IRC chat logs for #ltsp on irc.freenode.net (webchat)


Channel log from 30 November 2009   (all times are UTC)

00:07alkisg has joined #ltsp
00:55shawnp0wers has quit IRC
01:14Appiah_ is now known as Appiah
01:18FGXR6 has joined #ltsp
01:24Sarten-X has joined #ltsp
01:27mikkel has joined #ltsp
01:31F-GT has quit IRC
01:32mikkel has quit IRC
01:39mikkel has joined #ltsp
01:59foxbuntu has quit IRC
02:07foxbuntu has joined #ltsp
02:25vagrantc has quit IRC
02:35lucascoala has quit IRC
02:44
<lupine_85>
I'm hunting for places where netcat gets called in the initialisation of an LDM vt. I've found one in usr/share/ltsp/screen.d/ldm - anyone know of any other instances?
02:44
I seem to have a race condition... somewhere... that causes initialization of the X session to fail before the X server is started with nc complaining about an invalid option 'f' being passed
02:45
<alkisg>
/client/xinitrc.d/I00-cluster
02:45
<lupine_85>
( nc -w 5 $SRV 9571 > /var/run/ldm/$SRV is the one I've found so far - doesn't look like the right candidate, though I've taken to caching the ldm file in question)
02:45
<alkisg>
/client/screen-x-common
02:45
/screen.d/ldm
02:46
/rc.d/P00-ltsp-cluster
02:46
/rc.d/S99-ltsp-cluster
02:46
That's what grepping the source tells me...
02:48
<lupine_85>
I guess that would have been the most sensible option ;). Unless the $SRV variable is being set incorrectly, I'm not sure where it's getting 'f' from
02:55
<alkisg>
Well I guess you could replace nc with a script that would first echo a debugging message/spawn a shell etc and then call the actual nc ;)
02:57
(or better yet, first call nc and only stop if there's been an error)
02:57
<lupine_85>
oh, wait, got it
02:57
/var/run/ldm contains a file named "-f"
02:58
<alkisg>
Heh, so $SVR is actually -t!!!
02:58
<lupine_85>
so somewhere, I suspect, there's a broken if [-f ... ]
03:05* lupine_85 hunts, fails to find, looks more
03:07
<alkisg>
lupine_85: do you have nfs or nbd?
03:07Sarten-X has quit IRC
03:07
<lupine_85>
nbd
03:08
this is the Karmic interpretation of ltsp, for reference
03:09
so something, somewhere, is setting LDM_SERVER to be '-f'
03:09
<alkisg>
Or even SERVER
03:10
<lupine_85>
yeah, I don't have a /usr/share/ltsp/get_hosts
03:13
that gets set in ltsp_config...
03:14
server=$(pgrep -f -l $DEVICE | awk '{print $3}')
03:15
oof. that script scares me
03:16
however, it gives the right output at the moment
03:16
so I'd guess a race condition
03:16
<alkisg>
Put a break=init in pxelinux.cfg/default, and cat /tmp/net-eth0.cfg to see if it looks ok
03:19* lupine_85 sets up a test machine
03:24
<lupine_85>
thanks for following along :)
03:38ftherese has quit IRC
03:48
<lupine_85>
ok, everything looks fine in the initramfs
03:49* lupine_85 reboots some more
03:54Egyptian[Home] has quit IRC
03:55
<lupine_85>
aaaand now it doesn't want to die on me, unless removing the pretty splash screen works around the race
03:57Egyptian[Home] has joined #ltsp
03:59
<lupine_85>
ah well. I shall report a bug with ubuntu and chase more if it appears again
04:02nubae has quit IRC
04:12
<lupine_85>
hmm, if I set SERVER manually, I suspect I'll bypass this bug entirely
04:34Selveste1 has quit IRC
04:35
<alkisg>
Karmic switched to upstart, it might be a race condition, it won't be the first that emerged because of the switch...
04:35seppy has joined #ltsp
04:56shawnp0wers has joined #ltsp
05:00Selveste1 has joined #ltsp
05:02
<Appiah>
https://bugs.launchpad.net/ltsp/+bug/490283 can anyone set this to wishlist? I cant find it?
05:09Selveste1_ has joined #ltsp
05:13nubae has joined #ltsp
05:18Selveste1_ has quit IRC
05:22Selveste1 has quit IRC
05:34
<lupine_85>
alkisg: I filed a bug, anyways
05:38
mind you, I don't generally get much joy with LP
05:46nubae has quit IRC
05:46
<alkisg>
Heh no LP is ok... lots of people helping there
05:48
lupine_85: I don't think you should have put LTSP in the affected projects, though, e.g. if it is an upstart problem, I don't think the Fedora or OpenSuSE maintainers will be happy to be bothered about Ubuntu-specific problems... :)
05:48
<lupine_85>
ah... whoops
05:48
maybe I don't get much joy because I don't know how to use it ;)
05:53Faithful has joined #ltsp
05:53
<alkisg>
lupine_85: If you continue to experience the issue, do replace "nc" with a script somewhat like that:
05:53
#!/bin/sh
05:53
if ! /bin/nc.original "$@"; then
05:53
echo "nc problem, spawning a shell to look into it..."
05:53
/bin/sh
05:53
fi
05:54
There, look for /etc/mount problems or whatever else you suspect may be the problem
05:54
(cat /proc/<pid>/environ will get you access to the SERVER variable...)
05:55
<lupine_85>
well, I've set SERVER= in the lts.conf now, since my main concern is having a stable LTSP environment. If I get a slow day, or that doesn't fix it, I shall poke more
05:55
I'm pretty sure that'll fix the issue though
05:55
s/fix/workaround/
06:09scottmaccal has joined #ltsp
06:13GGD has joined #ltsp
06:14nubae has joined #ltsp
06:14vlt_asleep is now known as vlt
06:15vicd has quit IRC
06:48pmatulis has joined #ltsp
07:00GodFather has joined #ltsp
07:04mikkel has quit IRC
07:09Faithful has quit IRC
07:10mikkel has joined #ltsp
07:10GodFather has quit IRC
07:11GodFather has joined #ltsp
07:17GGD_ has joined #ltsp
07:34
<shawnp0wers>
Odd... This morning one of my LTSP servers suddenly wouldn't let anyone connect to it. SSH refused connections.
07:34
<Appiah>
O_o
07:35
<shawnp0wers>
A reboot fixed it
07:35
but still, that's odd
07:35GGD has quit IRC
07:37GodFather has quit IRC
07:39CAN-o-SPAM has joined #ltsp
07:41hersonls has joined #ltsp
07:46Selveste1 has joined #ltsp
07:51mikkel has quit IRC
08:03Gadi has joined #ltsp
08:04ogra has quit IRC
08:04seppy has quit IRC
08:04ogra has joined #ltsp
08:10Kicer86 has joined #ltsp
08:12ogra_ has joined #ltsp
08:13ogra has quit IRC
08:14ogra_ is now known as ogra
08:22ogra has quit IRC
08:22ogra has joined #ltsp
08:33
<sbalneav>
Morning all
08:35
<CAN-o-SPAM>
morning sbalneav!
08:36FGXR6 has quit IRC
08:37scottmaccal has quit IRC
08:41F-GT has joined #ltsp
08:41irule has quit IRC
08:42scottmaccal has joined #ltsp
08:46
<scottmaccal>
morning
08:47Appiah_ has joined #ltsp
08:48Appiah has quit IRC
09:05
<_UsUrPeR_>
morning all. I hope everybody had a good Thanksgiving
09:08GodFather has joined #ltsp
09:14dro has joined #ltsp
09:28
<sbalneav>
s/everybody/my US friends/ :)
09:29
<Appiah_>
:D
09:29Appiah_ is now known as Appiah
09:34
<CAN-o-SPAM>
how about that boxing day?
09:34cliebow_ has joined #ltsp
09:37squarepeg has quit IRC
09:47The_Code has joined #ltsp
09:50Hyperbyte has quit IRC
09:52
<dro>
have any of experienced random clients just powering off?
09:56GodFather__ has joined #ltsp
10:04
<sbalneav>
dro: sure
10:04
Usually due to hardware problems.
10:05Brian_H has joined #ltsp
10:07dro_ has joined #ltsp
10:07dro has quit IRC
10:08GodFather has quit IRC
10:08GodFather__ is now known as GodFather
10:09
<scottmaccal>
cliebow: was up your way yesterday
10:11
<MaRX-Mode>
johnny: it did complie the ltsp client for gentoo now properly (i think). It really did help after adding gtk+ and xorg-serverinto installaton, also put some others, but at least the client managed to compile itself successfully
10:11Ahmuck has joined #ltsp
10:11
<MaRX-Mode>
.. but now some prob with nfs. feels like can
10:11Selveste1 has quit IRC
10:12
<MaRX-Mode>
*can't load NIC modules or smthg else not working in kernel to get networking on. i'll continue debuggin. anycase: Thanks =)
10:12
<johnny>
MaRX-Mode, you have to manually edit the initramfs to include your nic
10:13
and rebuild it
10:13
<MaRX-Mode>
nice, something i ever done before. any "quick -howtos" in sites ?
10:13
<johnny>
genkernel initramfs includes a bunch of ata/scis drivers
10:13
but only 2 net drivers
10:14
chroot into the chroot via linux32 chroot
10:14
if you're on amd64 server
10:14
<MaRX-Mode>
i686
10:14
dual xeon ^^
10:14
<johnny>
and edit /usr/share/genkernel/x86/modules_load
10:14
add the relevant nic modules to the nic line
10:15
and then rebuild it via genkernel initrd or genkernel ramdisk, and copy our initramfs to your tftp directory
10:15
<MaRX-Mode>
any guesses how long it takes to load.. ~20 nic modules ?
10:15
<johnny>
no
10:15
<MaRX-Mode>
k, i'll see it then
10:15
<johnny>
the design of the initramfs is genkernel is pretty bad imo
10:15
we are probably going to switch to dracut for ltsp.. some day
10:16
<knipwim>
it's on the todo list :)
10:16
<MaRX-Mode>
heh, that's one step closer =)
10:17
<knipwim>
at least it is in portage now
10:17
<johnny>
ah knipwim good
10:18
knipwim, somebody needs to get the genkernel peeps to drop their own initramfs
10:18
:(
10:18
<knipwim>
and use dracut for initramfs generation?
10:19Kicer86 has quit IRC
10:19
<johnny>
yes
10:19
for gentoo overall
10:20Kicer86 has joined #ltsp
10:22
<knipwim>
btw, did you have a chance to look at the versioned ebuild?
10:22
<johnny>
sure
10:22
don't you have a git clone
10:22
why wouldn't you commit it there?
10:23
<knipwim>
didn't commit it yet because it's new
10:23
<johnny>
you should still commit it to your own git repo
10:25vvinet has joined #ltsp
10:26
<knipwim>
http://git.flowledge.nl/?p=ltsp-overlay.git;a=summary
10:28shawnp0wers has quit IRC
10:45cliebow_ has quit IRC
10:50shawnp0wers has joined #ltsp
10:59cliebow_ has joined #ltsp
10:59vagrantc has joined #ltsp
11:03* vagrantc still doesn't understand all the -fp business.
11:03
<vagrantc>
reboot/poweroff -fp ... what's the meaning of this? :)
11:03
seems to be a massive campaign
11:10
<sbalneav>
I think, the issue was one of, we used to use -fp, but that just does the equivalent of sync;sync;halt
11:10
but it left some stuff lying around on the server
11:10
so it was shanged to -p, which is like a shutdown
11:11
but as soon as networking turns off, we lose the / disk, either via NFS or NBD
11:11
but there's still shutdown tasks to do, so the box just sits there, grinding on the fact it has no /
11:11
<ogra>
vagrantc, -f is the important part ... means do what you were called as *immediately* ... i.e. poweroff/reboot -f doesnt call the initscripts
11:12
<sbalneav>
right
11:12
what he said, in less words than what I was saying :)
11:12
Morning ogra!
11:12
<ogra>
heh
11:12
evening :)
11:12
<sbalneav>
and vagrantc as well
11:12
<vagrantc>
nfs properly shuts down on debian
11:12
at least on lenny
11:13mikkel has joined #ltsp
11:13
<ogra>
well, once ubuntu ltsp switched completely to upstart i doubt it will be needed anymore
11:14
it should just DTRT then
11:14
<vagrantc>
upstart or upstart hooks need to not disable networking if the root filesystem is network based
11:14
<ogra>
right
11:15
there are still bugs in mountall ... but they should be shaken out for lucid
11:15
<vagrantc>
i seem to recall -fp behaving badly in the past, and was happy when we dropped it
11:15
<ogra>
thats why not everything is upstartified yet
11:15dro__ has joined #ltsp
11:16dro_ has quit IRC
11:18staffencasa has joined #ltsp
11:24dro__ has quit IRC
11:25Ahmuck has quit IRC
11:31otavio has quit IRC
11:32otavio has joined #ltsp
11:32
<shawnp0wers>
of course it would happen today... Anyone experience sshd quitting seemingly for no reason on their server, causing people to not be able to log in?
11:33
I have a big presentation to the community today, using this server
11:33
(OF COURSE)
11:33
<sbalneav>
Any errors in either syslog, daemon.log, or authlog?
11:34
All your updates up-to-date? DOS/hack attack etc.
11:34
<shawnp0wers>
none that I stick out...
11:34
Yes, all up to date, and it's an internal server, used by very few people
11:34
(it's my test server)
11:34
<sbalneav>
did the sshd daemon exit altogether?
11:34
<shawnp0wers>
running /etc/init.d/ssh restart makes things work again
11:35
I'll have to wait actually to see
11:35
I restarted in a panic
11:35
but this is my second time today...
11:35
I thought it was a fluke when I came in this morning after break
11:36
it looks like there are some sysv updates I could install...
11:36
odd that it would just "happen" though
11:36
<sbalneav>
sshd does most of it's loggin in auth.log
11:36
What was in there just before it died?
11:36
<shawnp0wers>
I'll go through there some more and look for anything fishy
11:36
I'm not sure when it died, I just came back from lunch and tried to log in
11:37
There are some errors about smbpass.so being missing, etc -- but I suspect it's just a slight misconfiguration, I dont' authenticate over smb anyway
11:37
(ie, I am fairly certain it's always been there)
11:37
I do authenticate over LDAP to an OSX server
11:38akuepker has quit IRC
11:39
<shawnp0wers>
sbalneav: whoa, quit again...
11:39
last things in log are: :received signal 15; terminating.
11:39
subsystem request for sftp
11:39
init pgpme version....
11:40
but no failure messages
11:40
<sbalneav>
where's the sftp coming from?
11:40
<shawnp0wers>
(from me, I'm trying to copy a file over)
11:41
it looks like sshd is still running...
11:41
sshd: root@pts/1
11:42
(when I ps waxx | grep sshd)
11:42
<sbalneav>
but you can't ssh into the box anymore?
11:42
<shawnp0wers>
no, I get "connection refused"
11:43
<sbalneav>
Bugs in the past have dealt with sshd crashing with pam_ldap:
11:43
http://www.mail-archive.com/debian-ssh@lists.debian.org/msg01161.html
11:43
<shawnp0wers>
strange that it would happen today though, right?
11:43
I've been using the server for a couple weeks...
11:44Hyperbyte has joined #ltsp
11:44
<sbalneav>
Well, the standard systems administrator rubrick you'd use in this case is, "what's changed recently?"
11:44
<shawnp0wers>
yeah, that's just it -- I came here after my sysadmin brain started smoking. :)
11:44
this hasnt been touched in over a week
11:45
do you know what sshd prvilege separation actually is?
11:45
<sbalneav>
Vaguely. I beleive it has to do with ssh dropping privs as soon as it can.
11:45
<shawnp0wers>
hm, and that bug is for logging in with keys, not passwords
11:46
I'm really befuddled...
11:46
<sbalneav>
Anything changed on your ldap server? New users, new OU's? etc.
11:46
could be bad data coming out of the ldap server.
11:46
<Gadi>
shawnp0wers: you can increase the log level of sshd - might tell you something
11:47
<sbalneav>
Gadi: yep.
11:47
<shawnp0wers>
will do
11:47
<scottmaccal>
ssh -v is helpful sometimes as well
11:47* Gadi says hi, btw
11:47
<shawnp0wers>
I I restarted the process again, that "sshd: root@pts/1" remained active even when the services was stopped
11:47
I killed if off, and restarted sshd
11:47
I'll go increase the log level
11:48
<Gadi>
yeah, listen to scottmaccal, too - you may even want ssh -vvv
11:49
<shawnp0wers>
wow, logging in once kills it
11:49
(via thin client)
11:50
I'll recreate this a bit and pick your collective brains a bit...
11:50* shawnp0wers goes off to gather info
11:54
<shawnp0wers>
well, I figured out what causes the crash, haven't figured out why yet...
11:54
Gadi, sbalneav: If I log into a thin client as root, I can do so -- but it crashes sshd
11:54
(I was logging in as root to globally install an openoffice template pack)
11:55
as long as I don't log into the thin client as root, all appears fine
11:55
<Gadi>
"doctor, it hurts when I do this." "Don't do that."
11:56
<shawnp0wers>
LOL, well yes -- I've stopped doing it
11:56
still, thought I'd report back. :)
11:57
it makes me laugh a bit WHY I was logging in as root. See, windows was acting odd, so I couldn't load the VSphere manager for ESX...
11:57
normally I'd do that on the console
11:57
(which, I just did, and all seems fine)
12:13* stgraber is working on converting LTSP's boot scripts to upstart
12:14
<Gadi>
stgraber: good - thats why I pulled everything into ltsp-init-common
12:14
:)
12:14* vagrantc just figured that out
12:14
<stgraber>
I'm actually splitting that from two init scripts to probably 20 or so ;)
12:15
so we can play with the event-based stuff
12:15
<Gadi>
stgraber: don't delete the initscripts from the branch, tho - other distros will need them
12:15* vagrantc waves
12:15
<stgraber>
ideally, I'd have cdpinger, jetpipe, sane, ... to start only when the device is actually plugged in
12:15cliebow_ has quit IRC
12:15
<stgraber>
vagrantc: Is it still in the plans for the next Debian release to move to upstart ?
12:16cliebow_ has joined #ltsp
12:16
<vagrantc>
i moved cdpinger to start from udev
12:16
or does upstart take over udev?
12:16
stgraber: looks like it's finally out of experimental
12:17
<stgraber>
udev can send some events to upstart, then we can start stuff from there
12:17
<vagrantc>
stgraber: i outsource all my future technology awareness to johnny
12:17
<stgraber>
or keep them in udev
12:18
<vagrantc>
there's been significant work on new non-upstart stuff, though last i heard, upstart wasn't totally out of the question
12:18
<johnny>
that sucks :(
12:18
would be nice if debian would just go wit upstart..
12:18
<vagrantc>
stgraber: pere would be the one to ask
12:18
<johnny>
then we'd have 4 major distros using it
12:18
<vagrantc>
johnny: ubuntu, fedora and .. ?
12:19
<johnny>
mandriva
12:19
<vagrantc>
that's a major distro?
12:19
<johnny>
oh wait.. maybe that's plymouth they are switching to only.. ..
12:19
well.. they were one.. :)
12:21
<vagrantc>
well, if it's any consolation, the same "upstream" version of upstart is in debian squeeze/testing as was in ubuntu karmic
12:21
but i don't think it's the default
12:21
not yet ported to GNU/kFreeBSD, though
12:35
<Gadi>
stgraber: you may get a big performance hit if you break it into 20 upstart scripts, each of which sources lts.conf
12:35
it may be better to group them into tasks with common dependencies
12:36
<johnny>
wouldn't it be better to reuse/subclass more stuff instead?
12:38
<stgraber>
Gadi: I'm basically splitting it so things that can be triggered by an event will be and some others will only start once ldm is started
12:38
<Gadi>
but, you will keep all functions in ltsp-init-common, yeah?
12:39alexqwesa_ is now known as alexqwesa
12:39
<stgraber>
Gadi: I'll try to only parse lts.conf (and the ltsp-cluster equivalent) once and then send event based on that. Other scripts will just wait for that event + some other condition to be triggered
12:39
for most of them, yes, that'll be using the functions from ltsp-init-common
12:39
some may be done a bit differently
12:40
anyway, I don't plan on breaking upstream soon, that's my goal for lucid, so I have a bit of time here ;)
12:44
<CAN-o-SPAM>
hehe
12:44
good goals :)
12:45
<stgraber>
Ubuntu's goal is to get a netbook booting in 10s, I think that I can probably get half of that for LTSP on the same hardware
12:46
(I currently have a X server started in 5s on an atom)
12:51
<CAN-o-SPAM>
wow very nice stgraber
12:52
<Gadi>
of course,his Xserver is a pre-recorded animation pasted to the framebuffer... ;)
12:56
<CAN-o-SPAM>
whatever works ?
12:57
heh
12:57
<stgraber>
Gadi: my X server at the moment is just a white root window ;) so I probably could do that with the fb driver instead to cheat a bit
12:58mushroomblue has quit IRC
13:00mushroomblue has joined #ltsp
13:15Sarten-X has joined #ltsp
13:25
<MaRX-Mode>
hmm.. still prob with nfs mount. does this client run dhcp wen built from scratch or should it also be turned on ?
13:27
<johnny>
uhmm.. sure
13:27
you need a dhcp server somewhere
13:27
and tell it to use nfs stuff
13:27
<MaRX-Mode>
do have it
13:27
<johnny>
does it tell servesr the proper boot info and the nfs path?
13:27
<MaRX-Mode>
pasting here allowed ?
13:27
<johnny>
no
13:27
pastebin
13:27
i don't really have much time tho
13:27
i'm pretty busy
13:28
<MaRX-Mode>
k, you continue though and we check it later on =)
13:28
i think dhcp server works well, client is still the prob
13:28
already able to boot it up with pxe
13:31The_Code has quit IRC
13:48GodFather has quit IRC
13:56Sarten-X has quit IRC
13:59vmlintu has quit IRC
14:15
<vagrantc>
well, installing upstart didn't break the ltsp client on lenny...
14:16
er, squeeze
14:30
oh, also, using nbd-client's initramfs hooks, networking doesn't break nbd either.
14:34Egyptian[Home] has quit IRC
14:40
<johnny>
alkisg, http://mso-chronicles.blogspot.com/2009/11/youtube-and-html5.html
14:44
<alkisg>
johnny: yup, I've seen it, on my Ubuntu it works with chromium but not with firefox
14:44
<johnny>
that's exactly what it says
14:44
it's a codec thing it seems
14:45
<alkisg>
Yeah... and unfortunately not all videos are accessible through that URL
14:45
(http://www.youtube.com/html5)
14:45
<johnny>
not yet..
14:46
<alkisg>
I don't know... I'm not too optimistic about this
14:46
Flash gives greater control, e.g. you can overlay an ad over a video, I don't know if html5 supports this
14:46
<johnny>
pretty sure you can
14:46
<alkisg>
So I'm not sure that youtube will switch to html5 videos any time in the future at all...
14:46
<johnny>
i know gstreamer itself supports it
14:47
i've done it with fluendo streaming media server
14:47
also .. btw.. i found a fast running canvas game
14:47
so you were wrong there too :)
14:47
<alkisg>
Heh
14:47
Which one?
14:47
<johnny>
i was at the store when looking ati t
14:47
i'll have to refind it
14:48
perhaps when my headache and sore throat go away
14:48
thinking about laying down for a bit
14:49
<alkisg>
Sure np... I still don't think canvas is any match for flash, for games. I hope I'm wrong, but... anyway we'll see :)
14:54Ahmuck has joined #ltsp
14:56Cristina1 has joined #ltsp
15:00
<Cristina1>
hi there!
15:00
is anyone able to open http://wiki.ltsp.org/twiki/bin/view/Ltsp/LtspDocumentationUpstream?
15:03
<Brian_H>
nope
15:03
4. Error type: 404
15:04
http://sourceforge.net/apps/mediawiki/ltsp/index.php?title=Ltsp_LtspDocumentationUpstream
15:04
can get to that though :)
15:05
<sbalneav>
Cristina1
15:05
!docs
15:05
<ltspbot>
sbalneav: "docs" :: For the most current documentation, see https://sourceforge.net/apps/mediawiki/ltsp/index.php?title=Ltsp_LtspDocumentationUpstream
15:05
<sbalneav>
it's moved.
15:07hersonls has quit IRC
15:07
<Cristina1>
thanks! :D
15:10
turns out that I'm trying to solve an NFS problem, and doing some research on google, I found a link to:
15:10
http://wiki.ltsp.org/twiki/bin/view/Ltsp/NFS#NFS_Server_not_responding
15:10
which might be very helpful for what I'm trying to do, but I can not get to it
15:11Kicer86 has quit IRC
15:12
<sbalneav>
It probably hasn't been moved over yet.
15:13
Our wiki crashed a few weeks ago, and we've been slowly moving pages over a few at a time.
15:13
Look at the cached copy in google :)
15:21
<Cristina1>
thanks, didn't have any luck though :(
15:23Jachim has joined #ltsp
15:23Jachim has quit IRC
15:25kusznir__ is now known as kusznir_
15:26Egyptian[Home] has joined #ltsp
15:29shawnp0wers has quit IRC
15:30Egyptian[Home] has quit IRC
15:37Egyptian[Home] has joined #ltsp
15:37
<Cristina1>
sbalneav: how long do u think it will take until all the pages are moved?
15:40vagrantc has quit IRC
15:43
<sbalneav>
Cristina1: no idea.
15:43
"as time permits" :)
15:44
I know jammcq does a few pages per day.
15:44
Rather than wait for the page, why not just describe your problem, and we'll see if we can help you with it?
15:46
<Cristina1>
thanks! So this is my problem... I'm working with git kernel 2.6.29 on leopard board. I'm using NFS as file system
15:47
but every once in a while I lose connectivity, getting: nfs: server 10.251.101.7 not responding, still trying
15:47Jachim has joined #ltsp
15:48nubae has quit IRC
15:49
<Cristina1>
so I've been dealing with this problem for several weeks now
15:51shawnp0wers has joined #ltsp
15:51
<Cristina1>
haven't been able to find how to solve it
15:56
<sbalneav>
Is the server, in fact, still responding? I.e. can other boxes mount it?
15:57
<Cristina1>
yes it is,
15:57Ahmuck-Jr has joined #ltsp
15:57
<Cristina1>
it just stops working after a while
15:58
which is pretty much random, since it does not follow any patterns
15:59
<sbalneav>
So it's the client that's losing it's connection, not the server?
16:00
What kinds of switch are the client and server connected with? Is it a smart switch? We've seen smart switches in the past decide to stop passing UDP traffic.
16:00
You may be able to solve it by doing TCP NFS mounts.
16:02
<Cristina1>
no it is not a smart switch
16:02
I've tried the same using another version of the kernel> 2.6.18
16:02
and I do not have this NFS issues with it
16:05Jachim has quit IRC
16:05
<sbalneav>
ok, so then clearly, the problem exists with the 2.6.29 kernel.
16:06
<Cristina1>
yes, so I believe it must be something with the nfs driver
16:06
<sbalneav>
If you're using a git version of the kernel, and compiling it yourself, you're well into "You're on your own" territory. :)
16:07
Solution would be to use either a distro supported kernel, or file bugs directly with LKML
16:08
<Cristina1>
yeah... git version for me!! thing is I have got to use it since there's a camera driver for this kernel I really need to use
16:08stevesmith1983 has joined #ltsp
16:08
<Cristina1>
so I am forced to fix this
16:10
<sbalneav>
Well, this clearly isn't an LTSP problem, so in that case, LKML's your best bet.
16:12mikkel has quit IRC
16:19korcan has joined #ltsp
16:22shawnp0wers has quit IRC
16:26
<Cristina1>
yeah, thought it might be helpful for me to see that page I posted at the beginning
16:30
<alkisg>
Cristina1: http://web.archive.org/web/20080507005104/http://wiki.ltsp.org/twiki/bin/view/Ltsp/NFS
16:37alkisg has quit IRC
16:40CAN-o-SPAM has quit IRC
16:41jach_ has quit IRC
16:41jach has quit IRC
16:48nubae has joined #ltsp
17:04Gadi has left #ltsp
17:06nubae has quit IRC
17:07Selveste1 has joined #ltsp
17:24Selveste1 has quit IRC
18:05Sarten-X has joined #ltsp
18:27Brian_H has quit IRC
18:37staffencasa has quit IRC
18:41stevesmith1983 has quit IRC
18:45ehrenpr has joined #ltsp
18:47
<ehrenpr>
is there anyone here who can help me configure italc and kiwi-ltsp?
18:52artista_frustrad has joined #ltsp
19:26panthera has quit IRC
19:26panthera has joined #ltsp
19:38kleewyck has joined #ltsp
19:40
<kleewyck>
I Know I'm nuts, but i'm trying to use the ppc chgroot for Fedora K12linux. I have it to the point the iBook boots starts to load, unfortunatly the HAL daemon fales to start. does anyone know a way to debug this?
19:45dkostousov has quit IRC
19:50vvinet has quit IRC
19:55ehrenpr has quit IRC
20:33pmatulis has quit IRC
21:20vicd has joined #ltsp
21:41GGD_ has quit IRC
21:41yanu has quit IRC
21:41yanu has joined #ltsp
21:57Egyptian[Home] has quit IRC
21:57
<kleewyck>
Figured it out cache/hald had the wrong protection..
21:58Egyptian[Home] has joined #ltsp
22:27yanu has quit IRC
22:30yanu has joined #ltsp
22:34mushroomblue has quit IRC
22:37mushroomblue has joined #ltsp
22:38kleewyck has quit IRC
22:47kleewyck has joined #ltsp
22:47kleewyck has quit IRC
23:12AlexJr has joined #ltsp
23:15alkisg has joined #ltsp
23:17rjune has quit IRC