00:09 | johnny has joined #ltsp | |
00:55 | alkisg has joined #ltsp | |
02:34 | F-GT has quit IRC | |
02:45 | vagrantc has quit IRC | |
02:50 | F-GT has joined #ltsp | |
03:26 | Lns has quit IRC | |
03:38 | jaime_david has joined #ltsp | |
03:39 | <jaime_david> Hi I'm using K12Linux, can i ltsp-build-client --arch ppc from a x64 server?
| |
03:40 | ogra has joined #ltsp | |
03:41 | <jaime_david> Hi
| |
03:41 | I'm using K12Linux, can i ltsp-build-client --arch ppc from a x64 server?
| |
03:47 | jaime_david has left #ltsp | |
04:09 | Selveste1 has joined #ltsp | |
04:37 | alkisg has quit IRC | |
05:31 | highvoltage has joined #ltsp | |
06:09 | hersonls has joined #ltsp | |
06:11 | artista_frustrad has quit IRC | |
06:24 | artista_frustrad has joined #ltsp | |
06:35 | artista_frustrad has quit IRC | |
06:47 | artista_frustrad has joined #ltsp | |
06:56 | artista_frustrad has quit IRC | |
06:57 | pmatulis has joined #ltsp | |
07:09 | artista_frustrad has joined #ltsp | |
07:18 | ogra has quit IRC | |
07:18 | _UsUrPeR_ has joined #ltsp | |
07:19 | ogra has joined #ltsp | |
07:19 | jammcq has quit IRC | |
07:22 | artista_frustrad has quit IRC | |
07:35 | artista_frustrad has joined #ltsp | |
07:35 | scottmaccal has joined #ltsp | |
07:36 | astigmata has joined #ltsp | |
07:36 | <astigmata> hi all
| |
07:37 | I m connected with ltsp atm
| |
07:37 | but my problem, I can't see or mount local harddisk
| |
07:38 | is possible to auto-mount HDD with ltsp ?
| |
08:00 | hersonls has quit IRC | |
08:00 | hersonls has joined #ltsp | |
08:09 | rjune has quit IRC | |
08:09 | try2free has joined #ltsp | |
08:15 | ogra_ has joined #ltsp | |
08:15 | ogra has quit IRC | |
08:16 | ogra_ is now known as ogra | |
08:16 | ogra has joined #ltsp | |
08:23 | try2free has left #ltsp | |
08:23 | epaphus has joined #ltsp | |
08:30 | astigmata has quit IRC | |
08:54 | jammcq has joined #ltsp | |
09:11 | epaphus has quit IRC | |
09:14 | Faithful has quit IRC | |
09:32 | CAN-o-SPAM has joined #ltsp | |
09:52 | dark4og has joined #ltsp | |
09:56 | <dark4og> We have been using Ubuntu LTSP for that past year, We are now in need of desktop scanners at all of our thin clients. Sane will not recognize the attached. thin client USB Scanner. I can get it to scan if I drop to a root on the thin client, (ctl-alt-f1) yet not available on the to any useres.. Any help would be GREAT!
| |
09:58 | I have set the thin client to a static IP and am running localapps, of firefox and flash
| |
10:00 | This is on a "Disklessworkstation" and PXE boot. LTSP5, Ubuntu 9.10.
| |
10:03 | <jammcq> dark4og: so the scanner works with SANE, but just not on the thin client?
| |
10:05 | dark4og: you'll need saned running on the thin client. that should make the scanner available via the network. Then, you need to set some environment variables on the server so that your scanner-aware apps can see it
| |
10:09 | romel has joined #ltsp | |
10:16 | staffencasa has joined #ltsp | |
10:46 | runout has left #ltsp | |
10:53 | loomis53 has quit IRC | |
10:57 | Faithful has joined #ltsp | |
10:58 | <moldy> dark4og: you could run the applications that need to access the scanner as local apps
| |
11:01 | Faithful has quit IRC | |
11:09 | Lns has joined #ltsp | |
11:34 | nubae has joined #ltsp | |
11:38 | nubae has quit IRC | |
11:38 | nubae has joined #ltsp | |
11:39 | epaphus has joined #ltsp | |
11:39 | nubae_ has quit IRC | |
11:39 | <epaphus> Hello, whats a good way to see if a particular video chipset is supported in LTSP ??
| |
11:40 | <johnny> if it's supported by linux
| |
11:40 | it's supported by ltsp
| |
11:40 | so that's not a useful question
| |
11:41 | the same could be said of most hardware
| |
11:41 | except local cd/dvd burners and scanners
| |
11:42 | hopefully the need for saned will die ..
| |
11:42 | <epaphus> hmm..
| |
11:42 | <johnny> and that won't be a problem anymore
| |
11:42 | for scanners
| |
11:42 | <epaphus> understood
| |
11:42 | <johnny> the cd/dvd burner stuff is alot more difficult tho :(
| |
11:44 | alkisg has joined #ltsp | |
11:44 | <epaphus> thanks johnny
| |
11:50 | <dark4og> jammcq: Yes the scanner works on the thin client if i log in as root, through (ctrl-alt-f1) then run scanimage.
| |
11:52 | moldy: When I run it as a localapps, i get nothing. nada... I added the sane to the image, and rebuilt the img files. still nothing.
| |
11:55 | Lns has quit IRC | |
11:58 | <moldy> dark4og: hm, that's weird
| |
12:01 | <alkisg> Maybe the user needs to be in some special group, like, I don't know, plugdev?
| |
12:01 | john has joined #ltsp | |
12:02 | <alkisg> crw-rw---- 1 root disk 21, 1 2010-04-05 19:35 /dev/sg1
| |
12:03 | <john> I'm having a login problem on a brand new install with Ubuntu 9.10, basically the LDM screen acts like it's going to login, then the session is closed. /var/log/auth.log offers no explanation
| |
12:03 | where else can I look to troubleshoot this issue?
| |
12:03 | <alkisg> john: do you see a successful authentication in the server auth.log?
| |
12:06 | <john> http://pastebin.com/Kcx71Gj5 will show the log
| |
12:06 | Looks like it authenticates just fine, but then during the login the session is closed
| |
12:07 | <alkisg> john: I've seen those symptoms with compiz problems. Can you try disabling compiz?
| |
12:07 | !compiz
| |
12:07 | <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
| |
12:08 | <john> that worked!
| |
12:08 | <alkisg> Wow, fast testing :)
| |
12:12 | vagrantc has joined #ltsp | |
12:14 | <john> I had two servers, both with the same problem. Now they are fixed, thanks!
| |
12:14 | * alkisg thinks that if the ltsp package shipped with a file in /usr/share/gconf/defaults/10-compiz, which made compiz==off by default, it would save people some trouble... | |
12:34 | <stgraber> alkisg: well, the way LTSP works currently is that it matches the distro's behavior that you'd have if you installed the distro directly on the thin client
| |
12:35 | alkisg: my goal for Ubuntu is to have LTSP's user experience identical to a regular desktop one, unfortunately this also means having the same bugs ;)
| |
12:35 | <alkisg> stgraber: sure, but some cards have compiz problems with LTSP while they don't have them as normal workstations
| |
12:35 | <stgraber> so in these cases, a bug should probably be opened against compiz so that this specific hardware is blacklisted in compiz itself
| |
12:36 | <alkisg> Because of the remote X that is involved
| |
12:36 | <stgraber> alkisg: really ? We had that kind of issues when we were using the compiz-launcher that bypassed most checks, now we are using compiz's own checks and it seems to work for most broken hardware here (as in, compiz gets disabled)
| |
12:37 | for the others, the bug should really be opened against compiz so that additional checks are added to blacklist these cards (or to fix the bug that prevent them from working properly)
| |
12:38 | ideally I'd add a COMPIZ=false in lts.conf that'd let people disable compiz though Ubuntu's completely frozen right now and that's a new feature so that won't go through feature freeze easily
| |
12:38 | <alkisg> Hmmm good idea though
| |
12:39 | <john> In our case compiz does not really accomplish much anyway, 6 year olds don't seem to care to much for the features
| |
12:40 | <alkisg> john, out of curiosity, what cards are these? lspci -nn...
| |
12:40 | !localxterm
| |
12:40 | <ltspbot`> alkisg: "localxterm" :: while sitting on a thin client, open a gnome terminal. In that, run: ltsp-localapps xterm. An xterm will open. That xterm runs locally, so any commands you enter there are executed directly on the client.
| |
12:40 | <stgraber> it's not that much compiz that I want by default, it's compositing and compiz is probably the best compositing manager out there
| |
12:41 | when a thin client runs a compositing manager, windows are stored as textures and cached locally. It essentially makes the network usage lower and makes the UI almost two times faster to use.
| |
12:42 | it's basically what makes the LTSP desktop feels as it was a regular desktop. No more lag when moving/maximizing windows, ...
| |
12:42 | <alkisg> But it also increases xorg ram requirements, I believe, so that COMPIZ=false option could have an additional use for low end PCs...
| |
12:42 | <john> VIA Technologies, Inc. CN400/PM800/PM880/PN800/PN880 [S3 UniChrome Pro]
| |
12:42 | <stgraber> alkisg: right, though in my tests, a whole desktop running with two monitors at 1680x1050 with 24bits color, take less than 90MB of RAM (for the whole system, not only X)
| |
12:44 | <john> looks like VIA chipset for everything on this client, the client is a Wyse V50
| |
12:45 | <alkisg> Hmmm I think I have a unichrome client with the same problem, but unfortunately it only has 128mb ram, so I cannot try it as a standalone client to see if the compiz problem is there as well...
| |
12:45 | klausade has joined #ltsp | |
12:46 | <alkisg> stgraber: do you have anything in ltsp cluster for executing commands on the clients?
| |
12:46 | Like, an ssh server on the chroot, or a daemon that listens for commands from the server etc?
| |
12:46 | <stgraber> we usually run a ssh server in the chroot
| |
12:47 | in the next generation of ltsp-cluster we have an XML+RPC daemon running on each thin clients and application servers with a plugin infrastructure that can be used to remotely execute actions or get data
| |
12:47 | <alkisg> stgraber: I'm working on something like that, that's why I"m asking - do you have any code atm?
| |
12:48 | ogra__ has joined #ltsp | |
12:48 | cliebow has joined #ltsp | |
12:48 | * alkisg has written server+client code in bash, and is rewriting them on C/Python... | |
12:48 | ogra_cmpc has quit IRC | |
12:48 | <johnny> no.. not xmlrpc stgraber :(
| |
12:48 | yuck
| |
12:48 | either use something stateful like xmpp
| |
12:49 | or something more restful like atompub or json
| |
12:49 | or use dbus
| |
12:49 | over tcp
| |
12:49 | please.. just not xmlrpc :(
| |
12:49 | <alkisg> I'm looking to use something with "select" on the server, because e.g. python's socketserver needed 10 MB RAM for each client
| |
12:49 | So for 100 clients 1Gb of RAM would be needed - yuck...
| |
12:49 | <johnny> alkisg, considered having something on the client that inotifies ?
| |
12:49 | like incron ?
| |
12:50 | <alkisg> johnny: inotify doesn't work over sshfs / nfs
| |
12:50 | <johnny> sure.. but why not watch for a tmpfs/aufsed writable dir?
| |
12:50 | <stgraber> alkisg: we have an xml+rpc server somewhere on launchpad with a relatively easy to use plugin infrastructure
| |
12:50 | <alkisg> With "select" no forks/threads are neccessary, so the RAM requirements will be very low
| |
12:50 | <johnny> stgraber, xmlrpc should die :(
| |
12:50 | seriously you should consider an xmpp server or dbus tho.. if you are keeping state..
| |
12:50 | <stgraber> the server side is PHP, doing XML+RPC for communication with the daemons and a REST+JSON interface for the web UI and scripting
| |
12:50 | <alkisg> johnny: I'm looking to implement reverse connections, for safety reasons
| |
12:51 | stgraber: I'm guessing that won't run for 64 MB clients... :-/
| |
12:51 | johnny: i.e. all clients connect to the server, they don't accept connections from anyone
| |
12:51 | I'm reusing the ssh channel so it's also encrypted...
| |
12:52 | stgraber: do you have some link handy?
| |
12:52 | <stgraber> alkisg: not really an issue for us right now as we don't have anything with less than 1GB ;) But worst case we'll rewrite the daemon running on the thin clients in C, that should make things better.
| |
12:53 | <alkisg> Oooh that would be perfect. I need to have this ready in a month or so, so if there's any code around, I might be able to utilize it / rewrite it / test it etc...
| |
12:53 | <stgraber> johnny: dbus over TCP doesn't work that well when you have your server running on Windows and JSON is great to use with AJAX web interfaces but not that good when you need proper support of variable types and exceptions
| |
12:55 | <johnny> ah.. server running on windows..
| |
12:55 | klausade has quit IRC | |
12:55 | <stgraber> not always but it can and our agent should ideally be the same for all appservs, be that Linux or Windows
| |
12:56 | <johnny> sure..
| |
12:56 | just didn't think people would want to do that.. but i guess they might..
| |
12:56 | <stgraber> well, we work with governments, large school districts and multi-nationals corporation, so ...
| |
12:56 | klausade has joined #ltsp | |
12:57 | <johnny> xmpp with pubsub still sounds like a good idea tho..
| |
12:57 | <stgraber> yeah, would need to look at that and see if I can get a single server to work with over 10000 thin clients
| |
12:58 | <johnny> prosody seems like it could with libev
| |
12:59 | of course.. the main prosody dev is always lookin for xmpp related work.. so.. if you ever need somebody to code it up.. he could :)
| |
13:02 | <stgraber> alkisg: bzr get lp:~ltsp-cluster-team/ltsp-cluster/ltsp-cluster-agent that's what I currently have in our internal bzr but I'm guessing it's missing quite a few revisions
| |
13:03 | <alkisg> thanks stgraber :)
| |
13:03 | <stgraber> alkisg: today is an holiday here so I'll poke the guy working on that tomorrow so he updates on LP (he always forget to push his changes ...)
| |
13:04 | <johnny> alkisg, btw.. prosody runs on an openwrt
| |
13:04 | so.. it's tiny
| |
13:04 | * alkisg googles for prosody... | |
13:04 | <johnny> as in.. it can run on that..
| |
13:23 | japerry has joined #ltsp | |
13:31 | john has quit IRC | |
13:41 | * alkisg doesn't see any remote control implementations using reverse connections... :-/ | |
13:41 | john has joined #ltsp | |
13:46 | <johnny> alkisg, cuz it's hard to do..
| |
13:46 | <alkisg> johnny: Running a remote shell through XMPP: http://snippets.dzone.com/posts/show/6097
| |
13:46 | 30 lines or less...
| |
13:47 | <johnny> it could also be done in lua as a prosody module
| |
13:47 | you can probably convince mattj to write it for ya :)
| |
13:47 | ejabberd has one, but the ram footprint is alot higher
| |
13:47 | <alkisg> If an XMPP server was running on the server, and the clients connected to it and gave it remote control capabilities, that would be reverse connections
| |
13:47 | I just don't want daemons running on the clients, as they're insecure by design..
| |
13:47 | *accepting connections
| |
13:47 | <johnny> not as insecure as you think tho..
| |
13:47 | i think you'er makign too much of it
| |
13:48 | you already run with ldm directx on anyways don't you??
| |
13:48 | <alkisg> Sure, but if I'm going to invest some time on something, I prefer to do it right if I can...
| |
13:48 | Also, reverse connections enable the server to keep track of each client/user
| |
13:49 | <johnny> well.. sure.. if the clients run an xmpp client on them.. of course you can trakc em
| |
13:49 | using telepathy would probably be a good idea here..
| |
13:49 | but i don't know if they have libraries and clients that are slim enough for your needs
| |
13:49 | that's probably the biggest problem
| |
13:49 | <alkisg> There's already an ssh connection, so it'd be nice if that was reused (I'm currently port-forwarding the ports I use, and connect to localhost instead)
| |
13:50 | <johnny> why don't you talk to the telepathy folks?
| |
13:50 | just to see
| |
13:50 | danielle madely seems like a reasonable person
| |
13:50 | (danni)
| |
13:50 | <alkisg> The client in C needs about the same ram as `sh`, so it's pretty light
| |
13:50 | <johnny> formerly davyd madeley
| |
13:50 | in case you find plenty of posts by davyd.. which you will..
| |
13:51 | telepathy stuff does run on nokia n900 and other tech
| |
13:51 | nubae_ has joined #ltsp | |
13:52 | <john> looks like VIA chipset for everything on this client, the client is a Wyse V50
| |
13:53 | oops
| |
13:53 | nubae has quit IRC | |
13:56 | nubae has joined #ltsp | |
13:57 | nubae_ has quit IRC | |
14:05 | dark4og has quit IRC | |
14:14 | nubae_ has joined #ltsp | |
14:15 | nubae has quit IRC | |
14:21 | <john> looks like VIA chipset for everything on this client, the client is a Wyse V50
| |
14:21 | john has quit IRC | |
14:23 | aliancemd has joined #ltsp | |
14:23 | <aliancemd> hello
| |
14:24 | Can someone help me with a LTSP problem?
| |
14:24 | <johnny> !ask
| |
14:24 | <ltspbot`> johnny: "ask" :: Don't ask to ask a question, simply ask it, and if someone knows the answer, they'll respond. Please hang around for at least 15 minutes after asking a question, as not everybody constantly monitors the channel.
| |
14:27 | <aliancemd> I need to boot over network from an Ubuntu 9.10 LTSP Server, computers with: 450 Mhz, 32 MB Ram, 8 MB Video. I had installed the LTSP Server, when I boot powerful computers it all works great. But when I try to boot this Thin Client, it boot to the login screen and there freezes, even the mouse can't move
| |
14:27 | <johnny> 32mb is not possible
| |
14:28 | 64mb is the lower lower limit
| |
14:28 | <aliancemd> I know that minimum has to be 48
| |
14:28 | hmm
| |
14:28 | <johnny> is it 48? hmm
| |
14:28 | didn't know we went that low
| |
14:28 | <aliancemd> Ye
| |
14:28 | <johnny> but i've never heard of 32
| |
14:28 | <alkisg> Nah... 48 is for the kernel, it won't be usable
| |
14:28 | <johnny> ah
| |
14:28 | <aliancemd> on 1 site, 32, on another 48, and another 64
| |
14:28 | <johnny> 64 is the lower limit
| |
14:28 | and that's tough
| |
14:28 | <aliancemd> In 1 pdf file about LTSP: 48 MB Ram and 16 MB Video+
| |
14:29 | <johnny> that was for the older version perhaps
| |
14:29 | ltsp 4
| |
14:29 | <alkisg> 4mb video is ok :)
| |
14:29 | <johnny> ltsp4 isn't supported anymore
| |
14:29 | aliancemd, the linux kernel itself is just too bi
| |
14:29 | <aliancemd> I tried to create a swap, but now it says connection lost
| |
14:29 | <johnny> big*
| |
14:29 | <aliancemd> when it tried to boot
| |
14:30 | * jammcq remembers the days of running a thin client with only 16mb | |
14:30 | <alkisg> aliancemd: try with the 64mb clients first
| |
14:30 | <jammcq> ah, the good ole days :)
| |
14:30 | * vagrantc remembers running thin clients with 8mb | |
14:30 | <jammcq> with X ?
| |
14:30 | <aliancemd> The problem is that I have 10+ Thin Clients with this configuration
| |
14:30 | * alkisg has run windows 3.11 with no hard disks with 2mb :D | |
14:30 | <aliancemd> and 4 with 64 MB Ram
| |
14:31 | I think the option is the swap?
| |
14:31 | How I understand the video card is not the problem, the Ram is
| |
14:32 | <johnny> aliancemd, the part that must bei n active ram is too big
| |
14:32 | for your 48mb clients
| |
14:32 | period
| |
14:32 | swap won't help
| |
14:32 | <cliebow> 4.2 Still Rocks..
| |
14:32 | <alkisg> aliancemd: for 32mb/48mb clients, ltsp 5 isn't an option. Either try the unsupported ltsp 4 (you're on your own there), or some other solution.
| |
14:32 | codyg has joined #ltsp | |
14:33 | <aliancemd> Ok, thx
| |
14:33 | <codyg> anyone have any luck getting sound out of the Citrix ICA Client? Possibly as a local app?
| |
14:33 | <cliebow> codyg, that sounds like a ? for Gadi..
| |
14:34 | jammcq: reaching the tipping point to move to the new digs 8~)
| |
14:35 | <jammcq> cliebow: awesome
| |
14:35 | cliebow: I need to call David Lloyd and setup the next bts
| |
14:35 | <aliancemd> One more time: So you say that I can't run 32 MB Ram on LTSP 5 anyway, even if I try to create a swap on the ThinClient?
| |
14:35 | <cliebow> still missig a few doors and such..
| |
14:35 | <johnny> aliancemd, yes
| |
14:35 | <cliebow> yeah...
| |
14:35 | <johnny> that is correct
| |
14:35 | <alkisg> aliancemd: you might be able to boot it in e.g. 20 minutes, but it won't be usable
| |
14:35 | <aliancemd> bad, thx
| |
14:37 | <codyg> cliebow, I'm not familiar with Gadi.
| |
14:38 | <alkisg> !seen Gadi
| |
14:38 | <ltspbot`> alkisg: Gadi was last seen in #ltsp 2 days, 22 hours, 40 minutes, and 57 seconds ago: <Gadi> !stgraber-ppa
| |
14:38 | <cliebow> he isnt handy today for some reason
| |
14:38 | <johnny> !seen sbalnaev
| |
14:38 | <ltspbot`> johnny: I have not seen sbalnaev.
| |
14:38 | <johnny> /me missed a letter
| |
14:39 | <cliebow> !seen sbalneav
| |
14:39 | <ltspbot`> cliebow: sbalneav was last seen in #ltsp 1 week, 4 days, 1 hour, 9 minutes, and 0 seconds ago: <sbalneav> No "process" magic.
| |
14:39 | <jammcq> is it still a holiday in Canada?
| |
14:39 | * jammcq thinks scotty was away last week | |
14:39 | <jammcq> on vacation
| |
14:41 | <codyg> Errorregedit
| |
14:41 | typing on wrong keyboard
| |
14:41 | <cliebow> ?
| |
14:41 | heh..
| |
14:41 | <codyg> <-- gets confused with kvm and other keyboard, sorry
| |
14:41 | <cliebow> i used to put my pw in all the time 8~)
| |
14:43 | <johnny> too bad.. cuz i wanted to tell him about packagekit and ltsp
| |
14:43 | <alkisg> In a mixed thin/fat/standalone clients lab, how would you call the standalone clients? Standalone clients? Workstations?
| |
14:43 | <johnny> perhaps thick is better than fat?
| |
14:43 | :)
| |
14:43 | <cliebow> chubby
| |
14:43 | <johnny> disked worked stations
| |
14:44 | disked workstations :)
| |
14:44 | <alkisg> Well, too late now, LTSP_FATCLIENT has been in the code for too long now :D
| |
14:44 | <johnny> lol
| |
14:44 | <cliebow> grotesqlient
| |
14:44 | <alkisg> Heh, a disked workstation can also be used as a thin/fat client...
| |
14:44 | <johnny> vs disklessworkstations :)
| |
14:44 | bobby_C has joined #ltsp | |
14:49 | <aliancemd> I found the differences between LTSP 4 and 5 are just in: Compression of the files transmited, Secured connection, Secured, Secured... But I found that I can configure to disable secured connection, compression... in the lts.conf file. How about now, can I run on 32 MB Ram? :) Or that's not the only difference?
| |
14:55 | <vagrantc> aliancemd: the difference is in the philosophy- ltsp5 is built from your host distro's packages. ltsp4 was essentially it's own distribution... which no longer recieves updates... so all the code running ltsp4 is very, very old.
| |
14:55 | <aliancemd> thx
| |
14:56 | <vagrantc> if you were to re-implement ltsp4 with current versions of things such as X.org, it would likely get a performance hit too
| |
14:57 | unfortunately, free software is not immune to software bloat.
| |
14:58 | <codyg> Not sure if anyone cares...after I installed the ICAClient as a localapp I was able to get sound by clicking the "Use audio devices" in the users-admin panel...makes sense, right?
| |
15:04 | scottmaccal has quit IRC | |
15:08 | <aliancemd> !senn the_code
| |
15:08 | <ltspbot`> aliancemd: Error: "senn" is not a valid command.
| |
15:09 | <aliancemd> !seen the_code
| |
15:09 | <ltspbot`> aliancemd: the_code was last seen in #ltsp 1 year, 1 week, 5 days, 0 hours, 23 minutes, and 57 seconds ago: <The_Code> ok, bad, i can't tell anymore, sorry
| |
15:18 | vagrantc has quit IRC | |
15:35 | CAN-o-SPAM has quit IRC | |
15:38 | nubae_ has quit IRC | |
15:39 | nubae has joined #ltsp | |
15:40 | ogra__ has quit IRC | |
15:49 | <_UsUrPeR_> Does sshfs understand acl and abide by the rules laid out with setfacl?
| |
15:50 | aliancemd has left #ltsp | |
15:51 | aliancemd has joined #ltsp | |
15:51 | aliancemd has left #ltsp | |
15:51 | aliancemd has joined #ltsp | |
15:51 | aliancemd has left #ltsp | |
15:52 | pmatulis has quit IRC | |
15:52 | ludug3r0 has joined #ltsp | |
15:52 | ogra__ has joined #ltsp | |
15:53 | <ludug3r0> alkisg: Hello. Do you know if there is some kinf of user limit that make the ldm hang up when using FATCLIENTS?
| |
15:55 | I'm having really strange behavior with LDM... I try to login with another user then it hang when starting the xsession or it can't authenticate.... When I shutdown 2 working machines and reboot the problamatic one, the problematic one now works fine...
| |
15:57 | <alkisg> ludug3r0: no, I've never heard of such a problem... and there aren't any relevant messages in the system logs?
| |
16:00 | ludug3r0 has quit IRC | |
16:04 | bobby_C has quit IRC | |
16:06 | ludug3r0 has joined #ltsp | |
16:06 | <ludug3r0> alkisg: Sorry!
| |
16:06 | alkisg: My system hanged.. :\
| |
16:06 | <alkisg> (11:50:01 μμ) alkisg: ludug3r0: no, I've never heard of such a problem... and there aren't any relevant messages in the system logs?
| |
16:07 | <ludug3r0> The only information I can get in the logs is that sometimes the gnome-manager reports timeout of applications
| |
16:07 | Let me get a example here...
| |
16:08 | Cannot register authentication agent...
| |
16:08 | Humm...
| |
16:08 | <alkisg> ludug3r0: you're using different users, right? not the same one on all 3 pcs..?
| |
16:09 | <ludug3r0> different users...
| |
16:09 | I'm using nfs...
| |
16:09 | I was thinking the timout problem was due to a clock skew...
| |
16:10 | I've seted a ntp server on the ltsp-server... setted up the nfs machine and clients to use the ltsp-server as the ntp-server...
| |
16:11 | I don know if this would solve my problem with clock skew...
| |
16:11 | hersonls has quit IRC | |
16:12 | <ludug3r0> But the fact is that I tryed everything with one machine... I couldn't login using it no matter what... Then I turned off two other machines and in the first try I got in...
| |
16:13 | I suspecting my server machine can't stand the load or the nsf machine...
| |
16:13 | <alkisg> ludug3r0: does it work if you put LTSP_FATCLIENT=False in lts.conf?
| |
16:14 | <ludug3r0> I didn't try to disable the fatclient...
| |
16:14 | what should i foresee disabling it? it will run application remotely, right?
| |
16:15 | <alkisg> Yup
| |
16:15 | <ludug3r0> Let me try...
| |
16:15 | <alkisg> That might tell you if you have a fat client related problem, or maybe a server problem...
| |
16:15 | (e.g. ssh limit?)
| |
16:16 | <ludug3r0> Humm.
| |
16:16 | I have another question...
| |
16:16 | When my fatclient logs at the server it creates a .Xauthority...
| |
16:16 | At the server I used the localdisk for the home users...
| |
16:17 | At the clients I use nfs...
| |
16:17 | So it ends up creating a .Xauthority at the server that I don't know if it is related with the one there is in the nfs-home...
| |
16:18 | So I tryed mounting the nfs-home at the ltsp-server...
| |
16:18 | end tryed not mounting the nfs-home at the ltsp-server...
| |
16:18 | I saw no difference...
| |
16:19 | ok... with the fat client turned off i got one machine to run..
| |
16:19 | let me try others...
| |
16:25 | alkisg: ok.. I'm breaking records here... Already 4 machines running with this flag disabled...
| |
16:25 | there are others fat clients logged too...
| |
16:29 | epaphus has quit IRC | |
16:30 | <ludug3r0> alkisg: ok.. I could log on five of then...
| |
16:31 | then the sixty one hunged up at the ldm... it keeps trying to see if the password is right then reports no response from server and restarts.
| |
16:33 | I will see your idea about ssh limits...
| |
16:35 | there is a lot of sshd spawned process...
| |
16:38 | Lns has joined #ltsp | |
16:39 | <alkisg> ludug3r0: sorry I was afk, reading...
| |
16:39 | <ludug3r0> No problem... I'm killing the sshds...
| |
16:39 | There are a lot of then...
| |
16:39 | <alkisg> yeah if you're having the same problem with thin clients (LTSP_FATCLIENT=False), then this isn't a fat client problem
| |
16:40 | <ludug3r0> Maybe zombies
| |
16:40 | <alkisg> Maybe you could try ssh'ing manually, to see the error message
| |
16:40 | (before you kill all sshd's, that is...)
| |
16:41 | <ludug3r0> I wasn heving problems with ssh..
| |
16:41 | I trying to use the fat clients now...
| |
16:41 | Ahh.
| |
16:41 | Can you comment over the home issue?
| |
16:41 | <alkisg> Let me see...
| |
16:42 | <ludug3r0> should I mount the same home at the server and at the clients?
| |
16:42 | <alkisg> Are your clients directly connected to the internet? Or only through the server?
| |
16:42 | hmmm ok if you installed ntp-server, no problem there...
| |
16:42 | <ludug3r0> They are connect to the internet trough another machine different than the ltsp-server
| |
16:43 | nubae_ has joined #ltsp | |
16:43 | <alkisg> (12:10:26 πμ) ludug3r0: So it ends up creating a .Xauthority at the server that I don't know if it is related with the one there is in the nfs-home... ==> I'm not sure what you mean, it should be in /home/user/.Xauthority no matter where you boot from (server, client, nfs or not)
| |
16:43 | <ludug3r0> Hummm...
| |
16:44 | Let me try to explain it better...
| |
16:44 | <alkisg> Did you put NFS_HOME=/home in lts.conf ?
| |
16:44 | <ludug3r0> Theres is a home directory at the ltsp-server...
| |
16:44 | <alkisg> Did you install nfs-client in the chroot?
| |
16:44 | nubae has quit IRC | |
16:44 | <ludug3r0> I put NFS_HOME = 192.168.8.253:/data/homes
| |
16:44 | Yeah...
| |
16:45 | I can get my nfs directory running fine.
| |
16:45 | Theres is a /home/user at the ltsp-server machine...
| |
16:45 | and there is a /home/user at the client mounted over nfs
| |
16:45 | jhutchins has quit IRC | |
16:46 | <alkisg> Wait wait. What do you get with getent user? /home/user or /data/home/user ?
| |
16:46 | *getent passwd username
| |
16:46 | <ludug3r0> I tryed leaving the /home/user at the ltsp-server on the local disk... And tryed using the /home/user at the ltsp-server over the nfs also...
| |
16:47 | I get /home/user
| |
16:47 | Both ways I end with /home/user
| |
16:47 | I read the scripts at the rc.d..
| |
16:47 | jhutchins has joined #ltsp | |
16:48 | <ludug3r0> It seens it looks if there is a "server:" part...
| |
16:48 | And it seens to work fine...
| |
16:48 | <alkisg> OK... did you try after a clean server boot? You're still able to logon only 2-3 clients then as well/
| |
16:48 | ?
| |
16:49 | (to make sure there aren't any zombie processes around...)
| |
16:49 | <ludug3r0> I did try a clean boot hours ago... But the problem persisted there...
| |
16:50 | <alkisg> Very strange...
| |
16:50 | <ludug3r0> I searched for processes executed by user now and killed then all...
| |
16:50 | I'm switching all the machines now...
| |
16:50 | And logging in without problems until now...
| |
16:50 | Ok...
| |
16:51 | I have a problem.. :(
| |
16:51 | At the third one...
| |
16:51 | Maybe I shouldn't turn then on at the same time...
| |
16:52 | <alkisg> No, there's a limit there, but it's at 10 users, not 3
| |
16:52 | <ludug3r0> They hang at "Verifying password. Please wait"
| |
16:52 | Where I can edit this limit?
| |
16:52 | <alkisg> Never mind, that's not it...
| |
16:53 | Do you have local shell access, to see /var/log/ldm.log on the clients?
| |
16:53 | (e.g. SCREEN_02=shell...)
| |
16:53 | !shell-screen
| |
16:53 | <ludug3r0> yes...
| |
16:53 | <ltspbot`> alkisg: "shell-screen" :: One way to get local root access to troubleshoot a thin client is to use a shell SCREEN: https://help.ubuntu.com/community/UbuntuLTSP/ClientTroubleshooting#Using%20a%20shell%20SCREEN
| |
16:53 | <alkisg> !localxterm
| |
16:54 | <ltspbot`> alkisg: "localxterm" :: while sitting on a thin client, open a gnome terminal. In that, run: ltsp-localapps xterm. An xterm will open. That xterm runs locally, so any commands you enter there are executed directly on the client.
| |
16:54 | <ludug3r0> Well... Iǘe counted the sshds forked here...
| |
16:54 | I have precisely 10 sshds forked...
| |
16:54 | <alkisg> Uhm... that doesn't sound right
| |
16:54 | 10 sshds for 3 clients?
| |
16:55 | <ludug3r0> just twop seconds.
| |
16:55 | ok...
| |
16:55 | it hanged.
| |
16:55 | I was trying to put another machine up...
| |
16:55 | <alkisg> Is that Lucid?
| |
16:55 | <ludug3r0> Now I have 12 forked sshds...
| |
16:55 | <alkisg> Or Karmic?
| |
16:56 | <ludug3r0> root 12211 0.0 0.1 10224 3292 ? Ss 18:47 0:00 sshd: raso [priv] raso 12267 0.0 0.0 10356 2304 ? S 18:47 0:00 sshd: raso@pts/4
| |
16:56 | No...
| |
16:56 | karmic
| |
16:56 | * alkisg hasn't tested on Karmic... | |
16:56 | <ludug3r0> I can't get raso to logon now...
| |
16:56 | Let me try to shutdonw another user...
| |
16:56 | <alkisg> The limit I was talking about is in /etc/ssh/sshd_config, MaxStartups 10:30:60
| |
16:57 | That means that if 11 users try to logon at *exactly* the same time, the 11th will be refused
| |
16:57 | <ludug3r0> strange...
| |
16:57 | the other user hanged up.... :/
| |
16:57 | I dont have 11 users loged at the same time..
| |
16:57 | :/
| |
16:57 | <alkisg> Something is pretty wrong there ...
| |
16:58 | <ludug3r0> But for one user, the system forks two sshds
| |
16:58 | root 3769 0.0 0.1 10224 3288 ? Ss 14:55 0:00 sshd: vak [priv] vak 3825 0.0 0.0 10356 2352 ? S 14:55 0:00 sshd: vak@pts/10
| |
16:58 | root 12211 0.0 0.1 10224 3292 ? Ss 18:47 0:00 sshd: raso [priv] raso 12267 0.0 0.0 10356 2304 ? S 18:47 0:00 sshd: raso@pts/4
| |
17:00 | <alkisg> That's usual, no problem there
| |
17:00 | <ludug3r0> well...
| |
17:00 | should I try with the lucid?
| |
17:00 | <alkisg> If that's an option, yeah, I'd say go for it
| |
17:01 | <ludug3r0> I'm having so much problems with this one I'm almost switch back to localboot machines. :P
| |
17:01 | Ok...
| |
17:01 | <alkisg> Lucid is working fine here
| |
17:01 | <ludug3r0> I will try with lucid...
| |
17:01 | At least I can exchange some ideas with you in abetter way.
| |
17:01 | <alkisg> Some machines had X or upstart problems, but that's unrelated to LTSP
| |
17:01 | <ludug3r0> Can I log around 20? :P
| |
17:01 | How many users can you log in at the same time?
| |
17:02 | <alkisg> (i had to put e.g. nomodeset or remove splash etc)
| |
17:02 | I didn't reach any limits, but I haven't tried with more than 5-6...
| |
17:02 | You should be able to login with some thousand users...
| |
17:03 | <ludug3r0> Ok...
| |
17:05 | I'm looking at the ldm.log now to see it there is some problem...
| |
17:05 | I will try to reproduce the bug..
| |
17:06 | jammcq has quit IRC | |
17:06 | Lns_ has joined #ltsp | |
17:09 | ludug3r0 has quit IRC | |
17:21 | <romel> is it possible to use kerberos to authenticate users are logging into ltsp-based system?
| |
17:29 | romel has quit IRC | |
17:39 | otavio has quit IRC | |
17:43 | <codyg> Never tried kerberos, I did use LDAP for quite a while, I think as long as it is PAM compatible its worth trying.
| |
17:46 | ludug3r0 has joined #ltsp | |
17:47 | <ludug3r0> alkisg: Ok.. I saw ldm.log...
| |
17:47 | I can login just 5 users....
| |
17:48 | When I try to log the sixth, the ldm hangs...
| |
17:48 | At the server it means 12 sshd forks...
| |
17:48 | The log seens fine...
| |
17:48 | There is a connection a the server...
| |
17:48 | <alkisg> If you try ssh at that point, does it succeed?
| |
17:48 | <ludug3r0> But it hangs inside the xsession...
| |
17:49 | There is a black screen with the task bar painted in gray...
| |
17:49 | codyg has quit IRC | |
17:49 | <ludug3r0> Yeah... It sesseds...
| |
17:49 | <alkisg> You may also check ~/.xsessionerrors for that user
| |
17:51 | <ludug3r0> It hangs here....
| |
17:51 | (polkit-gnome-authentication-agent-1:3299): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
| |
17:51 | This is the last line.. Tghis line is also present in sucessfull users....
| |
17:52 | after this.. a sucessfull user gets messages from the gnome-panel.
| |
17:52 | let me try to remove the configurations relatd to the xsession from this user.
| |
17:52 | johnny has left #ltsp | |
17:52 | <ludug3r0> removed .Xauthority too...
| |
17:55 | johnny has joined #ltsp | |
17:55 | <ludug3r0> strange...
| |
17:55 | I got this one running..
| |
17:55 | Now there are 6 machines up.
| |
17:57 | alkisg: do you know how can I switch of pam protections against DoS on the ltsp-server?
| |
17:57 | I'm starting to think this is the problem here...
| |
17:57 | <alkisg> No
| |
17:57 | <ludug3r0> But I don't know...
| |
17:57 | It shouldn't be this...
| |
18:00 | problems with the machine number seven.. Trying again... :\
| |
18:05 | Well... I will try with lucid...
| |
18:05 | alkisg: Thank you very much again. :)
| |
18:05 | <alkisg> ludug3r0: you're welcome, yeah, try with lucid without messing with any pam* stuff :)
| |
18:06 | * alkisg goodnights everyone... | |
18:06 | alkisg has quit IRC | |
18:18 | ludug3r0 has quit IRC | |
18:33 | Lns has quit IRC | |
18:48 | staffencasa has quit IRC | |
18:53 | Lns has joined #ltsp | |
18:53 | Faithful has joined #ltsp | |
19:58 | pmatulis has joined #ltsp | |
20:10 | Faithful has quit IRC | |
20:11 | Barbosa has joined #ltsp | |
20:15 | pmatulis has quit IRC | |
20:17 | try2free has joined #ltsp | |
20:27 | Faithful has joined #ltsp | |
20:32 | artista_frustrad has quit IRC | |
20:43 | Lns has quit IRC | |
20:44 | Faithful has quit IRC | |
20:45 | artista_frustrad has joined #ltsp | |
20:47 | Faithful has joined #ltsp | |
21:02 | Faithful has quit IRC | |
21:17 | try2free has quit IRC | |
21:19 | try2free has joined #ltsp | |
21:21 | Faithful has joined #ltsp | |
21:50 | TheProf has joined #ltsp | |
21:51 | <TheProf> Hello. I hope everyone is doing well today. I have a strange boot-up problem with a brand new thin client we received (first thin client that's not a pentium III or older!!). It is an Acer Aspire 1600 (Intel Atom chip) and
| |
21:52 | I am running k12Linux. Everythng boots up perfectly until the very last step
| |
21:52 | where you should see the login screen but instead the screen just flashes the same thing over and over:
| |
21:53 | "NET: Registered protocol family 10. lo: Disabled Privacy Extensions. Adding 65528k swap on /dev/mapper/swap0. Priority: -1 extents:1 across: 65528k
| |
21:54 | var/log/messages shows only one line that's unusual: Apr 5 22:39:26 ltsp in.tftpd[16429]: tftp: client does not accept options
| |
21:54 | So I'm not sure how to proceed? Any recommendations?
| |
21:55 | <johnny> that's not usually a fatal error
| |
21:55 | i see that one occasionally too
| |
21:56 | you should probably remove the quiet arguments and whatever splash arguments from /var/lib/tftpboot/ltsp/i3866/pxelinux.cfg/default
| |
21:56 | not sure what they are in fedora
| |
21:56 | but you should remove those 2 arguments
| |
21:56 | and boot normally
| |
21:56 | <TheProf> johnny, hello. I will check that file now to see what they are.
| |
21:57 | <johnny> also.. this isn't the best time for support on this channel.. most people are only active during the day of usa/europe
| |
21:57 | <TheProf> This is true-- it just happens it's the only time I could make it to the lab to set up this new machine
| |
21:57 | ok so in the file there is:
| |
21:58 | <johnny> also, a mailing list is good place when this channel is not active
| |
21:58 | <TheProf> johnny, I'm on the mailing list but weirdly, I can't access any Google services -- for some reason the firewall is blocking them all
| |
21:58 | figured I'd fix one issue at a time :0
| |
21:59 | :)
| |
21:59 | ok so there is the argument "quiet" after append in the file default.
| |
21:59 | <johnny> not sure if there is a splash or not
| |
21:59 | <TheProf> no argument for splash or such so I will remove that argument
| |
21:59 | <johnny> but try just rmeoving quiet
| |
21:59 | so you can see it
| |
21:59 | <TheProf> will do so now.
| |
22:01 | rebooting client to see what happens now
| |
22:02 | well is scrolls by VERY fast, I don't think there is a log of that since it's on the client, but everything looked OK.
| |
22:02 | Hmm..actually it looks like the flash you get on a screen when X is having resolution problems
| |
22:02 | that's the best way to describe it
| |
22:11 | <stgraber> TheProf: you may want to set SCREEN_02=shell in your lts.conf and try to grab /var/log/X*
| |
22:11 | (if that's indeed a X issue)
| |
22:12 | <TheProf> stgraber, I was doing exactly that -- just trying to figure out if it's the lts.conf in /var/lib/tftpboot/ltsp/i386/lts.conf or another lts.conf
| |
22:13 | <stgraber> the on in tftpboot is usually the right one
| |
22:13 | <TheProf> stgraber, great. Just set it and rebooting
| |
22:14 | Ok I've got the shell. so I'll look for the file you specified
| |
22:15 | Found an Xorg.log file
| |
22:15 | hmm..last line: Fatal server error: no screens found
| |
22:16 | <stgraber> Is that a laptop ?
| |
22:16 | <TheProf> No it's one of those small desktops
| |
22:16 | the small form factor ones
| |
22:17 | Acer ASpire. specs: http://www.acersupport.com/acerpanam/desktop/2009/acer/aspire/AspireR1600/AR1600sp2.shtml
| |
22:17 | <stgraber> hmm, ok, so it's not one more LVDS issue ;)
| |
22:17 | do you have a ltsp-xorg.conf (or similar) in /var/run/ ?
| |
22:17 | <TheProf> I can't seem to scroll back up on the "cat" output using shift+pgup on the shell
| |
22:19 | in the shell in /var/run there is lts.conf, ldm-xauth-FVGuXuxke and ltspfs_token but that's it
| |
22:19 | <stgraber> ok, that's good, so it's not some screen-session.d doing something weird
| |
22:19 | what happens if you simply start X with: X :1 ?
| |
22:19 | do you get the same error message ?
| |
22:20 | <TheProf> When I put "X :1" I get 3 (EE) lines
| |
22:20 | <stgraber> saying ?
| |
22:20 | <TheProf> 1) Unable to locate/open config file
| |
22:20 | 2) open /dev/fb0: no such file/directory
| |
22:20 | 3) no devices detected
| |
22:21 | then ends with the fatal server error: no screens found
| |
22:21 | <stgraber> hmm, ok ... I'd flag that as an issue with the distro you're using. It's either using a new LTSP with an old X server or has serious driver issues with X ...
| |
22:21 | <TheProf> pardon the delay -- typing on this keyboard is a bit slow.
| |
22:22 | <stgraber> I'm guessing the video card is an intel 945GM ?
| |
22:22 | (if you do lspci -nnn)
| |
22:22 | <TheProf> hmm..well I've been running this server for 1.5 years and about 15 thin clients -- all of them are old hardware. I'll check the video card now
| |
22:23 | shogunx has quit IRC | |
22:23 | <TheProf> it is an nVidia video chip
| |
22:24 | trying to get the specific one now - the flashing screen makes it hard to read
| |
22:24 | according to the specs website it is an NVIDIA ION LE graphics solution
| |
22:24 | <stgraber> doh, I hate these
| |
22:25 | you'll either need a very recent kernel (2.6.33 and higher) to have the nouveau driver that might support it, or have a second chroot with the binary driver installed in it
| |
22:26 | you can't install the driver in your regular chroot because it'll overwrite libGL.so and some other critical files and will break on any non-nvidia hardware after that
| |
22:26 | <TheProf> ew
| |
22:27 | <stgraber> at least we know what was wrong now ... I really liked when Atom-based hardware all had 945GM from Intel, these simply always work great ...
| |
22:27 | <TheProf> I assume you're talking about the Kernel associated with the ltsp-function, not the Fedora kernel?
| |
22:27 | this isn't something I can just yum-update can I? :)
| |
22:27 | <stgraber> I had to decline some new thin clients recently because they were ION-based and then force Asus into sending us some older thin clients that were still using the intel chipset
| |
22:27 | Lns has joined #ltsp | |
22:28 | <stgraber> TheProf: it's the fedora kernel you'll need to upgrade to something very recent (as in, maybe not even stable yet ;))
| |
22:28 | <TheProf> This was a donated machine for us, so it makes it a bit hard to turn down.
| |
22:28 | <stgraber> TheProf: I'd suggest that you actually try to run fedora on it directly and not through LTSP
| |
22:28 | TheProf: once you get it working, then you'll be able to figure out how to make it work with LTSP
| |
22:29 | I'm guessing fedora should have some way to at least give you a working VESA driver (though X will really be slow and uncaccelerated with that)
| |
22:30 | <TheProf> I can't force it to use even an older driver through specifying something in lts.conf for it's MAC, even to just get it to log in?
| |
22:30 | <stgraber> you can try forcing vesa
| |
22:30 | though X is supposed to always try it as fallback
| |
22:31 | <TheProf> I think I'll try that now.
| |
22:32 | wow it worked!
| |
22:33 | <stgraber> hehe, now you have a working thin client that's maybe slower than your old ones, thanks to nvidia ;)
| |
22:33 | <TheProf> it's funny because in terms of processor power, etc this thin client is faster than our server :)
| |
22:34 | <stgraber> yeah and with an intel video card it'd be extremely fast giving you a desktop-like experience (on the ones I have at the office which have intel video card, I run compiz, dual-head setup and 3D applications just like a normal desktop machine)
| |
22:35 | <TheProf> argh. so much for getting newer hardware being better. I am sure the people donating were thinking they were doing us a favour by getting us the latest and greatest
| |
22:41 | stgraber, do you know if there is another Xorg driver that's better than vesa, like the next step up?
| |
22:46 | <stgraber> that'd be nv (though I don't think it works with the ION chipsets), then nouveau (but it might not be in your current kernel), then the binary driver (with the other issues I mentioned)
| |
22:49 | jammcq has joined #ltsp | |
22:49 | <jammcq> hey all
| |
22:49 | <TheProf> I tried nv but it didn't boot. I didn't try nouveau
| |
22:49 | I was so excited to get a new thin client from this century (literally) but it doesn't work
| |
22:50 | jammcq, Hello.
| |
22:50 | <jammcq> howdie
| |
22:51 | <TheProf> thanks to stgraber's help was able to figure out the reason the Acer was not booting -- nVidia ion driver not supported.
| |
22:55 | stgraber, thanks for your help-- I will try posting to the mailing list to see if anyone has experience with these machines.
| |
22:55 | shogunx has joined #ltsp | |
22:58 | <TheProf> Goodbye.
| |
22:58 | TheProf has left #ltsp | |
23:04 | jammcq has quit IRC | |
23:41 | randomuser has joined #ltsp | |
23:43 | <randomuser> I just installed k12linux packages on my f12 system, started ltsp-build-client, and realized that it was building a 32bit client. how do i start over ? i see i can append --arch x86_64 , but what about the old build?
| |