00:02 | sweetpi has quit IRC | |
00:03 | gentgeen__ has quit IRC | |
00:03 | gentgeen__ has joined #ltsp | |
00:03 | sweetpi has joined #ltsp | |
00:15 | M4gic5t0rM has quit IRC | |
00:16 | sweetpi has quit IRC | |
00:20 | sweetpi has joined #ltsp | |
00:48 | cyberorg has quit IRC | |
01:07 | cyberorg has joined #ltsp | |
01:38 | ogra has quit IRC | |
01:53 | cyberorg has quit IRC | |
01:53 | ogra has joined #ltsp | |
01:54 | cyberorg has joined #ltsp | |
01:55 | map7 has quit IRC | |
02:02 | moose-tek has quit IRC | |
02:12 | alkisg has joined #ltsp | |
02:17 | dlezcano has quit IRC | |
02:50 | bobby_C has joined #ltsp | |
03:01 | Da-Geek has joined #ltsp | |
03:13 | sweetpi has quit IRC | |
03:17 | ogra_cmpc has quit IRC | |
03:23 | Selveste1____ has joined #ltsp | |
03:23 | Selveste1___ has quit IRC | |
03:30 | ogra_cmpc has joined #ltsp | |
03:41 | Selveste1____ has quit IRC | |
03:42 | Selveste1____ has joined #ltsp | |
04:00 | otavio has quit IRC | |
04:34 | dobber has joined #ltsp | |
05:03 | litlebuda has quit IRC | |
05:05 | litlebuda has joined #ltsp | |
05:15 | pmatulis has joined #ltsp | |
05:26 | alkisg has quit IRC | |
06:14 | wwx has quit IRC | |
06:15 | wwx has joined #ltsp | |
06:15 | Da-Geek has quit IRC | |
06:23 | ogra has quit IRC | |
06:27 | ogra has joined #ltsp | |
06:32 | cyberorg has quit IRC | |
06:35 | cyberorg has joined #ltsp | |
06:39 | elias_a has quit IRC | |
06:48 | cyberorg_ has joined #ltsp | |
06:50 | cyberorg has quit IRC | |
06:56 | otavio has joined #ltsp | |
06:59 | dlezcano has joined #ltsp | |
07:04 | Blinny has joined #ltsp | |
07:07 | elias_a has joined #ltsp | |
07:11 | alkisg has joined #ltsp | |
07:12 | <Blinny> alkisg: Well, I'm on the old kernel. I should know by this time tomorrow if it's a kernel issue.
| |
07:13 | <alkisg> Blinny: ok. Also, if you want, paste the output of this to pastebin: tcpdchk -v
| |
07:14 | dlezcano has quit IRC | |
07:14 | <alkisg> Urm never mind I thought that also listed the keepalive setting but it doesn't - post hosts.allow if you want
| |
07:15 | <Blinny> heh yeah you don't want my hosts.deny -- I use denyhosts
| |
07:16 | http://pastebin.com/QPvcFYEr
| |
07:19 | <alkisg> Yeah looks ok, I wonder if there's a way to check an already made connection to see if it has keep_alive = on...
| |
07:27 | dlezcano has joined #ltsp | |
07:38 | <alkisg> Btw, I wonder why we don't use `exec` on nbdrootd, that would save us from all the nbdrootd processes using up RAM...
| |
07:43 | stgraber, ogra, any reasons for not putting `exec` on the last line of nbdrootd? Seems to work fine here...
| |
07:43 | PATH=/usr/bin:/bin:$PATH exec nbd-server 0 $1 -r -C /dev/null $nbd_timeout > /dev/null 2>&1
| |
07:43 | denisesball has quit IRC | |
07:52 | cyberorg_ is now known as cyberorg | |
08:21 | wurzzero has joined #ltsp | |
08:26 | <wurzzero> Hi, is there a way to set the 'PRINTER_0_FLOWCTRL' to DTR/DSR? I saw that we have the Software (XON/XOFF) and Hardware (CTS/RTS) options...
| |
08:29 | dobber has quit IRC | |
08:31 | vagrantc has joined #ltsp | |
08:32 | <alkisg> wurzzero: you probably need to patch /opt/ltsp/i386/usr/sbin/jetpipe for that. See http://pyserial.sourceforge.net/pyserial_api.html about dsrdtr.
| |
08:32 | vagrantc: I'm thinking of putting an `exec` to the last line of nbdrootd, can you see any problems with that? Seems to work fine here...
| |
08:32 | PATH=/usr/bin:/bin:$PATH exec nbd-server 0 $1 -r -C /dev/null $nbd_timeout > /dev/null 2>&1
| |
08:33 | (to save us from all the nbdrootd processes + RAM)
| |
08:33 | <wurzzero> all right, i'll read that, tanks!
| |
08:33 | <vagrantc> alkisg: should be fine
| |
08:33 | <alkisg> OK, committing...
| |
08:34 | <vagrantc> alkisg: nbdswapd needs to run the cleanup bits... but nbdrootd doesn't
| |
08:34 | alkisg: if we ever switched to supporting the writeable cow file stuff, we might have to change that
| |
08:34 | <alkisg> Right, I won't do that for nbdswapd (although rm while the file is being used works, but it's too much of a hack :))
| |
08:35 | <vagrantc> alkisg: also, the newest version of nbd supports a single port with multiple images...
| |
08:35 | <alkisg> vagrantc: writeable cow stuff? you mean nbd-server -c?
| |
08:36 | vagrantc: about multiple images, aren't those only separated by the client IP? I.e. we can't use that, can we?
| |
08:38 | I once filed a feature request about it, and he said he'd like to implement it, https://bugs.launchpad.net/ubuntu/+source/nbd/+bug/593227
| |
08:40 | <vagrantc> alkisg: the new version supports the client requesting a particular image
| |
08:40 | <alkisg> Yey!!! /me looks...
| |
08:44 | nbd-server -c: Copy on write. ...This separate file is removed when the connection is closed...
| |
08:45 | So it's removed automatically, `exec` would still be valid there afaik
| |
08:45 | <vagrantc> ah, cool
| |
08:45 | <alkisg> Also, I wonder if we can use the same thing for swap files
| |
08:45 | Example: make a 1GB swap file in /tmp, and export the same file with -c in all the clients
| |
08:46 | They'll make temp files with only the changed data ==> small
| |
08:46 | Then, when the connection is closed, they'll be automatically deleted
| |
08:46 | <vagrantc> i think it adds considerable overhead
| |
08:46 | <alkisg> Sounds cool, but yeah that ^^^
| |
08:46 | <vagrantc> but the automatic deleting might be worth it
| |
08:46 | <alkisg> Also automatic size
| |
08:46 | (up to the max swap size we prepared)
| |
09:07 | dobber has joined #ltsp | |
09:10 | Gadi has joined #ltsp | |
09:18 | <alkisg> vagrantc: hdparm -tT /dev/nbd2 (==swap file using nbd-server -c) ==> 3.39 MB/sec
| |
09:18 | while hdparm -tT /dev/nbd0 (==swap file using nbdswapd) ==> 3.23 MB/sec
| |
09:18 | So I don't see any performance loss there. And also we save a lot of /tmp space on the server, as the .diffs are initially empty, until used.
| |
09:18 | <vagrantc> brutak!
| |
09:18 | brutal!
| |
09:18 | <alkisg> ...and they automatically get deleted
| |
09:18 | It sounds good to me...
| |
09:19 | We just need to mkswap /tmp/ltspswap.img once when the server boots (or when nbdswapd is first called, if no race conditions can happen there)
| |
09:20 | <vagrantc> need to use mktemp if you're using /tmp
| |
09:20 | <alkisg> Is there any problem if we create a very big swap file? E.g. 1 Gb? Only one needs to be created, the rest are just diffs...
| |
09:21 | (so we don't need to keep it low like we do now with 64MB, we may as well use it for fat clients too)
| |
09:21 | <vagrantc> can't you use an empty file 1GB long?
| |
09:21 | <alkisg> Well e.g. now 64MB x 20 clients ==> bigger than that
| |
09:21 | What's a good size? 512? 256?
| |
09:22 | Ah sorry misread
| |
09:22 | Yes, that 1 GB empty file was what I was talking about, just one file with mktemp / dd 1 Gb / mkswap
| |
09:23 | <vagrantc> but you can create a file that takes no space, but is treated as 1GB
| |
09:23 | <alkisg> Do all file systems support that?
| |
09:23 | <vagrantc> dd if=/dev/zero bs=1024k count=0 seek=1000
| |
09:23 | alkisg: don't know
| |
09:23 | <alkisg> Ah, yes, that's how I did it now, I didn't know seek is used for sparse files
| |
09:24 | <vagrantc> alkisg: oh, that's what we already do
| |
09:24 | sparse files, that's what it's call
| |
09:24 | <alkisg> The code in nbdswapd only needs a few minor changes to handle nbd-server -c
| |
09:24 | (e.g. not rm in the end)
| |
09:24 | But the most important question is, should nbd-server -c be the default?
| |
09:25 | Any downsides?
| |
09:27 | <vagrantc> depends how stable/reliable it is...
| |
09:27 | i'd want more real-world testing before making it the default
| |
09:28 | anyways, gotta run...
| |
09:28 | * vagrantc waves | |
09:28 | vagrantc has quit IRC | |
10:05 | * alkisg noted down that idea in https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/613868 | |
10:14 | Selveste1____ has quit IRC | |
10:14 | staffencasa has joined #ltsp | |
10:21 | Selveste1 has joined #ltsp | |
10:29 | Blinny has quit IRC | |
10:41 | vagrantc has joined #ltsp | |
10:44 | lordzhao has joined #ltsp | |
10:45 | <lordzhao> boa tarde a todos
| |
10:47 | alguem poderia me ajudar?
| |
10:47 | <vagrantc> this channel is mostly in english
| |
10:47 | there are a few portuguese speakers, though
| |
10:48 | <lordzhao> ok
| |
10:49 | ok, then in English, the translator of google :)
| |
10:49 | <vagrantc> oh boy
| |
10:49 | :)
| |
10:50 | <lordzhao> I need help with network dm9102 can someone help?
| |
10:51 | always of a error "dhcpcd fail"
| |
10:54 | <alkisg> https://help.ubuntu.com/community/UbuntuLTSP/AddingModules ?
| |
10:54 | vagrantc: I didn't see any new support for different images in the nbd git... only the old support that is based on the client ip, which we can't really use
| |
10:54 | <vagrantc> lordzhao: what linux distro? what version?
| |
10:56 | <lordzhao> use a distro based on debian
| |
10:56 | <vagrantc> lordzhao: i don't know any that use dhcpcd ...
| |
10:57 | [GuS] has joined #ltsp | |
10:57 | <vagrantc> lordzhao: what distro, specifically?
| |
10:57 | <lordzhao> | |
10:59 | I think the problem with port 67
| |
10:59 | is always much time on it ... when will catch for the second time ip
| |
11:00 | alkisg has quit IRC | |
11:02 | <lordzhao> Running dhcpd on port 67
| |
11:02 | ERROR! dhcpd failed!
| |
11:02 | Kernel panic - not syncing: Attempted to kill init
| |
11:04 | <vagrantc> ah
| |
11:05 | lordzhao: those errors all on the server, or on the thin client?
| |
11:05 | or both?
| |
11:06 | <lordzhao> thin client
| |
11:07 | other NIC works normally
| |
11:07 | davicom dm9102 not
| |
11:07 | <vagrantc> dhcpd shouldn't start on the thin client
| |
11:07 | that should be running on the server
| |
11:09 | <lordzhao> sorry, the error is actually dhcpcd fail
| |
11:10 | <vagrantc> and dhcpcd isn't normally installed on LTSP
| |
11:10 | at least on debian based systems
| |
11:10 | so something seems very unusual here
| |
11:11 | lordzhao: can you paste the output of "ltsp-info" to http://ltsp.pastebin.com
| |
11:15 | lordzhao has quit IRC | |
11:21 | lordzhao has joined #ltsp | |
11:24 | * Gadi guesses this is LTSP 4.2 that used dhcpdc | |
11:24 | * Gadi waves to vagrantc | |
11:27 | <lordzhao> know what can be?
| |
11:31 | * vagrantc waves to Gadi | |
11:31 | <vagrantc> aha, ltsp 4 ...
| |
11:31 | <Gadi> lordzhao: are you still on LTSP 4?
| |
11:32 | <lordzhao> yes
| |
11:32 | <Gadi> if so, the driver for that NIC is most likely not present in the initramfs
| |
11:32 | or it does not correctly detect the correct driver
| |
11:32 | what NIC?
| |
11:32 | ah, davicom dm9102
| |
11:32 | do you know what driver it uses?
| |
11:34 | <lordzhao> I believe it is the tulip...know how I can solve?
| |
11:34 | <Gadi> does it PXE or etherboot?
| |
11:35 | if it is PXE, you can try adding NIC=tulip in the appropriate pxlinux.cfg/ file
| |
11:35 | if it is etherboot, you can try adding NIC=tulip to the appropriate option in dhcpd.conf
| |
11:37 | if those don't work, then you may need to find the correct driver, get it into the initramfs and edit the "niclist" in the initramfs to load it based upon PCI ID
| |
11:37 | man, this takes me back....
| |
11:37 | ;)
| |
11:37 | bbiab
| |
11:38 | <lordzhao> Gadi, really what I saw in all forums...but is not working
| |
11:46 | Selveste1_ has joined #ltsp | |
11:46 | vagrantc has quit IRC | |
11:47 | Selveste1 has quit IRC | |
11:47 | <lordzhao> Gadi, Thank you for your attention and help ... I have to leave now but come back later.
| |
11:48 | lordzhao has quit IRC | |
11:48 | monteslu has quit IRC | |
11:54 | irule has joined #ltsp | |
12:06 | dobber has quit IRC | |
12:06 | alkisg has joined #ltsp | |
12:09 | monteslu has joined #ltsp | |
12:31 | Selveste1_ has quit IRC | |
12:33 | Selveste1_ has joined #ltsp | |
12:51 | Selveste1__ has joined #ltsp | |
12:53 | Selveste1_ has quit IRC | |
13:00 | vagrantc has joined #ltsp | |
13:04 | Damianos has joined #ltsp | |
13:05 | <Damianos> Installation question: Installing from the Karmic Alt CD, how do I set my wireless card as primary connection? I don't have a secondary NIC and I need the gigabit ethernet mobo port for the LTSP LAN
| |
13:07 | <alkisg> Set up static IPs and change /etc/ltsp/dhcpd.conf
| |
13:07 | (or you can use eth0 for both ltsp clients + internet)
| |
13:07 | dekelev has joined #ltsp | |
13:08 | <Damianos> ok
| |
13:08 | thanks a million!
| |
13:09 | What if I just skip the network detection phase? Will that still set up the LTSP on the ethernet port?
| |
13:10 | <dekelev> hi
| |
13:12 | <alkisg> Damianos: I'm not sure, but whatever you choose, all you have to do later on is check /etc/ltsp/dhcpd.conf
| |
13:12 | ...and /etc/network/interfaces. So no big deal.
| |
13:12 | <dekelev> is there a way to configure a default audio device for the clients to use, in case they have it plugged in ?
| |
13:12 | komunista has joined #ltsp | |
13:13 | <dekelev> or save every user's audio preferences so it will stay the same after reboot ?
| |
13:21 | <Damianos> @alkisg: ok, gotcha thanks for your help!
| |
13:25 | <dekelev> any ideas? i'm lost
| |
13:26 | <alkisg> dekelev: well I know you can setup some audio device, but not conditionally. I believe you'd need to write a script for this, and put it in a RC_FILE directive in lts.conf
| |
13:28 | <vagrantc> alkisg: i guess the nbd filename request stuff wasn't yet added
| |
13:29 | alkisg: might be included in nbd 2.9.17
| |
13:29 | <alkisg> vagrantc: I didn't see anything in git though... I wish it will
| |
13:30 | nbd-server -c for nbdrootd + a standarized port would be good to have
| |
13:30 | (cow also saves some ram on the client, as no tmpfs is needed)
| |
13:30 | <vagrantc> alkisg: he got a port from IANA for this purpose
| |
13:30 | <alkisg> Yey!
| |
13:30 | Which one?
| |
13:30 | <vagrantc> don't know
| |
13:31 | irule has quit IRC | |
13:39 | Appiah has quit IRC | |
13:39 | Appiah has joined #ltsp | |
13:40 | <dekelev> alkisg: thanks, I'm trying to mess with alsa-base.conf to configure usb sound as default
| |
13:40 | <alkisg> dekelev: I believe you can do that from lts.conf. You just can't conditionally select one or the other...
| |
13:41 | <dekelev> realy?
| |
13:41 | I check the lts.conf manual for this
| |
13:41 | checked
| |
13:41 | didn't find it
| |
13:45 | <alkisg> Ah, you're right, -c0 (first device) is hardcoded there.
| |
13:47 | komunista has quit IRC | |
14:00 | mgariepy has joined #ltsp | |
14:01 | komunista has joined #ltsp | |
14:02 | irule has joined #ltsp | |
14:13 | <Damianos> OK I'm kind of lost, wht exactly am I supposed to be looking for in the dhcp config file? All it says is primary interface is eth0
| |
14:17 | sweetpi has joined #ltsp | |
14:26 | [GuS] has quit IRC | |
14:28 | <Kyle__> Humm
| |
14:29 | Are there any good examples of how to setup thick clients? I've done the server setup (I think), but I'm not seeing much in the way of documenting the client setup at this point...
| |
14:29 | <alkisg> Damianos: /etc/ltsp/dhcpd.conf ==> the ip range there
| |
14:30 | Kyle__: the windows PCs?
| |
14:30 | Do they support "boot from network"?
| |
14:30 | If so, you don't need anything else on the clients...
| |
14:32 | <Kyle__> alkisg: Errr. In my setup and resetup etc, I don't appear to have a pxelinux.0 file anymore
| |
14:32 | <alkisg> Ah. Try `ltsp-update-kernels`
| |
14:33 | irule has quit IRC | |
14:34 | <Kyle__> Humm. Maybe my conf file is screwie.... re-reading.
| |
14:36 | nevermind. Those are relative paths to the ltftpd server... OK, he goes nofin.
| |
14:39 | alkisg: is ltsp-update-kernels supposed to take care of copying over the kernel to the tftpboot directory?
| |
14:39 | <alkisg> Yes
| |
14:39 | (of course if the chroot is already build and everything else is in order)
| |
14:39 | <Kyle__> It didn't copy over vmlinuz for some reason. Maybe my chroot isn't right.
| |
14:41 | <alkisg> ls -l /opt/ltsp/i386/boot
| |
14:44 | <Kyle__> Yea, there was no kernel there. I'm chrooted into it now to see why there's no kernel installed :)
| |
14:45 | Other than installing linux-image-xyzversion-generic, any suggestions on kernels?
| |
14:46 | <alkisg> I think linux-image-generic is more appropriate, that will pull the most recent version. But why wasn't it installed, and what else is missing?
| |
14:47 | <Kyle__> Ah ok....
| |
14:47 | Well, apparently the grub and firmware packages are missing too, since it's adding those now.
| |
14:48 | <alkisg> Maybe ltsp-build-client broke for some reason, and you need to run it from the start?
| |
14:49 | <Kyle__> Maybe...Not sure. I'll just wait and see how this one goes now.
| |
14:50 | M4gic5t0rM has joined #ltsp | |
14:50 | <Kyle__> So after this, I run ltsp-update-image && ltsp-update-kernels. After normal changes just ltsp-update-image is enough right?
| |
14:51 | <alkisg> ltsp-update-image --force should also take care of the kernels
| |
14:52 | <Kyle__> Sweet.
| |
14:52 | So I just treat the chroot however I would a normal machine, rebuild, and reboot the clients, and it's done. I like this.
| |
14:54 | <alkisg> Right. And with nfs, or with an nbd partition, no rebuilding is even needed...
| |
14:54 | <Kyle__> I take it because I have an /opt/ltsp/i386/images/*.img file I'm using an nbd file not a partition?
| |
14:55 | <alkisg> Yup, that's the normal way to go - it's just a little time consuming with fat chroots
| |
14:55 | M4gic5t0rM has quit IRC | |
14:56 | <Kyle__> OK. Well I'll get this going, then work on the fancy stuff :)
| |
14:57 | cascade has joined #ltsp | |
15:02 | cascade has quit IRC | |
15:12 | M4gic5t0rM has joined #ltsp | |
15:18 | otavio has quit IRC | |
15:18 | otavio has joined #ltsp | |
15:21 | tstafford_ has quit IRC | |
15:22 | * Kyle__ grumbles. Stupid small partition. | |
15:23 | <Kyle__> Died at 200mb to go cause the part was too small. At least it's not hard to make it larger. Grr.
| |
15:25 | <alkisg> Maybe that's why ltsp-build-client failed, too
| |
15:25 | M4gic5t0rM has quit IRC | |
15:30 | M4gic5t0rM has joined #ltsp | |
15:31 | <Kyle__> Eh, I don't think so. It was almost big enough to hold both.
| |
15:32 | shogunx has joined #ltsp | |
15:45 | <Kyle__> OK I had to create an nbd-server config file, maybe it's not quit right, cause it's still complaining that it can't connet to the NBD server.
| |
15:45 | M4gic5t0rM has quit IRC | |
15:45 | <alkisg> Kyle__: no, delete it, nbd-server starts from inetd
| |
15:45 | /etc/inetd.conf
| |
15:45 | <Kyle__> Errr. oh.
| |
15:46 | Out of curiocity, why?
| |
15:46 | curiosity rather.
| |
15:46 | Ahh, it had an old image name in there.. easy fix
| |
15:50 | <alkisg> inetd saves memory when nbd-server is not used, provides some security with /etc/hosts.allow/deny, allows easy scripting (nbdrootd/nbdswapd) because the sockets are redirected to stdin etc
| |
15:51 | otavio has quit IRC | |
15:51 | otavio has joined #ltsp | |
15:52 | tstafford has joined #ltsp | |
15:55 | <Kyle__> Fair enough. So if I shared a partition, does it have to be a block device like sdb2, or can it be an LVM partition?
| |
15:56 | * Kyle__ swears at himself: no users on that image. | |
15:57 | <alkisg> You use the server users
| |
15:57 | You don't add user accounts to the chroot
| |
15:57 | <Kyle__> Fat client.
| |
15:57 | tstafford has quit IRC | |
15:57 | <alkisg> No difference
| |
15:57 | Still, user accounts are on the server
| |
15:57 | <Kyle__> Odd. It doesn't see any of the servers users.
| |
15:57 | <alkisg> ?
| |
15:57 | Are you using gdm?
| |
15:58 | <Kyle__> I can try again, but it didn't the two times i tried.
| |
15:58 | Err, it's a graphical login manager, doesn't look like ubuntu's stock GDM, but I suppose it could be
| |
15:58 | <alkisg> Ah, you mean that username/password fails?
| |
15:58 | No, that's LDM
| |
15:58 | Check auth.log on the server while trying to login on the client
| |
15:59 | <Kyle__> Saying failed due to unknown user id
| |
15:59 | But that was on the screen, checking the log now.
| |
16:00 | <alkisg> No that "unknown user id" message comes from pulseaudio
| |
16:00 | <Kyle__> Oh.
| |
16:01 | the auth log doesn't seem to have anything recent enough to be from the clients.
| |
16:01 | <alkisg> Then try creating an lts.conf file to debug this, ie.
| |
16:02 | /var/lib/tftpboot/ltsp/i386/lts.conf with:
| |
16:02 | [Default]
| |
16:02 | SCREEN_02=shell
| |
16:02 | <vagrantc> originally ran nbd* from inetd because running nbd-server directly with multiple connections just crashed
| |
16:02 | <alkisg> SCREEN_07=ldm
| |
16:02 | Ouch
| |
16:02 | <vagrantc> though i wonder if that's still the case with newer versions...
| |
16:02 | <Kyle__> Hu.
| |
16:02 | <vagrantc> that was a *long* time ago
| |
16:03 | <Kyle__> If I read correctly, I don't have to do anything but reboot the clients for that type of change, correct?
| |
16:03 | <alkisg> Kyle__: yup
| |
16:03 | mgariepy has quit IRC | |
16:03 | <alkisg> Alt+ctrl+f2, and type: ssh user
| |
16:03 | ssh user@server
| |
16:04 | Change "user" to an existing username, leave "server" exactly as it is.
| |
16:05 | <Kyle__> I take it specifygin SCREEN_## it's just telling it what binary to run? So we just tossed a root shell in there, didn't we?
| |
16:05 | <alkisg> Kinda - these are special "screen scripts", you can't put arbitrary commands there
| |
16:06 | <Kyle__> Ahh. Well the SSH worked, but df -h as root there kindly let me know that I don't have /home mounted.
| |
16:07 | Which would cause that problem in most grpahical login managers.
| |
16:07 | <alkisg> It'll be mounted with sshfs
| |
16:07 | Did you see any messages/warnings at all with that ssh line?
| |
16:07 | tstafford has joined #ltsp | |
16:07 | <Kyle__> No, it just worked.
| |
16:08 | Oh. Humm. So it shoul dhave just worked, but didn't.
| |
16:08 | <alkisg> Hm. OK, switch to vt7 and try from LDM
| |
16:08 | Ah, ok
| |
16:09 | <Kyle__> Trying again with LDM. This time I'm seeing messages about creating .Xauthority files, etc....
| |
16:09 | I think it's working.
| |
16:10 | Wait...by default it's using sshfs? I never setup ssh keys. That would probably be the issue.
| |
16:11 | tstafford has quit IRC | |
16:12 | <alkisg> Those are done automatically
| |
16:12 | <Kyle__> Then I'm confused. Because when I sshed it did the usual first time auth stuff.
| |
16:13 | <alkisg> Then ltsp-update-sshkeys && ltsp-update-image is needed
| |
16:13 | <Kyle__> OK.
| |
16:16 | Thank you so much for this help.
| |
16:17 | <alkisg> np
| |
16:21 | shogunx has quit IRC | |
16:23 | tstafford has joined #ltsp | |
16:24 | vagrantc has quit IRC | |
16:27 | <Kyle__> Sweet, it's done. So does flipping the image require more than bouncing the clients?
| |
16:32 | tstafford has quit IRC | |
16:32 | alkisg has quit IRC | |
16:34 | vagrantc has joined #ltsp | |
16:34 | wurzzero has quit IRC | |
16:35 | tstafford has joined #ltsp | |
16:36 | Geeky1 has joined #ltsp | |
16:36 | <Geeky1> Hello
| |
16:37 | <hfrhryu34> HI
| |
16:37 | <Geeky1> If I don't want to PBX boot, is there a way to write the chroot image to local read-only media?
| |
16:38 | Sorry, mean PXE boot. :)
| |
16:40 | <Kyle__> In the exact opposit of normal: I now have a lab of 3Ghz P4's with gigabit networking, fat-client booting from a 900Mhz PIII server with a 100baseT card....
| |
16:40 | But it's still cool.
| |
16:41 | This is awesome.
| |
16:41 | vagrantc has quit IRC | |
16:41 | <Geeky1> Kyle__, Cool
| |
16:42 | <Kyle__> Geeky1: unfortunately I don't know enough to help you with that, but wait around!
| |
16:47 | Humm. 5 out of the 10 I booted like this are up and fine, but five are just kindof spinning in the ubuntu boot process?
| |
16:50 | <johnny> turn off quiet splash
| |
16:50 | in the pxelinux.cfg/default file
| |
16:50 | Geeky1, why wouldn't you want to pxeboot?
| |
16:59 | shogunx has joined #ltsp | |
17:04 | <Kyle__> johnny: I expected that to more tell me the problem, but it actually fixed it for 17 out of the 18 booted..
| |
17:05 | The one that isn't happy is complaining about an rsyslog timeout (sorry on the other side of the room, don't recall the exact message).
| |
17:05 | * Kyle__ is just going to bounce that one again. | |
17:05 | <johnny> hah
| |
17:06 | Geeky1 has quit IRC | |
17:08 | <Kyle__> Very neat.
| |
17:08 | Humm. So it's using sftp for the sshfs?
| |
17:09 | You realize that there's a known bug in the openssh sftp server. Not a security one, but a performance one. It's abysmally slower than scp through the same system.
| |
17:11 | lordzhao has joined #ltsp | |
17:12 | Geeky1 has joined #ltsp | |
17:13 | DanaG has joined #ltsp | |
17:13 | <DanaG> Say, how well does ltsp support ARM clients? Most tutorials I can find use the "create root" thingy for x86.
| |
17:17 | <johnny> good question :)
| |
17:18 | the person who can answer isn't here
| |
17:18 | <Kyle__> DanaG: I'm only just starting with it myself, but from what I see, you need to build the image for the target system on the target system... after that I think you should be able to serve it from any system.
| |
17:18 | That help?
| |
17:18 | <johnny> you could need to create it from arm yes
| |
17:18 | would*
| |
17:18 | unless you setup some emulator
| |
17:18 | ogra, are you here?
| |
17:19 | kinda late in germany.. but he might be here
| |
17:19 | <DanaG> Ah. I would have local storage -- an SD card -- for the root on the ARM.
| |
17:19 | <johnny> you don't need it
| |
17:19 | just serve it pxe like normal if posible
| |
17:27 | Geeky1 has quit IRC | |
17:28 | Geeky1 has joined #ltsp | |
17:28 | chupacabra has quit IRC | |
17:34 | Gadi has left #ltsp | |
17:36 | japerry has joined #ltsp | |
17:44 | DanaG has quit IRC | |
17:52 | lordzhao has quit IRC | |
17:52 | DanaG has joined #ltsp | |
18:03 | DanaG has quit IRC | |
18:03 | DanaG1 has joined #ltsp | |
18:03 | DanaG1 is now known as DanaG | |
18:07 | DanaG has quit IRC | |
18:16 | lordzhao has joined #ltsp | |
18:17 | komunista has quit IRC | |
18:18 | lordzhao has quit IRC | |
18:23 | lordzhao has joined #ltsp | |
18:25 | <lordzhao> how do I remove the kernel module tulip ltsp?
| |
18:35 | staffencasa has quit IRC | |
18:35 | <lordzhao> ?
| |
18:40 | shamino has quit IRC | |
18:41 | <lordzhao> how do I remove a module of the network card?
| |
18:44 | shamino has joined #ltsp | |
18:58 | japerry has quit IRC | |
18:58 | <lordzhao> algum brasileiro?
| |
19:03 | lordzhao has quit IRC | |
19:30 | ogra_cmpc has quit IRC | |
19:30 | ogra has quit IRC | |
19:31 | vagrantc has joined #ltsp | |
19:32 | Selveste1___ has joined #ltsp | |
19:32 | Selveste1__ has quit IRC | |
19:33 | pmatulis has quit IRC | |
19:37 | ogra_cmpc has joined #ltsp | |
19:37 | vagrantc has quit IRC | |
19:38 | vagrantc has joined #ltsp | |
19:39 | ogra has joined #ltsp | |
19:39 | * vagrantc waves to ogra | |
19:43 | ogra_cmpc has quit IRC | |
19:47 | Geeky1 has quit IRC | |
19:56 | ogra_cmpc has joined #ltsp | |
20:03 | sweetpi has quit IRC | |
20:21 | dlezcano has quit IRC | |
20:47 | EDinNY has joined #ltsp | |
20:48 | <EDinNY> does JAM stop by anymore?
| |
20:48 | Anyone at the Linux show in Boston next week?
| |
21:40 | <vagrantc> EDinNY: jim mcquillan?
| |
21:40 | !seen jammcq
| |
21:40 | <ltspbot`> vagrantc: jammcq was last seen in #ltsp 2 days, 23 hours, 49 minutes, and 43 seconds ago: <jammcq> hey friends
| |
21:57 | M4gic5t0rM has joined #ltsp | |
22:22 | litlebuda has quit IRC | |
22:24 | litlebuda has joined #ltsp | |
22:26 | japerry has joined #ltsp | |
22:34 | chupacabra has joined #ltsp | |
22:44 | Damianos has quit IRC | |
22:45 | Damianos has joined #ltsp | |
22:47 | M4gic5t0rM has quit IRC | |
22:48 | Damianos has quit IRC | |
22:49 | Damianos has joined #ltsp | |
23:11 | alexqwesa has quit IRC | |
23:24 | vagrantc has quit IRC | |
23:36 | try2free has joined #ltsp | |
23:40 | try2free has left #ltsp | |
23:53 | RiXtEr has joined #ltsp | |