ltsp-update-image, correct?
| |
| 15:43 | Leolo_3 has joined #ltsp | |
| 15:43 | <johnny> easy to find out..
| |
| 15:44 | <Kyle__> Actually looks like it is, if you tell it to force. I keep forgetting these are all scripts and easy to just read :)
| |
| 15:44 | <Leolo_3> OK, I have a mad request; how can one use x11vnc to look at a user's session? Hard mode : using ltsp-4
| |
| 15:45 | <johnny> nobody knows ltsp4 anymore.. sorry..
| |
| 15:45 | <Leolo_3> nobody but me, it seems
| |
| 15:45 | <alkisg> Leolo_3: why don't you enable vino-server in gnome settings?
| |
| 15:46 | <Leolo_3> gnome?
| |
| 15:46 | using xfce4
| |
| 15:47 | <alkisg> Hmmm ok... here's an easy workaround: put x11vnc -noshm -loop connect_or_exit localhost in /etc/xdg/autostart/some-desktop-file.desktop
| |
| 15:47 | Then just run vncviewer -listen...
| |
| 15:47 | Ugly, but works :D
| |
| 15:47 | <Leolo_3> *blink*
| |
| 15:48 | that requires user participation.
| |
| 15:48 | or am I missing something
| |
| 15:48 | <alkisg> No, /etc/xdg/autostart runs automatically in the user session
| |
| 15:48 | <Leolo_3> hmmmm...
| |
| 15:49 | M4gic5t0rM has quit IRC | |
| 15:49 | <alkisg> (of course if anyone runs vncviewer -listen, he'll see all the user sessions :D)
| |
| 15:49 | There are password options in x11vnc though
| |
| 15:50 | M4gic5t0rM has joined #ltsp | |
| 15:51 | <Kyle__> Leolo_3: Have you played with the VNC module that's part of xorg?
| |
| 15:51 | I thought I'd heard of one before, so I looked it up real quick, and sure enough, there is one.
| |
| 15:54 | ogra_ac has quit IRC | |
| 15:55 | ogra_ac has joined #ltsp | |
| 15:56 | * ogra_ac pokes his wlan card | |
| 15:57 | thunsucker has quit IRC | |
| 15:58 | <Leolo_3> woah
| |
| 15:58 | ok, x11vnc in /etc/xdg/autostart worked
| |
| 15:58 | but... isn't really what I want
| |
| 15:59 | but might be able to fill in for what I want
| |
| 15:59 | what I thought would work su - bob -c "x11vnc -display 10.0.0.111:0.0"
| |
| 16:04 | hersonls has quit IRC | |
| 16:05 | alexqwesa has quit IRC | |
| 16:09 | <alkisg> Leolo_3: you'd need to retrieve the user's XAUTHORITY and DISPLAY for that to work, with DISPLAY=localhost:something
| |
| 16:09 | (that is after the user has logged in)
| |
| 16:11 | Something like eval $(tr '\0' '\n' < /proc/$(pidof some-user-process)/environ | egrep '^DISPLAY=|^XAUTHORITY=')
| |
| 16:12 | Good night all
| |
| 16:12 | alkisg has quit IRC | |
| 16:14 | * Kyle__ frowns | |
| 16:14 | alexqwesa has joined #ltsp | |
| 16:15 | <Kyle__> OK, if you put a new kernel on your ltsp-image, and ltsp-update-image -f, that should be everything you need, right?
| |
| 16:16 | mgariepy has joined #ltsp | |
| 16:16 | <Kyle__> ls
| |
| 16:16 | <Sorinan> Kyle__, my client says that the server (CUPS) don't have any queues, but the browsing is enabled, do you know why?
| |
| 16:17 | <Kyle__> Sorinan: Maybe the server isn't setup to share it's queues?
| |
| 16:21 | <Sorinan> Browsing On\n BrowseOrder allow,deny\n BrowseAllow from 192.168.*\n
| |
| 16:21 | that's what I have
| |
| 16:21 | share queues=browsing, right?
| |
| 16:22 | <Kyle__> Sorinan: Off the top of my head Ican't recall the exact syntax. Have you tried using the web-browser to hit the cups-server at :631, to use the gui?
| |
| 16:22 | * Kyle__ grins | |
| 16:22 | <Kyle__> sorry I'm fighting my own problems right this second.
| |
| 16:23 | Like why this silly kernel won't update.
| |
| 16:24 | <Sorinan> yes, i had some problems with permissions, but now all works on gui
| |
| 16:24 | <Kyle__> OK.
| |
| 16:24 | Humm.
| |
| 16:34 | <Sorinan> Kyle__, here's my cupsd.conf on CUPS server (LTSP client) http://pastebin.com/5g9PWrqW
| |
| 16:36 | ogra_ac has quit IRC | |
| 16:38 | ogra_ac has joined #ltsp | |
| 16:39 | sweetpi has joined #ltsp | |
| 16:41 | <Kyle__> Sorinan: at a glance hat looks right...Dunno why it's not working
| |
| 16:42 | <Sorinan> hmm, the printer must be added in CUPS server config? or just on client?
| |
| 16:42 | <Kyle__> Sorinan: On the server.
| |
| 16:43 | <Sorinan> but that way, whenever I want add a printer, I would have to edit the LTSP image?
| |
| 16:44 | <Kyle__> Err.
| |
| 16:44 | Sorinan: The image that the CUPS server uses yes. If it's the same image the rest of the clients use, then, I guess yes.
| |
| 16:45 | <Sorinan> hmmm, creating a different image for each terminal isn't a smart thing....
| |
| 16:45 | there's another solution?
| |
| 16:46 | i just want that all machines running the image shares all their connected printers, with hotswap, without having to change the image
| |
| 16:46 | korcan_ has joined #ltsp | |
| 16:46 | <Sorinan> there's a way to that?
| |
| 16:46 | something like a direct "USB tunnel", maybe?
| |
| 16:47 | <Kyle__> Sorinan: If your cupsd.conf is setup so that it always shares every printer it has, that would do it. And I think that's pretty much what yours has, but I'm not sure without further thought/investigation.
| |
| 16:49 | korcan has quit IRC | |
| 16:50 | <Sorinan> yes, the printer is detected (when i try to mannually add), but there's a way to auto-add it?
| |
| 16:51 | korcan_ has quit IRC | |
| 16:58 | Sorinan has quit IRC | |
| 16:59 | mgariepy has quit IRC | |
| 17:04 | din_os has quit IRC | |
| 17:06 | <Kyle__> Well, for Sorinan, if he comes back. yes. The lpadmin program will let you add printers from the command line, and set their permissions, drivers, etc.
| |
| 17:11 | ogra_ac has quit IRC | |
| 17:11 | ogra_ac has joined #ltsp | |
| 17:12 | <Kyle__> ogra_ac: Still wlan problems?
| |
| 17:12 | <ogra_ac> yes thats what you get with weird setups
| |
| 17:14 | <stgraber> ogra_ac: got my mail ?
| |
| 17:15 | ogra_cmpc has quit IRC | |
| 17:15 | <Kyle__> ogra_ac: Weird? Like cheapie-no-name-ordered-from-china laptops?
| |
| 17:17 | ogra_ac has quit IRC | |
| 17:17 | ogra_ac has joined #ltsp | |
| 17:17 | <ogra_ac> grmpf
| |
| 17:17 | stgraber, yeah, i dont think i can make it that early
| |
| 17:18 | still havent found the time to look it up though, but i think i arrive in boston in the afternoon
| |
| 17:18 | <stgraber> ogra_ac: hmm, ok. I'm sure we'll find someone to come pick you up at the airport anyway
| |
| 17:18 | <ogra_ac> worst case i'll pick a rental car
| |
| 17:19 | but jammcq already said he might have somebody
| |
| 17:33 | * vagrantc waves | |
| 17:33 | <ogra_ac> vagrantc, !!!
| |
| 17:36 | bobby_C has quit IRC | |
| 17:37 | <vagrantc> y'all getting together for LTSP BTS or something elsewise?
| |
| 17:37 | ogra_ac has quit IRC | |
| 17:37 | ogra_ac has joined #ltsp | |
| 17:40 | <Leolo_3> alksig : oooo! I never thought to grep /proc/$PID/env for that info
| |
| 17:41 | alexqwesa has quit IRC | |
| 17:42 | ogra_ac has quit IRC | |
| 17:44 | ogra_ac has joined #ltsp | |
| 17:45 | vagrantc has quit IRC | |
| 17:48 | alexqwesa has joined #ltsp | |
| 17:50 | sweetpi has quit IRC | |
| 17:53 | M4gic5t0rM has quit IRC | |
| 17:54 | <Leolo_3> solution to my problem : -noshm
| |
| 17:54 | alexqwesa has quit IRC | |
| 17:56 | alexqwesa has joined #ltsp | |
| 18:02 | alexqwesa has quit IRC | |
| 18:04 | alexqwesa has joined #ltsp | |
| 18:05 | gecko_zo6 has joined #ltsp | |
| 18:05 | <gecko_zo6> good afternoon
| |
| 18:06 | cannot boot my ltsp clients
| |
| 18:06 | seems that the problem is in tftp
| |
| 18:06 | I install the package tftpd-hpa
| |
| 18:07 | I try to run LTSP in debian Lenny
| |
| 18:07 | can someone help me
| |
| 18:08 | <Leolo_3> is tftpd activated?
| |
| 18:08 | <gecko_zo6> yes
| |
| 18:08 | I see it with netstat -tulp
| |
| 18:08 | and the daemon is already running
| |
| 18:09 | and i try to connect from a local client and seems to work
| |
| 18:10 | but when my ltsp clients start the dhcp3-server gives an IP and when start tftp nothing happend
| |
| 18:10 | tftp arp time out
| |
| 18:14 | when I restart the tftp server I can see this in syslog
| |
| 18:14 | cannot bind to local socket
| |
| 18:14 | address already in use
| |
| 18:15 | <Leolo_3> iptables?
| |
| 18:15 | <gecko_zo6> do not have iptables installed
| |
| 18:15 | <Leolo_3> lsof -i | grep tftp # find out who is using the tftp port
| |
| 18:16 | <gecko_zo6> inetd
| |
| 18:17 | knipwim has joined #ltsp | |
| 18:17 | <Leolo_3> right
| |
| 18:17 | <gecko_zo6> :)
| |
| 18:17 | <Leolo_3> good.
| |
| 18:18 | knipwim_ has quit IRC | |
| 18:18 | <gecko_zo6> so it's inetd who's ussing the service
| |
| 18:18 | <Leolo_3> yeah. though that's what I would expect. but maybe ftpd-hpa wants to run as a daemon, not from inetd
| |
| 18:19 | <gecko_zo6> so i have to comment from /etc/default right? :)
| |
| 18:20 | so I have to run it from inetd?
| |
| 18:23 | jhutchins_lt has quit IRC | |
| 18:24 | rjune has quit IRC | |
| 18:24 | <gecko_zo6> I stop the daemon and put no in daemon mode the tftp and run the service from inetd but nothing happends
| |
| 18:25 | AndyGraybeal has quit IRC | |
| 18:25 | tessier has quit IRC | |
| 18:25 | <gecko_zo6> is still do not working the tftp
| |
| 18:26 | or is the daemon mode the way to go???
| |
| 18:27 | any ideas???
| |
| 18:29 | tessier has joined #ltsp | |
| 18:29 | AndyGraybeal has joined #ltsp | |
| 18:38 | <gecko_zo6> ??
| |
| 18:40 | <Leolo_3> first : what tftp server was inetd launching?
| |
| 18:42 | second : do you want to run from inetd or as a daemon. CentOS runs it from inetd. maybe tftpd-hpa wants to act differently
| |
| 18:42 | CentOS uses xinetd, rather. but that's beside the point
| |
| 18:42 | third, simply try service inetd stop ; service tftpd-hpa start and look at your syslog.
| |
| 18:43 | or better : deactivate tftpd in inetd, service inetd reload ; then try to restart tftpd-hpa
| |
| 18:44 | gecko_zo6 has quit IRC | |
| 18:44 | gecko_zo6 has joined #ltsp | |
| 18:46 | irule has joined #ltsp | |
| 18:52 | <johnny> you're supposed to use inetd
| |
| 18:52 | Leolo_3, no..
| |
| 18:52 | the default is inetd
| |
| 18:52 | gecko_zo6, you need to turn off the the tftp-hpa daemon
| |
| 18:52 | inetd takes care ofi t
| |
| 18:56 | gecko_zo6 has quit IRC | |
| 19:00 | pmatulis has joined #ltsp | |
| 19:12 | cliebow has joined #ltsp | |
| 19:12 | cliebow_ has joined #ltsp | |
| 19:20 | cliebow has quit IRC | |
| 19:20 | cliebow has joined #ltsp | |
| 19:38 | MorningSon has quit IRC | |
| 19:42 | MorningSon has joined #ltsp | |
| 20:07 | rjune has joined #ltsp | |
| 20:11 | alexqwesa has quit IRC | |
| 20:12 | alexqwesa has joined #ltsp | |
| 20:14 | cliebow has joined #ltsp | |
| 20:14 | cliebow_ has joined #ltsp | |
| 20:17 | <cliebow> seen jammcq
| |
| 20:17 | <ltsplogbot> I don't know who jammcq is.
| |
| 20:17 | try2free has joined #ltsp | |
| 20:17 | <cliebow> !seen jammcq
| |
| 20:17 | <ltspbot> cliebow: jammcq was last seen in #ltsp 3 days, 3 hours, 50 minutes, and 7 seconds ago: <jammcq> sometimes that's the funniest part
| |
| 20:28 | try2free has left #ltsp | |
| 20:29 | zamba has quit IRC | |
| 20:37 | jhutchins_lt has joined #ltsp | |
| 20:47 | jhutchins_lt has quit IRC | |
| 20:51 | sweetpi has joined #ltsp | |
| 20:57 | irule has quit IRC | |
| 21:03 | cliebow_ has quit IRC | |
| 21:03 | cliebow has quit IRC | |
| 21:16 | jappiecr has joined #ltsp | |
| 21:17 | <jappiecr> hello, can I ask questions here?
| |
| 21:18 | <highvoltage> isn't that a question you asked right there?
| |
| 21:19 | <jappiecr> can I get answers here? :-)
| |
| 21:19 | <highvoltage> isn't there just one way to find out?
| |
| 21:22 | <jappiecr> I am TRYING to set up a ltsp classroom server for a local school in Costa Rica. My server runs and when I boot a thin client it gets up to the login screen and when I login I get failure due to unknown user id
| |
| 21:22 | <highvoltage> you probably want to run the following on the server:
| |
| 21:23 | sudo ltsp-update-sshkeys
| |
| 21:23 | sudo ltsp-update-image
| |
| 21:23 | jappiecr: it's slightly dependent on which distribution you use. is it ubuntu?
| |
| 21:25 | <jappiecr> I installed edubuntu 10.04
| |
| 21:25 | M4gic5t0rM has joined #ltsp | |
| 21:28 | <highvoltage> jappiecr: great, then that should work just fine
| |
| 21:28 | ogra has quit IRC | |
| 21:28 | <jappiecr> great it works! thank u
| |
| 21:28 | oeps...no it does not
| |
| 21:29 | ogra has joined #ltsp | |
| 21:29 | <highvoltage> jappiecr: can you see the login attempt in /var/log/auth.log?
| |
| 21:29 | <jappiecr> process 256 glib warning: **: getpwuid_r (): failed due to unknown user id (0)
| |
| 21:30 | <highvoltage> yeah, unfortunately LDM's error codes aren't all that helpful
| |
| 21:30 | (not that it really has error codes)
| |
| 21:30 | <johnny> that's not ldm is it
| |
| 21:30 | <highvoltage> jappiecr: ah, did you reboot the thin client? I think you might need to
| |
| 21:31 | <johnny> that's pulse audio i think.. shouldn't be fatal
| |
| 21:31 | so it's not an ldm errror at all
| |
| 21:31 | it should be supressed, or not generated at all unless asked, but that requires patches
| |
| 21:31 | <highvoltage> ah yes, indeed. either way that error is unrelated :)
| |
| 21:32 | *nod* I agree it's kind of annoying
| |
| 21:32 | gu has quit IRC | |
| 21:33 | <jappiecr> LDM?
| |
| 21:33 | <highvoltage> jappiecr: LDM is the login manager on the thin clients that connects you to the server
| |
| 21:33 | !LDM
| |
| 21:33 | <ltspbot> highvoltage: Error: "LDM" is not a valid command.
| |
| 21:33 | <highvoltage> ltspbot: you're useless!
| |
| 21:33 | <ltspbot> highvoltage: Error: "you're" is not a valid command.
| |
| 21:33 | <highvoltage> pfft.
| |
| 21:33 | !info LDM
| |
| 21:33 | <ltspbot> highvoltage: Error: No factoid matches that key.
| |
| 21:35 | <highvoltage> ltspbot: LDM is the LTSP Display Manager, the login client that runs on the thin clients that connects you to your desktop session on the application server
| |
| 21:35 | <ltspbot> highvoltage: Error: "LDM" is not a valid command.
| |
| 21:35 | * highvoltage doesn't know ltspbot | |
| 21:36 | * highvoltage bbl | |
| 21:36 | <highvoltage> jappiecr: hope you get an answer to your problem!
| |
| 21:37 | <jappiecr> ok...I rebooted and got to login. Than when logined and the login tune finished it crashed to the same error...fatal audio?
| |
| 21:42 | ...and yes I can see the login attempt in /var/log/auth.log...
| |
| 21:42 | Sep 27 20:35:41 administrador-desktop sshd[31238]: pam_unix(sshd:session): session opened for user administrador by (uid=0)
| |
| 21:42 | Sep 27 20:35:41 administrador-desktop sshd[31368]: subsystem request for sftp
| |
| 21:42 | Sep 27 20:35:42 administrador-desktop polkitd(authority=local): Registered Authentication Agent for session /org/freedesktop/ConsoleKit/Session6 (system bus name :1.99 [/usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1], object path /org/gnome/PolicyKit1/AuthenticationAgent, locale en_US.utf8)
| |
| 21:42 | Sep 27 20:35:47 administrador-desktop sshd[31368]: channel 27: open failed: administratively prohibited: open failed
| |
| 21:42 | Sep 27 20:35:48 administrador-desktop gnome-keyring-daemon[31463]: dbus failure unregistering from session: Connection is closed
| |
| 21:42 | Sep 27 20:35:48 administrador-desktop sshd[31368]: Received disconnect from 10.0.30.20: 11: disconnected by user
| |
| 21:42 | Sep 27 20:35:48 administrador-desktop sshd[31238]: pam_unix(sshd:session): session closed for user administrador
| |
| 21:42 | Sep 27 20:35:48 administrador-desktop polkitd(authority=local): Unregistered Authentication Agent for session /org/freedesktop/ConsoleKit/Session6 (system bus name :1.99, object path /org/gnome/PolicyKit1/AuthenticationAgent, locale en_US.utf8) (disconnected from bus)
| |
| 21:42 | Sep 27 20:35:48 administrador-desktop gnome-keyring-daemon[31463]: dbus failure unregistering from session: Connection is closed
| |
| 21:55 | MorningSon has quit IRC | |
| 21:59 | robbie_ has joined #ltsp | |
| 21:59 | robbie_ is now known as eibbor | |
| 21:59 | <jappiecr> highvoltage, I am first time user and so thank you for trying to help me! Can I just leave xchat open and wait for somebody who knows
| |
| 22:07 | johnny has left #ltsp | |
| 22:08 | johnny has joined #ltsp | |
| 22:11 | Mixologic has joined #ltsp | |
| 22:14 | alkisg has joined #ltsp | |
| 22:14 | alkisg has joined #ltsp | |
| 22:24 | M4gic5t0rM has quit IRC | |
| 22:28 | eibbor has quit IRC | |
| 22:41 | alkisg has quit IRC | |
| 23:01 | mistik1 has quit IRC | |
| 23:12 | pmatulis has quit IRC | |
| 23:14 | mistik1 has joined #ltsp | |
| 23:43 | stgraber has quit IRC | |
| 23:43 | stgraber has joined #ltsp | |