IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 5 January 2009 (all times are UTC)
| 00:03 | <alkisg> Wouldn't it be better if SSH_ASKPASS was used in ldm instead of expecting "Password:"?
| |
| 00:05 | <Ryan52> probably :)
| |
| 00:05 | what would you use?
| |
| 00:05 | <alkisg> ldm again!
| |
| 00:05 | I mean, the greeter would get the password and keep it,
| |
| 00:06 | <Ryan52> err, what?
| |
| 00:06 | <alkisg> and the ssh_askpass program would be ldm (maybe with a command line switch) that asks the greeter about the stored password
| |
| 00:06 | <Ryan52> okay.
| |
| 00:06 | <alkisg> So no gui involved in SSH_ASKPASS
| |
| 00:06 | <Ryan52> well, no, it would tell the greeter to ask for it with the prompt that is giving in ARGV.
| |
| 00:06 | for password expiry and stuff.
| |
| 00:07 | wow, that makes a lot of sense :)
| |
| 00:07 | how did we never think of that?
| |
| 00:07 | <alkisg> Heh
| |
| 00:08 | <Ryan52> probably we could just have a fifo (or something stupid like that) that lets a new program called ldm-askpass (or something stupid like that) tell ldm to talk to the greeter.
| |
| 00:10 | <alkisg> Yeah... So, I think it should go like this: greeter gets username + password, then runs ssh with SSH_ASKPASS set to ldm-askpass, then ldm-askpass is called and signals ldm to ask the greeter about the stored password
| |
| 00:11 | <Ryan52> ya.
| |
| 00:12 | <alkisg> (I'm trying to find a way to minimize the timeouts, they're a major problem if the students keep forgeting their passwords... :))
| |
| 00:12 | <warren> isn't ssh_askpass only for ssh key auth?
| |
| 00:12 | <Ryan52> ya.
| |
| 00:12 | warren: works for passwords too.
| |
| 00:21 | <alkisg> About password timeout: the ssh connection could be initiated when the greeter has both username + password. So the password timeout would be a greeter-only concern. E.g. if I enter a username and just wait for a timeout, ssh doesn't have to be called.
| |
| 00:22 | <Ryan52> password timeout?
| |
| 00:23 | <alkisg> I think there is one, isn't there? :)
| |
| 00:23 | <Ryan52> I don't think so :)
| |
| 00:23 | <alkisg> If I enter a username and wait for 10 minutes without entering the password, what happens?
| |
| 00:23 | OK, let me try...
| |
| 00:25 | But is my understanding correct that ssh gets run before the user enters his password?
| |
| 00:25 | <Ryan52> nope.
| |
| 00:25 | * Ryan52 just checked the code | |
| 00:26 | <alkisg> OK, sorry! :)
| |
| 00:29 | <Ryan52> alkisg, are you intending on implementing the SSH_ASKPASS thing? if not, I will be happy to (after discussing with other people, and playing with it a bit) :). that would be a major improvement to some of the hackish stuff that we're doing...
| |
| 00:30 | <alkisg> Ryan52: please do!!! I don't have much experience when it comes to glibc and stuff, I'm new to Linux
| |
| 00:30 | <Ryan52> o i c. ok.
| |
| 00:30 | <alkisg> Hmm... I just tried this: I entered the username, waited for 2 minutes, entered the password and I got "No response from server, restarting..."
| |
| 00:31 | So I guess there is a timeout somewhere...
| |
| 00:31 | <Ryan52> it's not always no response..
| |
| 00:31 | if you entered stuff wrong it also says that.
| |
| 00:32 | <alkisg> I think if you enter wrong stuff then it keeps prompting for a password for 3 times
| |
| 00:32 | <Ryan52> not anymore.
| |
| 00:32 | they changed that at the last hackfest.
| |
| 00:32 | <alkisg> Ah, ok
| |
| 00:55 | qcwp has joined #ltsp | |
| 00:55 | |Paradox| has quit IRC | |
| 00:56 | qcwp is now known as |Paradox| | |
| 01:16 | F-GT has quit IRC | |
| 01:17 | F-GT has joined #ltsp | |
| 01:29 | <vagrantc_> the current behavior with incorrect passwords is a little wonky ... i think it just says "no response from server" and then dies.
| |
| 01:29 | <warren> yes
| |
| 01:30 | <vagrantc_> i seem to recall since the hackfest re-write it behaving kinda ok with multiple password attempts
| |
| 01:31 | <Ryan52> so, vagrantc_ and warren, what do you think about alkisg and my crazy SSH_ASKPASS idea? :)
| |
| 01:32 | rcy has quit IRC | |
| 01:32 | <Ryan52> do you see any reason for me to not just go for it? afaik, everybody hates the expect method :)
| |
| 01:33 | <warren> Ryan52: could you please post the entire idea to the list?
| |
| 01:33 | * Ryan52 nods | |
| 01:34 | <vagrantc_> Ryan52: i think it already uses SSH_ASKPASS ...
| |
| 01:34 | <Ryan52> 0.o
| |
| 01:34 | <vagrantc_> maybe that was the python ldm ...
| |
| 01:34 | <warren> Was Ryan in kindergarten back then? =)
| |
| 01:35 | <Ryan52> recursive grep shows no matches.
| |
| 01:35 | <vagrantc_> yeah. nothing in the current code.
| |
| 01:35 | <Ryan52> warren, probably :)
| |
| 01:35 | does anybody know what the ml's address is? :)
| |
| 01:35 | <vagrantc_> i *think* SSH_ASKPASS is probably a good idea, though i'm curious why it was dropped. i *know* it was implemented at one point.
| |
| 01:36 | * vagrantc_ heads to sleep | |
| 01:36 | <Ryan52> bye
| |
| 01:37 | ah. ltsp-developer@lists.sourceforge.net
| |
| 01:39 | okay, I should sleep too ... tomorrow is the first day back to school after winter break :(
| |
| 01:43 | alkisg has quit IRC | |
| 01:59 | vagrantc_ has quit IRC | |
| 02:03 | Egyptian[Home] has joined #ltsp | |
| 02:08 | dirigeant has quit IRC | |
| 02:14 | leio has quit IRC | |
| 02:20 | leio has joined #ltsp | |
| 02:26 | cyberorg has quit IRC | |
| 02:29 | tjikkun_work has joined #ltsp | |
| 02:37 | durfjmli has joined #ltsp | |
| 02:37 | |Paradox| has quit IRC | |
| 02:38 | durfjmli is now known as |Paradox| | |
| 03:21 | cyberorg has joined #ltsp | |
| 03:42 | mzduyiiuztw has joined #ltsp | |
| 03:42 | |Paradox| has quit IRC | |
| 03:43 | mzduyiiuztw is now known as |Paradox| | |
| 03:45 | mikkel has joined #ltsp | |
| 03:49 | Appiah has joined #ltsp | |
| 04:10 | fnldpjm has joined #ltsp | |
| 04:10 | |Paradox| has quit IRC | |
| 04:11 | fnldpjm is now known as |Paradox| | |
| 04:19 | mraudsepp has joined #ltsp | |
| 04:20 | <mraudsepp> ooh, Gentoo is getting close? :)
| |
| 04:20 | <warren> that topic hasn't been updated in months
| |
| 04:20 | <mraudsepp> anyway, folks with Geode hardware and latest graphics stack around here?
| |
| 04:21 | <warren> mraudsepp: I have both, but will not have access until I fly home in 3 days.
| |
| 04:21 | it was working fine when I tested it
| |
| 04:21 | <Appiah> ;)
| |
| 04:21 | <mraudsepp> wonder if others are seeing freezes on rotation with geode-2.11.0 driver
| |
| 04:21 | <warren> mraudsepp: no
| |
| 04:21 | mraudsepp: what version of X.org server?
| |
| 04:21 | <mraudsepp> 1.5.x or so
| |
| 04:21 | <warren> xorg-x11-server-Xorg-1.5.3-6 here
| |
| 04:22 | <mraudsepp> I think it's 1.5.3 with lots of EXA backports
| |
| 04:22 | including the glyph cache, which speed things up up to 5 times with some drivers and hardware, but slows geode down considerably.
| |
| 04:22 | <warren> I haven't seen any issues, I tested the Geode LX thincan and craptastic uh.... another Geode LX thin client
| |
| 04:22 | <Appiah> I'm troubleshooting a problem with my thin clients not being able to PXE Boot , the DHCP log and wireshark just shows the clients DHCP Discover over and over again. This happends when I start about 15 clients at the same time
| |
| 04:23 | <warren> Appiah: not a bandwidth issue?
| |
| 04:23 | <Appiah> it it can take about 20 restarts for a new client to be able to boot
| |
| 04:23 | warren: I dont think so
| |
| 04:23 | Beacuse I just boot them up
| |
| 04:23 | dont log on or anything
| |
| 04:23 | <mraudsepp> warren: right, ThinCan is my reference too, being an Artec employee. Other high priority item is the EXA Composite gradient rendering bug
| |
| 04:24 | <Appiah> I can the 15 clients at login for a hour , and then try to fire up a 16th client but it just wont get an lease
| |
| 04:24 | no DHPACK
| |
| 04:24 | <mraudsepp> I'll try to update my stack a bit more in the background and get some traces later if it still freezes
| |
| 04:24 | <warren> I don't know anything about the driver
| |
| 04:24 | I only test it
| |
| 04:25 | <mraudsepp> yeah, I guess I will be needing to dig into the code myself at some point
| |
| 04:25 | <nubae> Appiah: have u got remote logging enabled?
| |
| 04:26 | that can eat up your bandwidth and stop the clients getting connections
| |
| 04:26 | <Appiah> I have no idea nubae
| |
| 04:27 | how can I force it off?
| |
| 04:27 | <nubae> u would have done this: https://help.ubuntu.com/community/UbuntuLTSP/TCSyslogToLTSPServer
| |
| 04:28 | make sure /etc/default/syslogd looks right
| |
| 04:28 | <mraudsepp> warren: http://bugs.freedesktop.org/show_bug.cgi?id=15700 was the gradient bug I had in mind of course :)
| |
| 04:29 | <warren> mraudsepp: " I guess I will be needing to dig into the code myself at some point" does not inspire confidence
| |
| 04:29 | <nubae> lol
| |
| 04:29 | <warren> mraudsepp: yes, I'm the one that submitted that bug
| |
| 04:29 | mraudsepp: largely thanks to your company's thincan
| |
| 04:30 | <Appiah> remote logging is not on
| |
| 04:30 | <mraudsepp> warren: yeah, aware you filed it :)
| |
| 04:30 | <nubae> Appiah: dunno then... gigabit network?
| |
| 04:31 | <mraudsepp> warren: the guessing was that if no-one else is doing any geode X work, I will be forced to jump into X.org development that I haven't done before, that's all. The guessing bit is that I have no idea if OLPC folks will care to or not and at what capacity
| |
| 04:31 | <warren> mraudsepp: OLPC let go their only person who was even remotely working on geode
| |
| 04:31 | mraudsepp: AMD's geode team is gone
| |
| 04:32 | mraudsepp: geode upstream is effectively dead now
| |
| 04:32 | <Appiah> nubae: yupp
| |
| 04:32 | well atleast the server is on gigabit
| |
| 04:32 | bit unsure about the client
| |
| 04:32 | <nubae> and the switch?
| |
| 04:33 | <warren> Appiah: it is fine with only one client?
| |
| 04:33 | <mraudsepp> warren: exactly. Might have to become "upstream" for some bits as ThinCan is alive and needing various love for Geode
| |
| 04:33 | <nubae> olpc no longer even pays anyone to work on sugar anymore
| |
| 04:33 | <Appiah> this can happen to all clients
| |
| 04:33 | <warren> nubae: what!?
| |
| 04:33 | <mraudsepp> there's SugarLabs though?
| |
| 04:33 | <Appiah> after it gets it IP, I can restart it and it will get an IP directly
| |
| 04:33 | <mraudsepp> not sure if they pay anything..
| |
| 04:33 | <nubae> yup... thats independent
| |
| 04:33 | <Appiah> I dont think this is a BW issue ...
| |
| 04:33 | <mraudsepp> OLPC dumps windows on stuff? ;d
| |
| 04:33 | <nubae> the last sugar dev stopped being paid in november
| |
| 04:33 | <warren> Appiah: the switch could suck
| |
| 04:34 | Appiah: it is almost always the switch's fault
| |
| 04:34 | <nubae> well, that said, there are only really 200 laptops currently running windows, so its not really about that
| |
| 04:34 | <Appiah> I got one switch here thats pretty close to the server switch
| |
| 04:34 | the other clients are bit further away but they all show the same problem
| |
| 04:35 | <nubae> Appiah: I'd try and see if its a hardware problem first... try a different switch different cables...
| |
| 04:35 | <Appiah> I wonder which switch I should check then? the one the server is connected to?
| |
| 04:35 | <warren> Appiah: did it work on this switch before?
| |
| 04:35 | <Appiah> it is working
| |
| 04:35 | this can happen to any of the 20 clients
| |
| 04:36 | did not have this problem when we were using older hardware , and 8.04 and less clients
| |
| 04:36 | about 15
| |
| 04:37 | what hardware did you have in mind nubae ?
| |
| 04:37 | <nubae> older hardware, u mean older clients?
| |
| 04:37 | <Appiah> server hardware
| |
| 04:37 | was on a poor P4 ;)
| |
| 04:37 | <nubae> hmmm, well thats indeed odd, normally upgrading the server should definitly improve things
| |
| 04:38 | <Appiah> well it's a fresh install on a new server
| |
| 04:38 | and it's the only DHCP server on the network
| |
| 04:39 | <nubae> firewall?
| |
| 04:40 | <Appiah> not on this network
| |
| 04:40 | it's only the ltspserver and the clients
| |
| 04:40 | maybe a cisco phone beacuse I saw some skinny packages on wireshark
| |
| 04:41 | I restarted dhcp3 now and now I got the problem again
| |
| 04:42 | just DHCPDISCOVER DHCPOFFER but no dhpack
| |
| 04:42 | |Paradox| has quit IRC | |
| 04:42 | |Paradox| has joined #ltsp | |
| 04:44 | <nubae> check /var/lib/dhcp3, maybe there's something in there about the leases
| |
| 04:45 | <Appiah> what about them?
| |
| 04:45 | <nubae> well, not sure, but its another place to look
| |
| 04:46 | <Appiah> funny now it boots again
| |
| 04:46 | this is so friggin anoying
| |
| 04:46 | <nubae> your cables are properly grimmped
| |
| 04:46 | I've had problems like that with badly grimped cables
| |
| 04:46 | that and broken switches
| |
| 04:47 | <Appiah> so we got about 20 bad cables
| |
| 04:47 | or a bad switch? :P
| |
| 04:47 | <nubae> nah, but one bad cable can effect the others
| |
| 04:47 | <warren> one bad switch can work for some things but not others
| |
| 04:47 | 95% of the time it is the switches fault
| |
| 04:47 | <Appiah> I tried some dhcp snooping on a client switch but I didnt get anything
| |
| 04:49 | wait a sec
| |
| 04:49 | this client got like 4 leases O_O
| |
| 04:50 | for the same IP
| |
| 04:50 | |Paradox| has quit IRC | |
| 04:50 | <Appiah> could that have to do with anything?
| |
| 04:50 | |Paradox| has joined #ltsp | |
| 04:51 | otavio has joined #ltsp | |
| 04:53 | sepski has joined #ltsp | |
| 04:54 | <mraudsepp> warren: while we were on the geode topic, feel free to give me a list with stuff that needs hacking on for geode, so I can have a better technical overview of what's the status of stuff. Soon we need to make some decisions on what to work on at company level, and it helps having a good list of additional stuff (I have lost of coreboot and few Xorg stuff only in my list right now). That goes for anyone using geode :)
| |
| 04:54 | <warren> mraudsepp: well, the only issue I've had personally was that EXA composite thing
| |
| 04:55 | mraudsepp: I haven't had issues with coreboot ever since we wrote the intelligent patch for mkelfimage
| |
| 04:56 | mraudsepp: IMHO, the EXA issue and making it run stable should be done quickly, and then your company should just move on. Geode is near dead now. Too slow and obsolete, with much better stuff becoming lower power and cheaper.
| |
| 05:10 | <Appiah> I found 2 things that are suspicious about dhcpd3 (i'm really trying evertything here)
| |
| 05:10 | it writes multiple leases for the same client (tried a new clean lease file)
| |
| 05:10 | and the time seams to be wrong
| |
| 05:11 | it says the client got a lease 11:07 but its 12:07
| |
| 05:11 | system clock says 12:07 too
| |
| 05:16 | cyberorg_ has joined #ltsp | |
| 05:18 | cyberorg has quit IRC | |
| 05:18 | cyberorg_ is now known as cyberorg | |
| 05:20 | <mraudsepp> warren: the too slow part is something that could be improved in Xorg, tbh. But there's only so much you can do
| |
| 05:21 | <warren> mraudsepp: running it side by side with Athlon + ATI thin clients in Japan, it was noticably uglier. lower bit depth and slower to update screen...
| |
| 05:22 | <mraudsepp> warren: if it wouldn't software fallback to cause pixmap migration ping-pong, I bet it would be a lot better as far as screen updates go
| |
| 05:22 | warren: glyph drawing is a joke with modern Xorg, for instance
| |
| 05:23 | <warren> mraudsepp: I don't know the details, I only prodded upstream geode into being a public project and submitted testing data
| |
| 05:23 | <mraudsepp> due to no A8 pixmap support. Solution should be simple enough - make EXA glyph stuff use ARGB instead of A8 when driver requests. For any other A8 cases you could do hardware conversion from A8 to ARGB - still should be better than software fallback and the related migration costs
| |
| 05:23 | <warren> mraudsepp: even if these are software limitations to squeeze out a little more life out of geode hardware, realistically the manpower to fix it and keep it working...
| |
| 05:24 | <mraudsepp> yeah, there's so much you can do
| |
| 05:25 | though I'm thinking to do some from home. There's low hanging fruit around the geode X.org stuff to get into X.org development more for OSS development stuff. And doing the important stuff from company time, the rest home to get experience to hack on the high hanging fruit radeon/intel stuff for my own machines benefits
| |
| 05:26 | <warren> ok, well I'm interested in testing patches
| |
| 05:26 | I can turnaround test data on two different LX hardware quickly (if I'm home)
| |
| 05:26 | <mraudsepp> I'm into improving GNOME performance stuff for everything. For that I need a slow machine to test what is slow - ThinCan sounds perfect, but not if performance profiles are full of X.org level stuff
| |
| 05:27 | <warren> what is your role at Artec?
| |
| 05:29 | <mraudsepp> software engineer
| |
| 05:29 | I've been ending up working on everything we have non-ARM, mostly everything ThinCan related.
| |
| 05:50 | alkisg has joined #ltsp | |
| 05:52 | <Appiah> I'm thinking , either the DHCPoffer dont reach the client (the server does send it and writes a lease) or the DHCPRequest dont reach the server
| |
| 05:59 | elisboa has joined #ltsp | |
| 06:03 | ccherret1 has joined #ltsp | |
| 06:04 | ccherrett has quit IRC | |
| 06:11 | <mraudsepp> warren: for coreboot, there's just lots of stuff possible to do to get it viable for more scenarios, and all that in upstream code in coreboot-v3. After I fix a NAND setup problem there (hopefully this week), v3 will work better than our internal branch whose binaries you probably have, when not counting the lack of VGA BIOS there. So beyond that there's all the other stuff necessary for other use cases than X.org with LTSP - USB boot, free VGA BIOS
| |
| 06:11 | with cb-v3, BIOS interrupts (SeaBIOS integration), PXE, USB support (keyboard at boot time), setup menu and so on and on
| |
| 06:11 | mikkel has quit IRC | |
| 06:12 | <warren> I'm not here.
| |
| 06:12 | or my mind isn't here
| |
| 06:12 | <ogra> your fingers are though ...
| |
| 06:12 | happy new year #ltsp
| |
| 06:19 | <cyberorg> happy new year to you too ogra :)
| |
| 06:21 | <Appiah> So I've still got a problem with thin clints PXE booting , it apears it has nothing to do with the amount of clients on.
| |
| 06:21 | DHCPdiscover -> DHCPOffer then it stops there
| |
| 06:21 | from sniffing in wireshark and DHCP log
| |
| 06:21 | this happens "sometimes"
| |
| 06:21 | you can reboot the client once or 15 times and it will work again
| |
| 06:22 | worked fine in 8.04 , this came in 8.10
| |
| 06:22 | <nubae> Appiah: have u tried changing switches?
| |
| 06:22 | <cyberorg> Appiah, also check number of nbdrood processes running
| |
| 06:23 | nubae has left #ltsp | |
| 06:25 | dxo has joined #ltsp | |
| 06:25 | |Paradox| has quit IRC | |
| 06:26 | dxo is now known as |Paradox| | |
| 06:28 | nubae has joined #ltsp | |
| 06:41 | <alkisg> Appiah, try with wireshark on the client side. Of course not on the client itself, but with a pc connected to the client switch.
| |
| 06:46 | <Appiah> with my laptop connected to the switch now I can only see broadcasted stuff
| |
| 06:46 | DHCP offer and DHCP Ack
| |
| 06:47 | <alkisg> So there is a DHCP ACK?
| |
| 06:48 | <Appiah> for one that's working yes ;)
| |
| 06:48 | <alkisg> OK, try another one! :)
| |
| 06:48 | And keep the log, to upload it to pastebot
| |
| 06:49 | <Appiah> I'm also getting a HUB
| |
| 06:52 | Faithful has joined #ltsp | |
| 06:54 | wgyztjsvqqv has joined #ltsp | |
| 06:54 | |Paradox| has quit IRC | |
| 06:55 | wgyztjsvqqv is now known as |Paradox| | |
| 06:55 | <Appiah> well here is one that's not working
| |
| 06:55 | BrunoXLambert has joined #ltsp | |
| 06:56 | <Appiah> http://pastebin.ca/1300073
| |
| 06:56 | that's all
| |
| 06:56 | gentgeen__ has joined #ltsp | |
| 06:56 | <Appiah> discover/offer all the time but nothing happens
| |
| 06:57 | wireshark gives me all the offers
| |
| 06:58 | <alkisg> and it's plain pxe? no floppy, etherboot, gpxe etc?
| |
| 06:58 | <Appiah> plain pxe
| |
| 06:58 | if thats what HP thin clients use
| |
| 07:00 | Blinny has joined #ltsp | |
| 07:00 | <alkisg> Do you have a wireshark log, with timings etc?
| |
| 07:03 | ccherrett has joined #ltsp | |
| 07:04 | ccherret1 has quit IRC | |
| 07:06 | <alkisg> ogra, do you remember why SSH_ASKPASS, ldm-askpass etc were dropped from ldm? I thought it was a good idea... :-/
| |
| 07:06 | <ogra> i think that was dropped by sbalneav while he did the new chat protocol
| |
| 07:07 | <alkisg> ...and Ryan52 was interested to re-implement it (we didn't know that it was used sometime in the past, vagrantc told us)
| |
| 07:07 | pscheie has quit IRC | |
| 07:07 | <ogra> ldm chats directly to ssh now
| |
| 07:07 | that would break a lot i think, talk to sbalneav
| |
| 07:07 | <alkisg> OK, thanks
| |
| 07:07 | <ogra> Ryan52 was in portland at the hackfest wheer scottie made the change
| |
| 07:08 | <alkisg> I just thought it would be cleaner and get rid of some big delays
| |
| 07:09 | R4f4 has joined #ltsp | |
| 07:09 | <Appiah> well alkisg it says pretty much the same , I see no differnce in the packages that I get
| |
| 07:09 | <alkisg> Appiah, the timings are important
| |
| 07:09 | <Appiah> what?
| |
| 07:10 | <alkisg> If the client broadcasts a new dhcp_request right after the dhcp_offer, then the client thinks it's not suitable. If it's much later, it doesn't receive the offer.
| |
| 07:11 | <Appiah> ok
| |
| 07:11 | We are gonna try another switch
| |
| 07:11 | then I'll look more on the time
| |
| 07:12 | <alkisg> Appiah, also, is there a boot filename in this offer packet?
| |
| 07:12 | <Appiah> root path =
| |
| 07:12 | /opt/ltsp/i386
| |
| 07:13 | <alkisg> No, not root path, boot filename
| |
| 07:13 | <Appiah> ah
| |
| 07:13 | yes there is one
| |
| 07:17 | kind of sucks when the clients starts working again
| |
| 07:28 | cliebow has joined #ltsp | |
| 07:30 | hanthana has joined #ltsp | |
| 07:40 | RobertMLaptop has quit IRC | |
| 07:42 | RobertLaptop has joined #ltsp | |
| 07:55 | six2one has joined #ltsp | |
| 08:06 | |Paradox| has quit IRC | |
| 08:07 | |Paradox| has joined #ltsp | |
| 08:16 | <rjune> ogra, how was vacation?
| |
| 08:19 | pmatulis has joined #ltsp | |
| 08:19 | pscheie has joined #ltsp | |
| 08:27 | topslakr has quit IRC | |
| 08:28 | mikkel has joined #ltsp | |
| 08:32 | topslakr has joined #ltsp | |
| 08:33 | likuidkewl has joined #ltsp | |
| 08:33 | likuidkewl is now known as dmaran | |
| 08:34 | dirigeant has joined #ltsp | |
| 08:46 | CAN-o-SPAM has joined #ltsp | |
| 08:47 | rootchick has joined #ltsp | |
| 08:47 | rootchick has left #ltsp | |
| 08:47 | rootchick has joined #ltsp | |
| 08:53 | Gadi has joined #ltsp | |
| 08:54 | six2one has quit IRC | |
| 08:56 | six2one has joined #ltsp | |
| 09:03 | pmatulis has quit IRC | |
| 09:05 | pmatulis has joined #ltsp | |
| 09:06 | <CAN-o-SPAM> warren, ping
| |
| 09:21 | dirigeant has quit IRC | |
| 09:39 | <Appiah> alkisg: the boot filename is in the offer packet
| |
| 09:39 | it still does not boot
| |
| 09:39 | We tried a new switch (cisco 29xx) now none of the 20 clients boot
| |
| 09:40 | <alkisg> OK, maybe you have to compare the packets when it doesn't boot and when it does boot. It's pretty weird. Did you try with a hub?
| |
| 09:42 | CAN-o-SPAM has quit IRC | |
| 09:43 | <Appiah> tried in a hub
| |
| 09:44 | didnt see a diffrence in the packages between when it works and when it does
| |
| 09:44 | atleast not in the offer
| |
| 10:10 | alekibango has quit IRC | |
| 10:11 | CAN-o-SPAM has joined #ltsp | |
| 10:12 | Ahmuck has joined #ltsp | |
| 10:18 | <Ryan52> ogra, well, ldm would still run ssh ... then the SSH_ASKPASS program would come back and talk to ldm again.
| |
| 10:18 | so it's probably not like it was before. Though I dunno how it worked before..
| |
| 10:32 | topslakr has quit IRC | |
| 10:37 | topslakr has joined #ltsp | |
| 10:41 | <Gadi> Ryan52: we moved away from SSH_ASKPASS due to passwod expiry handling iirc
| |
| 10:41 | (SSH_ASKPASS does not support it)
| |
| 10:43 | R4f4 has quit IRC | |
| 10:50 | tjikkun_work has quit IRC | |
| 10:56 | vagrantc has joined #ltsp | |
| 10:56 | dberkholz_ is now known as dberkholz | |
| 10:56 | spectra has joined #ltsp | |
| 10:57 | CrypTom has joined #ltsp | |
| 10:57 | mraudsepp has quit IRC | |
| 10:58 | <CrypTom> Hi, I've installed Kubuntu 8.10 and LTSP. New users get GNOME, when they log in for the first time, but I would like to get them KDE by default. How can I change this?
| |
| 10:58 | * vagrantc wonders how many professional artists really would be motivated by winning a thin client ... | |
| 10:59 | * CAN-o-SPAM is determined to find out | |
| 11:00 | <vagrantc> :)
| |
| 11:00 | staffencasa has joined #ltsp | |
| 11:04 | vvinet has joined #ltsp | |
| 11:06 | <cliebow> CrypTom, you could install kde-desktop..
| |
| 11:08 | <CrypTom> cliebow: you mean "kubuntu-desktop", that is already installed!
| |
| 11:09 | but I also installed ubuntu-desktop to provide a choice for my pupils
| |
| 11:09 | but I would like them to get KDE by default
| |
| 11:12 | <vagrantc> CrypTom: sudo update-alternatives --config x-session-manager
| |
| 11:12 | CrypTom: or maybe x-desktop-environment
| |
| 11:12 | <CrypTom> vagrantc: im chroot (/opt/ltsp/) oder normal?
| |
| 11:12 | <vagrantc> CrypTom: on the server
| |
| 11:14 | <cliebow> not in chroot...
| |
| 11:19 | <CrypTom> vagrantc: thanks, it was x-session-manager!
| |
| 11:20 | <vagrantc> CrypTom: good good :)
| |
| 11:21 | <CrypTom> do I need a special setting in order to make USB-Sticks on the thin-clients work?
| |
| 11:21 | I dont see any reaction, when I plug in a stick
| |
| 11:22 | <vagrantc> CrypTom: there should be mounts in /media/USERNAME
| |
| 11:22 | CrypTom: i don't think KDE registers them by default
| |
| 11:23 | CrypTom: there is a bit of a hackish script to add them to the user's Desktop dir...
| |
| 11:23 | * vagrantc tries to remember how to enable it | |
| 11:23 | <CrypTom> vagrantc: /media would be fine, but they dont appear
| |
| 11:24 | <vagrantc> CrypTom: user in the fuse group?
| |
| 11:24 | * CrypTom will check | |
| 11:26 | <CrypTom> vagrantc: cool, it works, the user was not in the group...
| |
| 11:26 | vagrantc: thanks again!
| |
| 11:27 | <vagrantc> CrypTom: look in /usr/share/doc/ltspfs/examples for a script to add the icons to the desktop
| |
| 11:28 | <ogra> Ryan52, well, i thought there were issues with the error handling and password update if you use SSH_ASKPASS
| |
| 11:28 | <vagrantc> ah, yes.
| |
| 11:28 | <ogra> but only scott can really answer this, it was his change
| |
| 11:29 | <vagrantc> SSH_ASKPASS doesn't handle the password expiry stuff at all
| |
| 11:29 | <ogra> yeah
| |
| 11:29 | <vagrantc> so you'd have to do screen scraping/passthrough anyways
| |
| 11:29 | <ogra> which is what his change was supposed to avoid
| |
| 11:30 | wow, i just had to get a 2G Sd card (not SDHC) for some special HW that can only handle SD and only up to 2G ... these things are three times as expensive as the 4 or 8G ones nowadays
| |
| 11:31 | <vagrantc> well, it basically does a tiny bit of screen scraping, and mostly passthrough
| |
| 11:31 | <CrypTom> vagrantc: cool, the icon stuff works, but it show an ugly ? instead of a nice icon, can that be changed easily?
| |
| 11:31 | <vagrantc> CrypTom: you'd have to edit the script
| |
| 11:32 | <johnny> will kde ever do the right thing here?
| |
| 11:32 | or whos fault is it?
| |
| 11:32 | <vagrantc> CrypTom: look at all the lines referencing icon
| |
| 11:32 | cor3 has joined #ltsp | |
| 11:32 | <cor3> Nice work on the ubuntu integration
| |
| 11:32 | <vagrantc> CrypTom: find a kde icon name that you prefer, and let us know what to change it to
| |
| 11:33 | <cor3> its become my 1st choice in os for ltsp deployments
| |
| 11:33 | <CrypTom> I see "if dev.startswith('usb'): icon = 'usbpendrive_mount'
| |
| 11:34 | so, do I have to install some icon-theme package to get usbpendrive_mount icon?
| |
| 11:34 | <vagrantc> no idea
| |
| 11:35 | CrypTom: you either need to install the icons it's referencing, or change the references to match what's installed on your system
| |
| 11:35 | * vagrantc is no KDE fan | |
| 11:36 | <vagrantc> johnny: i think it was designed on GNOME, and nobody has ever proposed a clean way to get it working on KDE. it may require some changes in KDE to get it to work right.
| |
| 11:37 | <cor3> xfce all the way
| |
| 11:37 | <johnny> sure.. but suprising that the kde people themselves haven't wanted to do it right
| |
| 11:37 | perhaps nobody has ever talked to them about it? or what?
| |
| 11:38 | <CrypTom> vagrantc: It works with: icon='drive-removable-media-usb-pendrive'
| |
| 11:38 | <vagrantc> XFCE has the same issue as KDE, last i looked.
| |
| 11:39 | CrypTom: what version of KDE do you have?
| |
| 11:41 | <CrypTom> vagrantc: kubuntu 8.10 with kde 4.
| |
| 11:42 | <vagrantc> i suspect KDE4 changed the icon names or something like that
| |
| 11:42 | as the script was written on kde3.x, i'm guessing
| |
| 11:43 | <CrypTom> ok
| |
| 11:43 | dirigeant has joined #ltsp | |
| 11:49 | <CrypTom> vagrantc: thanks and bye
| |
| 11:50 | CrypTom has quit IRC | |
| 11:51 | topslakr_ has joined #ltsp | |
| 11:58 | cor3 has left #ltsp | |
| 11:59 | topslakr has quit IRC | |
| 12:02 | vvinet has quit IRC | |
| 12:12 | <_UsUrPeR_> In Ubuntu 8.04, what needs to be changed in order to get USB flash drives to appear on client desktops?
| |
| 12:12 | is that in the udev rules someplace?
| |
| 12:12 | <Gadi> you just need the user to be in the fuse group
| |
| 12:13 | <_UsUrPeR_> gadi: ok, thank you
| |
| 12:15 | gadi: is there a difference between Flash and a USB hdd?
| |
| 12:15 | <Gadi> many - one has 5 letters the other has 3, ....
| |
| 12:16 | <_UsUrPeR_> :)
| |
| 12:16 | <Gadi> a USB hdd may not be considered "removable"
| |
| 12:16 | by udev
| |
| 12:16 | you should check
| |
| 12:17 | <_UsUrPeR_> just made an 8.04 install. It appears that, though my user is in the fuse group, neither a flash drive or USB hdd is working properly.
| |
| 12:17 | <Gadi> other than that.... if it has an external power supply, use it
| |
| 12:17 | <_UsUrPeR_> ahh.
| |
| 12:17 | hmm~~~
| |
| 12:18 | is that in 60-persistent-storage.rules?
| |
| 12:19 | <Gadi> udevinfo -qenv -n ${DEVICENAME}
| |
| 12:19 | on the thin client
| |
| 12:19 | where $DEVICENAME is something like sda
| |
| 12:19 | or sda1
| |
| 12:20 | <_UsUrPeR_> k
| |
| 12:21 | artista_frustrad has joined #ltsp | |
| 12:22 | hanthana has quit IRC | |
| 12:24 | CAN-o-SPAM has quit IRC | |
| 12:34 | Egyptian[Home] has quit IRC | |
| 12:34 | Egyptian[Home] has joined #ltsp | |
| 12:39 | <rootchick> hello all...is there a trick to getting ldm rc.d scripts to work when you're using LDM_DIRECTX=True?
| |
| 12:40 | CAN-o-SPAM has joined #ltsp | |
| 12:46 | Lns has joined #ltsp | |
| 12:47 | dirigeant has quit IRC | |
| 12:47 | dirigeant has joined #ltsp | |
| 12:50 | <vagrantc> rootchick: they should just work
| |
| 13:24 | nubae has quit IRC | |
| 13:27 | Tetrispt has joined #ltsp | |
| 13:27 | <Tetrispt> hello
| |
| 13:28 | anyone knows where i can donwload ltsp for RHEL 5 (.rpm) ?
| |
| 13:28 | thanks
| |
| 13:36 | <Gadi> Tetrispt: I dont think LTSP5 is made for RHEL5 yet - I think you need to use LTSP4.2
| |
| 13:37 | Tetrispt: http://wiki.ltsp.org/twiki/bin/view/Ltsp/LTSP-42#Installation_of_LTSP_4_2
| |
| 13:39 | leio_ has joined #ltsp | |
| 13:44 | <Tetrispt> Gadi: i found this link - https://fedorahosted.org/k12linux/wiki/RHEL5Server
| |
| 13:45 | Gadi: it seems it works...
| |
| 13:45 | i will test it :)
| |
| 13:45 | <warren> Tetrispt: very old
| |
| 13:46 | Tetrispt: it might work, but very old and unmaintained
| |
| 13:46 | Tetrispt: working on real RHEL5 support but will take time
| |
| 13:46 | Eghie has joined #ltsp | |
| 13:46 | <Gadi> Tetrispt: meet warren, the guy responsible for all things LTSP and fedora/RHEL :)
| |
| 13:52 | leio has quit IRC | |
| 13:56 | <Tetrispt> hi warren
| |
| 13:56 | so you recomend me to install LTSP 4.2 ?
| |
| 13:59 | leio_ has quit IRC | |
| 13:59 | leio__ has joined #ltsp | |
| 14:02 | openstep has joined #ltsp | |
| 14:03 | <openstep> hi
| |
| 14:03 | anyone here?
| |
| 14:04 | leio__ is now known as leio | |
| 14:04 | <rjune> openstep, nobody here but us chickens
| |
| 14:05 | <openstep> i need som help
| |
| 14:05 | :)
| |
| 14:06 | * Gadi thinks of the Beatles song | |
| 14:06 | <openstep> I need to finish tomorrow with an ltsp deployment
| |
| 14:07 | and my problem is I can not set the default session on the cleints
| |
| 14:07 | <Gadi> set it to what?
| |
| 14:07 | <openstep> xfce
| |
| 14:07 | <Gadi> sudo update-alternatives --config x-session-manager
| |
| 14:07 | (on ubuntu)
| |
| 14:08 | <openstep> yes it is ubuntu
| |
| 14:08 | wow
| |
| 14:08 | so how would it know what session do I want?
| |
| 14:08 | <warren> Tetrispt: no
| |
| 14:09 | Tetrispt: I would recommend using Fedora 10 with LTSP5, and upgrade into RHEL6 when it becomes available later.
| |
| 14:09 | Tetrispt: http://alt.fedoraproject.org/pub/alt/ltsp/k12linux/f10/rc1/ release candidate of Fedora 10 media
| |
| 14:10 | Tetrispt: burn to DVD or install on a usb stick for usb boot
| |
| 14:10 | bobby_C has joined #ltsp | |
| 14:10 | <openstep> Gadi: so what is this command?
| |
| 14:10 | <Tetrispt> warren: i will test it. thanks for the help.
| |
| 14:12 | Blinny has quit IRC | |
| 14:12 | <Gadi> openstep: what happened when you did the above?
| |
| 14:12 | <openstep> well, I am not at the server now
| |
| 14:12 | <Gadi> ah, well do that
| |
| 14:12 | <openstep> I need to find the soultion and accopmplish tomorrow
| |
| 14:13 | <Gadi> you should be able to update the x-session-manager alternative to point to xfce
| |
| 14:14 | <openstep> so this command will sync the clients x-session manager on /opt/... with the one on the server????
| |
| 14:14 | <CAN-o-SPAM> warren: do i need to join fedora-art-mailing list to post, or can you bypass that for me?
| |
| 14:15 | pmatulis_ has joined #ltsp | |
| 14:15 | <warren> CAN-o-SPAM: I will not get involved (you need to join)
| |
| 14:15 | <Gadi> the client's default session is the default session on the server
| |
| 14:15 | <cliebow> openstep, what Gadi says
| |
| 14:15 | <Gadi> this merely sets the default session on the server
| |
| 14:15 | <cliebow> Gadi Rocks!
| |
| 14:15 | <warren> CAN-o-SPAM: did you keep my earlier recommendation? (url to how it looks like in fedora)
| |
| 14:16 | <Gadi> cliebow: you're better than those confidence-building cassette tapes
| |
| 14:16 | :)
| |
| 14:16 | <CAN-o-SPAM> yea, i tried to send, it said not-a-member
| |
| 14:16 | <openstep> Gadi: is that means that when I sit down by the server and set the default session I have to run this command to set it on the cleint???
| |
| 14:16 | <CAN-o-SPAM> guess i'll join up
| |
| 14:17 | <Gadi> openstep: right - run this command on the server and it will change the server's system-wide default setting for X session
| |
| 14:17 | the client just goes by the server system default
| |
| 14:17 | <cliebow> .your default session is what evr it is on the server
| |
| 14:17 | <Gadi> unless it chooses something else
| |
| 14:17 | <cliebow> openstep, what Gadi says
| |
| 14:18 | <Gadi> openstep: if you are running ubuntu right now, go ahead and run the command from a terminal
| |
| 14:18 | it won't bite
| |
| 14:18 | <openstep> because right bow when I set xfce4 on the server to be default...then log out and back in it stays default but not on the clients
| |
| 14:18 | <Gadi> :)
| |
| 14:18 | <cliebow> ANd i look even better than those Cowboy cheerleaders
| |
| 14:19 | <Gadi> openstep: well, you prolly set it as default for a gdm session
| |
| 14:19 | not an ldm session
| |
| 14:19 | <openstep> well, I am not at the server now
| |
| 14:19 | <Gadi> ldm goes by the x-session-manager alternatives default setting
| |
| 14:19 | openstep: are you on an ubuntu laptop?
| |
| 14:19 | PC?
| |
| 14:20 | doesnt have to be the server
| |
| 14:20 | <openstep> it is an IBM x226
| |
| 14:20 | in a school
| |
| 14:20 | <Gadi> running ubuntu?
| |
| 14:20 | <openstep> yes
| |
| 14:20 | <Gadi> that you are on right now?
| |
| 14:20 | <openstep> nowww
| |
| 14:20 | <Gadi> pop open a terminal window
| |
| 14:20 | <openstep> nope sorry
| |
| 14:20 | <Gadi> and run the command
| |
| 14:21 | oh
| |
| 14:21 | <openstep> I am not at the server now, but tomorrow
| |
| 14:21 | <Gadi> ok
| |
| 14:21 | yeah, this command doesnt work on Windows ;)
| |
| 14:21 | <openstep> just need the info so that in the morning I can be done before the staff and students arrive
| |
| 14:22 | <Ahmuck> stgraber:
| |
| 14:22 | <openstep> Gadi: thanks, and jut let me repeat once to make sure I understand it
| |
| 14:22 | <stgraber> Ahmuck: busy, sorry. May be available tonight.
| |
| 14:22 | <Gadi> openstep: go for it
| |
| 14:23 | <Ahmuck> stgraber: k
| |
| 14:23 | <openstep> so when I would like to set a default session (like xfce) I do so on the server's login manager. This does not take any effect on the clients until....
| |
| 14:24 | this command:
| |
| 14:24 | sudo update-alternatives --config x-session-manager
| |
| 14:24 | pmatulis has quit IRC | |
| 14:24 | <Gadi> A+ for execution, A- for understanding the theory
| |
| 14:24 | lets bolster the theory:
| |
| 14:24 | <openstep> then every time I would like something different to be default I should repeat
| |
| 14:25 | <Gadi> the only login manager you care about is the client's login manager which is "ldm"
| |
| 14:25 | (LTSP Display Manager)
| |
| 14:25 | the server's (in your case, gdm) plays no role here
| |
| 14:25 | LDM will use whatever the server's system-wide default X session is
| |
| 14:26 | (which need not be the same that you chose for your user in gdm)
| |
| 14:26 | To change the system-wide X session default, you run that command you quoted
| |
| 14:26 | What you do when you login at the server through gdm is between you and your Maker
| |
| 14:26 | :)
| |
| 14:27 | oh, and: Pete and RePete were on a boat. Pete falls out. WHo is left?
| |
| 14:27 | <openstep> without that command it is only applys to gdm on the server
| |
| 14:27 | <Gadi> openstep: gdm is only affected by whatever you set when you logged in thru gdm
| |
| 14:27 | but that does not affect the clients in the least
| |
| 14:27 | only your local gdm login when sitting at the server
| |
| 14:28 | <openstep> yes and here comes this command, which somehow reads gdm's default
| |
| 14:28 | <Gadi> nope
| |
| 14:28 | <openstep> and applys it to be system wide
| |
| 14:28 | <Gadi> this command simply sets the system-wide X session default
| |
| 14:28 | it will ask you what to set it to
| |
| 14:28 | it wont read it from anywhere
| |
| 14:28 | <openstep> oh, t
| |
| 14:28 | <Gadi> and if you started with stock ubuntu, the default will be gnome
| |
| 14:29 | <openstep> that part i did not understand
| |
| 14:29 | <Gadi> ah, my bad
| |
| 14:29 | * Gadi always misses a step | |
| 14:29 | <Gadi> or an openstep, as the case may be
| |
| 14:29 | now: Pete and RePete were on a boat. Pete falls out. WHo is left?
| |
| 14:29 | :)
| |
| 14:30 | <openstep> then sorry for asking one more
| |
| 14:30 | <Gadi> go for it
| |
| 14:31 | <openstep> running that command will ask me to type in my preferred session, right?
| |
| 14:32 | <Gadi> something like that - it should be interactive
| |
| 14:32 | <openstep> so for example in the case of xfce4 I just type in "xfce4"
| |
| 14:32 | oh, many thanks
| |
| 14:32 | <Gadi> if it isn't, come back here and vent
| |
| 14:32 | we'll vent with you
| |
| 14:32 | :)
| |
| 14:32 | we like to vent
| |
| 14:32 | <openstep> :) thanks for your patience
| |
| 14:32 | <Gadi> so formal...
| |
| 14:33 | :)
| |
| 14:33 | anytime
| |
| 14:33 | Appiah_ has joined #ltsp | |
| 14:33 | <Gadi> you still havent told me who is left
| |
| 14:33 | :)
| |
| 14:33 | <openstep> just a sec
| |
| 14:34 | :)
| |
| 14:34 | RePete
| |
| 14:34 | <Gadi> Pete and RePete were on a boat. Pete falls out. WHo is left?
| |
| 14:34 | * Gadi couldnt resist | |
| 14:34 | Appiah has quit IRC | |
| 14:34 | <openstep> hehe
| |
| 14:34 | ...
| |
| 14:34 | is there a tool to set default session in the menu?
| |
| 14:35 | <Gadi> which menu?
| |
| 14:35 | on the LDM login screen?
| |
| 14:36 | <openstep> oh, never mind
| |
| 14:36 | try it tomorrow
| |
| 14:36 | I am confused
| |
| 14:38 | thanks I need to go
| |
| 14:38 | openstep has quit IRC | |
| 14:56 | spectra has quit IRC | |
| 14:58 | warren has quit IRC | |
| 14:58 | <cliebow> chuck is still on the boat..though he looks like he just got off it
| |
| 15:03 | dmaran has left #ltsp | |
| 15:10 | Faithful has quit IRC | |
| 15:10 | cliebow has quit IRC | |
| 15:25 | <rjune> !g
| |
| 15:25 | <ltspbot> rjune: "g" is Gadi!!!!!!!!!!!!!!!!!!!!!!!!
| |
| 15:44 | pmatulis_ has quit IRC | |
| 15:46 | Mip5 has joined #ltsp | |
| 15:50 | <Mip5> I've got a few Dell machines that will pxe boot, and appear to load initrd.img, go to the splash screen, but not present a prompt. When I disable splash and remove quiet, it appears to get to Starting LTSP client, but then scrolls by too fast to see where it's hanging.
| |
| 15:51 | I'm running edubuntu hardy
| |
| 15:52 | Additionally, if I tail -f /var/log/syslog, I see multiple connections from the client in ldminfod
| |
| 15:54 | If I hit ctrl-alt-f1, I'm presented with a blinking login prompt, but I can't log in.
| |
| 16:00 | johnny has left #ltsp | |
| 16:01 | BrunoXLambert has quit IRC | |
| 16:01 | rootchick has left #ltsp | |
| 16:01 | CAN-o-SPAM has quit IRC | |
| 16:06 | vvinet has joined #ltsp | |
| 16:12 | Gadi has left #ltsp | |
| 16:20 | EAG has joined #ltsp | |
| 16:29 | <Mip5> I've created a unique pxeconfig for this machine, and it's pulling it correctly. Are there features I can disable to make facilitate the booting?
| |
| 16:33 | <alkisg> Mip5: what's the last thing you see in the client screen before it crashes?
| |
| 16:33 | six2one has quit IRC | |
| 16:33 | LinusU has joined #ltsp | |
| 16:34 | <LinusU> any good sites for local apps setup on ubuntu 8.10?
| |
| 16:37 | warren__ has joined #ltsp | |
| 16:39 | warren__ is now known as warren | |
| 16:39 | <Mip5> alkisg: Starting LTSP client - but it scrolls by pretty quickly
| |
| 16:40 | <alkisg> And then what, blank screen?
| |
| 16:41 | <EAG> does anyone else have problems with getting "nautilus scripts" to show in the context menu when right clicking?
| |
| 16:41 | in a nautilus-window..
| |
| 16:41 | <Mip5> Yes - with a blinking cursor in teh upper lefthand corner
| |
| 16:41 | Eghie has quit IRC | |
| 16:42 | <alkisg> So maybe it's X-related? What graphics card do the clients have?
| |
| 16:43 | <Mip5> I wondered about that too.
| |
| 16:44 | They are intel boards on Dell OptiplexGX270s. I'm not sure of the card (it's integrated into the mobo)
| |
| 16:44 | <alkisg> Maybe you could try setting X_SERVER=vesa in lts.conf
| |
| 16:45 | !lts.conf
| |
| 16:45 | <ltspbot> alkisg: "lts.conf" is http://wiki.ltsp.org/twiki/bin/view/Ltsp/LtsConf
| |
| 16:45 | <Mip5> okay - I have other clients that working fine (different hardware)
| |
| 16:46 | Can i specify that only this class of clients use the vesa in lts.conf?
| |
| 16:46 | <alkisg> Yes, either for each one based on hostname/ip/mac, or with a LIKE stanza
| |
| 16:46 | See the url above
| |
| 16:46 | <Mip5> okay - I'll dig into that. Thanks a bunch
| |
| 16:46 | <alkisg> np
| |
| 16:49 | <Mip5> okay - well I see that I had already set that previously (though I did it for all clients).
| |
| 16:51 | <alkisg> Then maybe you should disable it! google says you have intel graphics card, which are supposedly supported.
| |
| 16:52 | You could also try to get a shell in the client and run X -configure and read the logs
| |
| 16:52 | i.e. put SCREEN_02=shell in lts.conf
| |
| 16:54 | <Mip5> Wow - that's cool
| |
| 16:54 | ogra has quit IRC | |
| 16:55 | <Mip5> I'll try disabling vesa first, and then will try to add the SCREEN_02 line to the stanza for this machine. brb
| |
| 16:55 | ogra has joined #ltsp | |
| 16:57 | johnny has joined #ltsp | |
| 16:57 | <Mip5> killer! I've got a root prompt
| |
| 16:58 | <alkisg> Heh! So run an X -configure :0 and then check at /var/log/X.?.log
| |
| 17:01 | <Mip5> okay - so the log was very short - didn't look like typical Xlog to me.
| |
| 17:03 | <alkisg> There are two file there I think, look at the other one
| |
| 17:03 | Xorg.0.log and Xorg.7.log or something
| |
| 17:03 | <Mip5> I then tested it with X -config //xorg.conf.new and got the gray screen with an X (that moved via the mouse)
| |
| 17:04 | <alkisg> Ah, right...
| |
| 17:04 | <Mip5> Sorry - I don't really know much about this stuff (as you can probably tell!)
| |
| 17:04 | <alkisg> So one way to solve this would be to ssh this xorg.conf.new to the server and use it with X_CONF=xxx in lts.conf
| |
| 17:05 | (but you'll also have to copy it into the chroot and run ltsp-update-image)
| |
| 17:05 | <Mip5> I'm not sure I follow - are you saying to copy this xorg.conf.new to the stanza in lts.conf that defines the params for this machine?
| |
| 17:05 | <alkisg> Yes, but first scp it to the server
| |
| 17:06 | scp /root/xorg.conf.new username@server:...
| |
| 17:06 | <Mip5> okay - I can't seem to get back to my shell - though I could reboot the beast
| |
| 17:06 | I'd reboot the client -not the server
| |
| 17:06 | <alkisg> You'd loose the xorg.conf then
| |
| 17:06 | <Mip5> doh!
| |
| 17:06 | Okay - I haven't done it yet
| |
| 17:06 | <alkisg> What if you press ctrl+alt+back space?
| |
| 17:07 | (as you can see I'm not really experienced either)
| |
| 17:07 | <Mip5> I get ^[^H
| |
| 17:07 | but it doesn't restart x
| |
| 17:07 | <alkisg> And with Ctrl+Alt+F1 ?
| |
| 17:08 | <Mip5> Well - I get a login prompt, but I can't login with the credentials I would expect to work...
| |
| 17:08 | <alkisg> Yes, you'd have to unlock the root account for the thin clients
| |
| 17:08 | sepski has quit IRC | |
| 17:08 | <alkisg> Bah, anyway, just reboot the client, run X -configure and scp it without running X -config
| |
| 17:08 | <Mip5> Good idea - brb
| |
| 17:17 | bummer -
| |
| 17:17 | I can't ssh or scp from this shell to my server
| |
| 17:17 | nor the other way around
| |
| 17:17 | I'm going to have to pick this up tomorrow =
| |
| 17:17 | Thanks a lot for your help.
| |
| 17:18 | I feel as if it's within reach - at least I know it's X related now
| |
| 17:18 | <alkisg> Ok... still thinking it should be autodetected though... g'night.
| |
| 17:19 | mikkel has quit IRC | |
| 17:19 | <Mip5> Right - it does work when I boot from the ubuntu CD - but that's not how I want these machines set up
| |
| 17:21 | bobby_C has quit IRC | |
| 17:21 | EAG has left #ltsp | |
| 17:23 | alkisg has quit IRC | |
| 17:37 | johnny has left #ltsp | |
| 17:38 | johnny has joined #ltsp | |
| 17:38 | johnny is now known as Guest87249 | |
| 17:39 | Guest87249 has left #ltsp | |
| 17:48 | Mip5 has quit IRC | |
| 18:16 | johnny has joined #ltsp | |
| 18:22 | <Ryan52> oh, I had assumed that password expiry would also use the SSH_ASKPASS, but I could defidently see why it wouldn't..
| |
| 18:23 | * Ryan52 was going to test these things before he emailed the ml. | |
| 18:57 | staffencasa has quit IRC | |
| 18:59 | Lns has quit IRC | |
| 19:03 | johnny has left #ltsp | |
| 19:14 | vagrantc has quit IRC | |
| 19:16 | ikij has joined #ltsp | |
| 19:16 | |Paradox| has quit IRC | |
| 19:16 | nothingman has joined #ltsp | |
| 19:16 | ikij is now known as |Paradox| | |
| 19:16 | <nothingman> hi, all
| |
| 19:19 | hi, all
| |
| 19:20 | <Ryan52> hi, nothingman
| |
| 19:20 | <nothingman> what's new tonight?
| |
| 19:21 | I just installed Ubuntu ltsp server on an older Dell and now I get a blank desktop when I log in
| |
| 19:21 | seems to be frozen, but my mouse batteries died, so there ya go
| |
| 19:23 | not edubuntu, ltsp from the grub menu and then the rest
| |
| 19:46 | brendan0powers has joined #ltsp | |
| 19:53 | otavio has quit IRC | |
| 19:56 | LinusU has quit IRC | |
| 19:56 | johnny has joined #ltsp | |
| 20:00 | RobertLaptop has quit IRC | |
| 20:10 | LinusU has joined #ltsp | |
| 20:33 | nothingman has quit IRC | |
| 20:37 | haz1 has joined #ltsp | |
| 20:39 | <haz1> does anyone know how to correctly edit lbus_event_handler.sh to specify "allow_root" as a fuse option? if i change the add line to be '/usr/sbin/ltspfs ${WS}:/tmp/drives/${SHARENAME} "${HOME}/${DRIVEDIR}/${SHARENAME}" allow_root' it stops working :(
| |
| 20:39 | trying to follow scott's advice here: http://www.mail-archive.com/ltsp-discuss@lists.sourceforge.net/msg29623.html
| |
| 20:40 | i have user_allow_other in /etc/fuse.conf
| |
| 20:41 | this is with ltsp 4.2, and i'm trying to get fuse to allow root to access the local device mounts
| |
| 20:44 | rcy has joined #ltsp | |
| 20:50 | nothingman has joined #ltsp | |
| 21:03 | hanthana has joined #ltsp | |
| 21:15 | <nothingman> awful quiet in here
| |
| 21:15 | I feel I might be eaten by a grue
| |
| 21:16 | GRUE_ has joined #ltsp | |
| 21:17 | * GRUE_ eats nothingman | |
| 21:17 | GRUE_ has quit IRC | |
| 21:17 | LinusU has quit IRC | |
| 21:17 | <Ryan52> :p
| |
| 21:28 | <haz1> think i've answered my own question; it should be "-o allow_root".
| |
| 21:28 | dirigeant has quit IRC | |
| 21:37 | <nothingman> lol
| |
| 21:39 | haz1 has left #ltsp | |
| 21:51 | brendan0powers has quit IRC | |
| 21:51 | vvinet has quit IRC | |
| 22:02 | CaScAdE^FarAway has joined #ltsp | |
| 22:17 | pscheie has quit IRC | |
| 22:19 | CaScAdE^1arAway has quit IRC | |
| 22:40 | pscheie has joined #ltsp | |
| 22:54 | tinpardo has quit IRC | |
| 22:57 | tinpardo has joined #ltsp | |
| 22:58 | <tinpardo> h
| |
| 23:23 | pscheie has quit IRC | |
| 23:26 | rzlagwqfpckx has joined #ltsp | |
| 23:26 | |Paradox| has quit IRC | |
| 23:26 | cyberorg has quit IRC | |
| 23:26 | rzlagwqfpckx is now known as |Paradox| | |
| 23:38 | warren has quit IRC | |
| 23:43 | pscheie has joined #ltsp | |
| 23:48 | warren has joined #ltsp | |
| 23:55 | alkisg has joined #ltsp | |