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


Channel log from 9 May 2014   (all times are UTC)

00:03gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
00:09gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 240 seconds)
00:23adrianorg has left IRC (adrianorg!~adrianorg@179.179.79.129, Ping timeout: 245 seconds)
00:25adrianorg has joined IRC (adrianorg!~adrianorg@177.134.62.14)
00:30calisto has joined IRC (calisto!~calisto@unaffiliated/calisto)
00:53andygraybeal has left IRC (andygraybeal!~andy@h72.214.22.98.dynamic.ip.windstream.net, Ping timeout: 250 seconds)
00:59calisto has left IRC (calisto!~calisto@unaffiliated/calisto, Ping timeout: 265 seconds)
01:00calisto has joined IRC (calisto!~calisto@unaffiliated/calisto)
01:05andygraybeal has joined IRC (andygraybeal!~andy@h224.192.213.151.dynamic.ip.windstream.net)
01:05gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
01:07calisto has left IRC (calisto!~calisto@unaffiliated/calisto, Ping timeout: 258 seconds)
01:08gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Read error: No route to host)
01:09gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
01:13gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 240 seconds)
01:35imox has left IRC (imox!~imox@p57A96336.dip0.t-ipconnect.de, Quit: imox)
01:47adrianorg has left IRC (adrianorg!~adrianorg@177.134.62.14, Ping timeout: 245 seconds)
01:50adrianorg has joined IRC (adrianorg!~adrianorg@177.134.62.14)
02:09gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
02:14gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 245 seconds)
02:46Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 252 seconds)
02:48Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
03:10gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
03:15gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 276 seconds)
03:55Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 276 seconds)
04:11gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
04:11Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
04:15gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 264 seconds)
04:46Ark74 has joined IRC (Ark74!~Ark74@189.214.42.128.cable.dyn.cableonline.com.mx)
05:01lmds_ has left IRC (lmds_!~lmds@tui.pi-et-ro.net, Ping timeout: 276 seconds)
05:02lmds_ has joined IRC (lmds_!~lmds@tui.pi-et-ro.net)
05:12gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
05:16gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 265 seconds)
05:27Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Ping timeout: 264 seconds)
05:52work_alkisg has left IRC (work_alkisg!~alkisg@ubuntu/member/alkisg, Read error: Connection reset by peer)
05:59work_alkisg has joined IRC (work_alkisg!~alkisg@plinet.ioa.sch.gr)
06:07
<work_alkisg>
So, ops in this channel are: Mistik1, jammcq, jstraw, sbalneav and knipwim
06:07work_alkisg is now known as alkisg
06:07
<alkisg>
sbalneav: can you make the bot op?
06:08
Otherwise we're not able to e.g. change the /topic, and e.g. remove the stackexchange proposal...
06:13gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
06:17gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Ping timeout: 252 seconds)
06:18alexxtasi has joined IRC (alexxtasi!~alex@unaffiliated/alexxtasi)
07:01piet has joined IRC (piet!~piet@dyndsl-085-016-207-157.ewe-ip-backbone.de)
07:11gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)
07:24vmlintu has joined IRC (vmlintu!~vmlintu@83.145.240.143)
07:47khildin has joined IRC (khildin!~khildin@ip-213-49-86-81.dsl.scarlet.be)
07:58ogra_ has left IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de, Ping timeout: 240 seconds)
08:02bennabiy has left IRC (bennabiy!~Thunderbi@96-37-209-0.dhcp.leds.al.charter.com, Ping timeout: 255 seconds)
08:03bennabiy has joined IRC (bennabiy!~Thunderbi@96-37-209-0.dhcp.leds.al.charter.com)
08:08ogra_ has joined IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de)
08:20vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Ping timeout: 246 seconds)
09:43Grembler has joined IRC (Grembler!~Ben@cpc29-aztw22-2-0-cust128.18-1.cable.virginm.net)
10:04Grembler has left IRC (Grembler!~Ben@cpc29-aztw22-2-0-cust128.18-1.cable.virginm.net, Quit: I Leave)
10:17lelele has joined IRC (lelele!502128c2@gateway/web/freenode/ip.80.33.40.194)
10:17
<lelele>
hi everybody
10:18
aniwhere know the place where the LTSP client save the log? i'm using fat clients
10:18
<alkisg>
In /var/log, like normal
10:20lelele has left IRC (lelele!502128c2@gateway/web/freenode/ip.80.33.40.194, Client Quit)
10:26piet has left IRC (piet!~piet@dyndsl-085-016-207-157.ewe-ip-backbone.de, Quit: Verlassend)
10:26alkisg is now known as work_alkisg
10:36workingcats has left IRC (workingcats!~workingca@212.122.48.77, Quit: Leaving)
10:43adrianorg has left IRC (adrianorg!~adrianorg@177.134.62.14, Read error: Connection reset by peer)
10:47adrianorg has joined IRC (adrianorg!~adrianorg@177.134.58.129)
10:55workingcats has joined IRC (workingcats!~workingca@212.122.48.77)
10:56adrianorg has left IRC (adrianorg!~adrianorg@177.134.58.129, Ping timeout: 255 seconds)
10:58adrianorg has joined IRC (adrianorg!~adrianorg@177.134.58.129)
11:28
<khildin>
what would be the best way to secure iTalc... I am in a situation that it sometimes occurs that teachers forget to log out their session or forget to lock their screen....
11:28
and it already happened some brat was able to get the private key... :-/
12:05telex has left IRC (telex!~telex@freeshell.de, Remote host closed the connection)
12:06telex has joined IRC (telex!~telex@freeshell.de)
12:54adrianorg has left IRC (adrianorg!~adrianorg@177.134.58.129, Ping timeout: 240 seconds)
12:55adrianorg has joined IRC (adrianorg!~adrianorg@187.113.248.212)
13:16calisto has joined IRC (calisto!~calisto@unaffiliated/calisto)
13:27Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas)
13:35brianr has joined IRC (brianr!~brianr@out.ewbc.com)
13:58alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
14:08
<brianr>
good morning!
14:09
yesterday I was asking about an nbd issue i was having
14:10
basically i run into an issue where my thin clients, while booting, will be dropped to a busybox
14:10
they get an error about mounting /dev/nb0 on /root failed invalid argument
14:11
/scripts/init-bottom/ltsp /usr/bin/test
14:11
/scripts/init-bottom/ltsp /usr/bin/test line 27 panic not founf
14:12
<alkisg>
Write this in the client busybox, and tell us its output: grep ROOTSERVER /*/net*.conf
14:13
<brianr>
ok, so i cannot like make one fail cause its a production environment
14:14
<alkisg>
You don't have a test client that fails?
14:15
<brianr>
no, because that would mean that a production client is probably not working
14:15
<alkisg>
If you cannot provide real-time feedback, irc is not suitable, try the ltsp mailing list
14:15
<mmetzger>
I have the same problem (Ubuntu 14.04)
14:15
Get:
14:15
<brianr>
to correct the issue, as root i have to run /bin/nbd-server and then some more can login
14:15
<mmetzger>
grep: /tmp/net0-eth0.conf: No such file or directory
14:16
<alkisg>
brianr: that doesn't help in troubleshooting, sorry
14:16
<brianr>
or more precisely more can boot
14:16
<alkisg>
Try again in IRC when you have a client in busybox, or try the mailing list where you can give feedback when the problem happens again.
14:17
In IRC one can't wait 5-6 hours until the problem happens again, to get the next line of feedback...
14:18alexxtasi has left IRC (alexxtasi!~alex@unaffiliated/alexxtasi)
14:18
<brianr>
so, that problem is pretty unique? your not seeing anyone else with a similar problem?
14:18
<alkisg>
There are 3-4 similar problems, and all require troubleshooting
14:18
<mmetzger>
I am running into the same error.
14:19
<alkisg>
And I'm sure there are others that we haven't yet seen, like the one with missing /run at init-premount, that mmetzger mentions
14:19
<mmetzger>
(or not)
14:19
<alkisg>
mmetzger: /tmp/net-eth0.conf is a symlink to /run/net-eth0.conf, and that second file isn't there?
14:19
ls -l /*/net*.conf
14:20
<mmetzger>
alkisg: Nothing in /run
14:20
<alkisg>
And what is /tmp/net-eth0.conf?
14:20
A symlink or a file?
14:20
<mmetzger>
Symlink as you mentioned
14:21
<alkisg>
Yeah it's a new issue, first heard of it 3 days ago, was I talking with you again or with someone else?
14:21
<mmetzger>
I think you were speaking with Chap
14:21
<alkisg>
Right, same issue there
14:21
Is it always reproducible? Or it only happens some times?
14:22
<mmetzger>
Happens for me everytime. 64bit server, 64bit client.
14:22
<brianr>
just some additional observations ... after rebooting the LTS box, i went around and restarted the thin clients. 23 booted, then the 2th failed and i had to start another nbd-server process, then about 6 more booted, i started another nbd-server process and the remaining 6 were able to boot
14:22
*24th
14:22
<alkisg>
brianr: on the server, check /var/log/syslog for messages about nbd-server segfaulting
14:22
<brianr>
nada
14:23
<alkisg>
If so, update your nbd-server, there was a known problem about that
14:24
mmetzger: put break=premount in pxelinux.cfg/default, replacing 'quiet splash' there
14:24
When you get a busybox shell, check the messages that you see *after* the prompt, if any, and then, run: /scripts/init-premount/udhcp
14:25
And then ls -l /*/net*.conf
14:25
<mmetzger>
alkisg: Will do - will take a few moments. And thank you for the assistance.
14:25
<vmlintu>
brianr: if the problem is that with a trusty client first boot works and second not, it's a kernel bug
14:27
<brianr>
its 13.10, all up to date
14:27
<vmlintu>
brianr: ok, then it's another bug
14:28
<brianr>
screenshot of error: https://drive.google.com/file/d/0B2mh9UJ2G6hHVnAtSkpWNy1VLVU/edit?pli=1
14:29
<vmlintu>
brianr: which kernel are you using?
14:30
<brianr>
i wasn't really expecting live troubleshooting ... but i can't really find any good leads on the issue to troubleshoot itt
14:30
inux version 3.11.0-20-generic
14:31
<mmetzger>
alkisg: Only message post prompt is "random: nonblocking pool is initialized"
14:31
alkisg: Ran udhcp script, got IP
14:31
alkisg: Now have existing /run/net-eth0.conf file
14:32
<alkisg>
mmetzger: now type exit
14:32
See if it boots that way
14:32
<vmlintu>
brianr: we did have similar problems with nbd-client on 13.10 earlier and those were solved by updating to newer nbd-client. If you want to try a newer nbd-client, the versio we use is here: http://archive.opinsys.fi/git-master/pool/quantal/main/n/nbd/
14:33
<alkisg>
brianr: what's your ltsp server ip?
14:34
<brianr>
10.28
14:34
<mmetzger>
alkisg: Unfortunately no - got the same error of mount: mounting /dev/nbd0 on /root failed: Invalid argument.
14:35
<alkisg>
And you're in busybox now, right? What's the output of this? grep ROOTSERVER /*/net*.conf
14:35adrianorg has left IRC (adrianorg!~adrianorg@187.113.248.212, Ping timeout: 264 seconds)
14:36
<brianr>
no, im not in busy box .. i snapped a pic of one that had the error last night
14:36
<alkisg>
brianr: when you're in a position to troubleshoot, run the same command, and see if ROOTSERVER is 10.28
14:36
<mmetzger>
alkisg: Back to the "No such file or directory" and /run is empty.
14:36adrianorg has joined IRC (adrianorg!~adrianorg@179.179.76.222)
14:36
<brianr>
i will run that next time one fails
14:36
<alkisg>
brianr: the previous line was for mmetzger
14:36
<brianr>
ok
14:37
<alkisg>
mmetzger: and I suppose the nbd-client mounting line succeeds, right?
14:37
<mmetzger>
alkisg: Symlink in /tmp for net-eth0.conf to /run
14:37
<vmlintu>
alkisg: btw, there's now a kernel bug in kernels =>3.13 that causes nbd-client to fail on second boot.. The first boot works, but second boot on the same client always fails unless you restart nbd-server in between or wait some 15 minutes..
14:37
<alkisg>
!nbd-client
14:37
<ltsp>
nbd-client: To try mounting the NBD image from the client initramfs: nbd-client 192.168.67.1 -N /opt/ltsp/i386 /dev/nbd0
14:37
<alkisg>
vmlintu: client-side kernel bug that persists reboots?
14:37
<vmlintu>
alkisg: yep
14:37
<alkisg>
Wow... is there a bug report or explanation about it?
14:38
<brianr>
so upgrading to 14.04 isn't gonna fix my issue ;)
14:38
<vmlintu>
alkisg: we are working on it.. we've spent two days on it already
14:38
<mmetzger>
alkisg: nbd-client seems to work, though it does write a message "nbd0: unknown partition table"
14:38
<alkisg>
vmlintu: and restarting nbd-server on the server doesn't fix it? it doesn't sound right...
14:38
mmetzger: that's normal
14:38
<vmlintu>
alkisg: nbd-server restart fixes it
14:39
<alkisg>
So, your issue is "/run gets cleaned after ipconfig runs, for some reason"
14:39
vmlintu: it sounds like a server-side bug to me
14:39
wouter: have you seen that one? ^
14:40
Client boots fine the first time in kernel >=3.13, and it fails the second time
14:40
<vmlintu>
alkisg: nbd-server is not the problem, it's the client
14:40
<alkisg>
vmlintu: I mean, nbd-server might be keeping some state for that client
14:40
So it breaks on second mount because of that client-specific state that iskept on the server, not on the client
14:41
mmetzger: do the same thing with break=premount again. At that point, just create some file in run, and type exit. See if that file is cleaned too.
14:41
<mmetzger>
alkisg: Will do.
14:41
<alkisg>
mmetzger: i.e., check if everything in /run is deleted, or if it's only net-eth0.conf
14:42
mmetzger: also, cat /proc/mounts at both times
14:42
On break, and on failure
14:42
See what is mounted on /run at both times
14:42
<vmlintu>
alkisg: this is probably caused by it: http://sourceforge.net/p/ltsp/mailman/message/32303113/
14:43
alkisg: only the change of the client kernel breaks it. It happens also on 13.10 if you use kernel 3.13 or newer
14:43
<alkisg>
vmlintu: vagrantc there suggests dhcp, not nbd
14:44
And, if it's the same issue that mmetzger mentions, it could be related to kernel, udev, /run etc, but again not nbd
14:45
vmlintu: when it happens, does manually running nbd-client work?
14:45
!nbd-client
14:45
<ltsp>
nbd-client: To try mounting the NBD image from the client initramfs: nbd-client 192.168.67.1 -N /opt/ltsp/i386 /dev/nbd0
14:47
<vmlintu>
alkisg: after it has failed from initramfs, running nbd-client manually does work
14:48
<alkisg>
vmlintu: then it doesn't sound like an nbd issue at all
14:48
It sounds like mmetzger's issue
14:50
<vmlintu>
all ubuntu 14.04 installations are probably affected by this
14:50
<alkisg>
vmlintu: ls -l /*/net*.conf
14:50
<mmetzger>
alkisg: Ok, it looks like /run is no longer mounted after the manual udhcp and exit.
14:51
<alkisg>
Do you also see the broken symlink there?
14:51
<vmlintu>
I'm on a laptop now, so I cannot check anything
14:51
I was debugging this on 12.10 originally with kernel 3.13
14:51
<alkisg>
mmetzger: it's mounted at break=premount, and not mounted at the busybox panic shell?
14:51
<mmetzger>
alkisg: Correct.
14:52
alkisg: Before it's "tmpfs /run tmpfs rw,nosuid,noexec,relatime,size=710428k,mode=755 0 0"
14:52
<alkisg>
Ah
14:52
<mmetzger>
alkisg: After, there's no /run listed in the output of "cat /proc/mounts"
14:52
<alkisg>
Hmm no, it couldn't have been moved to the root file system because it doesn't exist yet...
14:53
mmetzger: now try something else: break=mount
14:53
And again the grep ROOTSERVER line
14:53
or the ls -l line
14:53
Don't run udhcp anymore manually
14:53
Just inspect things at break=mount
14:53
<mmetzger>
Got it - results in a few.
14:56||cw has left IRC (||cw!~chris@phpgroupware/cw, Read error: Connection timed out)
14:57||cw has joined IRC (||cw!~chris@phpgroupware/cw)
14:57
<mmetzger>
alkisg: /run is mounted at the break=mount with the same options as before and the net-eth0.conf file has the correct ROOTSERVER defined.
14:58
<alkisg>
mmetzger: there are 2 net-eth0.conf files, one symlink in /tmp and a real file in /run, right?
14:58
<mmetzger>
alkisg: Correct
14:59
<alkisg>
mmetzger: is that a clean install?
14:59
Or an upgrade?
15:00
<mmetzger>
alkisg: Clean install
15:00
<alkisg>
!nbd-client
15:00
<ltsp>
nbd-client: To try mounting the NBD image from the client initramfs: nbd-client 192.168.67.1 -N /opt/ltsp/i386 /dev/nbd0
15:01
<alkisg>
And you said that runs, and the only thing you changed, was i386 to amd64?
15:01
<mmetzger>
alkisg: Yes
15:01
<alkisg>
And the ip...
15:01
cat /proc/cmdline
15:01
<mmetzger>
10.10.10.2
15:01
<alkisg>
What do you see in /proc/cmdline, that mentions nbd?
15:02
Only root=/dev/nbd0? Or also ltsp_amd64 somewhere?
15:02
<mmetzger>
nbdroot=:ltsp_amd64
15:02
and root=/dev/nbd0
15:02
<alkisg>
OK, go to pxelinux.cfg/default and remove nbdroot completely
15:03
<mmetzger>
Reboot with break=mount option?
15:03
<alkisg>
Without it, normal boot
15:03
I.e. the problem could be gone with that change now...
15:05
<mmetzger>
Sweet! It booted to login screen
15:06
<alkisg>
mmetzger: put $CHROOT/etc/ltsp/update-kernels.conf to pastebin
15:06
<mmetzger>
alkisg: From the client or the server?
15:06
<alkisg>
Server
15:07
<mmetzger>
Duh, nevermind. Sorry.
15:07
<alkisg>
It doesn't matter, on the client it's /etc/ltsp/update-kernels.conf
15:07
Without the $CHROOT part...
15:08
I.e. now we want to find out what put ltsp_amd64 there... that was the old way for nbd roots
15:10* alkisg wonders how that affected /run, though...
15:11
<mmetzger>
alkisg: Looking at this again, this may have been an upgrade from 12.04, but the client image was a brand new build.
15:11
<alkisg>
OK, that does make sense, 12.04 had that part
15:12
<mmetzger>
I'll build a fresh 14.04 version and see if it's the same. Would you still like the pastebin of the current update-kernels.conf?
15:12
<brianr>
regarding the nbd-server update ... i am on 3.3-3, should i get to 3.4-5 ?
15:14
<alkisg>
brianr: don't try random things without proper troubleshooting
15:15
You'll just make things worse
15:15
mmetzger: yes, please pastebin it
15:16
<brianr>
there doesnt seem to be any way to force nbd-server to be more verbose other than running it in the foreground
15:16
which still isn't very useful
15:20
<mmetzger>
alkisg: http://pastebin.com/gFKNFLqE
15:21
<alkisg>
mmetzger: is your pxelinux.cfg/default a symlink or a real file?
15:21
<mmetzger>
alkisg: It's a real file.
15:22
<alkisg>
OK, nowadays it's a symlink to ltsp
15:22
Delete it from $CHROOT/boot/pxelinux.cfg/default completely, if it exists there,
15:22
and make it a symlink to the ltsp file in $TFTP/ltsp/amd64/pxelinux.cfg/default
15:22
That was there because of the 12.04 => 14.04 upgrade
15:23
<mmetzger>
symlink in chroot, on server, or both?
15:24
<alkisg>
delete in chroot, symlink on server
15:24* alkisg waves
15:25alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
15:25adrianorg has left IRC (adrianorg!~adrianorg@179.179.76.222, Read error: Connection reset by peer)
15:26brianr has left IRC (brianr!~brianr@out.ewbc.com, Quit: Leaving)
15:28
<mmetzger>
alkisg - For what it's worth, symlinking default to ltsp in the tftpboot results in a successful boot.
15:30adrianorg has joined IRC (adrianorg!~adrianorg@187.115.107.51)
15:39vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc)
16:26brianr has joined IRC (brianr!~brianr@out.ewbc.com)
16:28
<brianr>
good news everyone, my test thin client (a vm) failed to boot with the same error i was talking about earlier
16:30
grep ROOTSERVER /*/net*.conf yeilds No Such File or Directory
16:36
<vagrantc>
multiple DHCP servers?
16:36
<brianr>
there is a net-eth0.conf in /run though
16:36
<vagrantc>
bad DHCP settings?
16:37
multiple DHCP servers *and* bad DHCP Settings?
16:37
<brianr>
vagrantc, if that were the case then why does starting another nbd-server instance correct the problem?
16:38
<vagrantc>
i maybe don't recall the whole backstory
16:38
brianr: what do you mean by starting another nbd-server?
16:38
<brianr>
https://drive.google.com/file/d/0B2mh9UJ2G6hHVnAtSkpWNy1VLVU/edit?pli=1
16:39
so thats what i get
16:39
<vagrantc>
it actually puts a ROOTSERVER value into net-eth0.conf ?
16:39
<brianr>
yes
16:39
and its the right IP
16:39
<vagrantc>
that could happen for dozens of reasons
16:39
could you paste the results of net-eth0.conf when it's broken and when it's working?
16:40
brianr: you using ipappend 3 ?
16:41
<brianr>
i am going to check the ipappend ... cause i know i added that when i was troubleshooting a problem with blank host names from the clients
16:42
<vagrantc>
i still don't understand what you meant by "starting another nbd-server instance correct the problem?"
16:42
maybe nbd-server isn't running at all, and that's why it corrects the problem?
16:42
<brianr>
i goto the console of the LTS box and run /bin/nbd-server
16:44
for instance, last night i update the lts box, ubuntu 13.10, and reboot it. run aroun drebooting the thin clients .. 23 boot up, 24th fails , i goto server run /bin/nbd-server 6 more connect, then they start failing, run /bin/nbd-server again and the next 7-10 all connect
16:48
so, it looks like the nbd-server process being spawned as a thin client connects is being lost at some point, so when someone logs out of the thin client it goes to black screen and then they reboot it but there is already a nbd-server process running for that thin client and it shits the bed
16:49
there is no indication in syslog of any crashes
16:51
though ... i see this "failed to setup servers: failed to open a modern socket: failed to bind an address to a socket: Address already in use"
16:52
<vagrantc>
are you running nbd-server from inetd, or from init?
16:53
<brianr>
i have been assuming it wass init
16:53
this is a relatively new box
16:58
ipappend 2
16:58
<vagrantc>
try ipappend 3
17:05adrianorg has left IRC (adrianorg!~adrianorg@187.115.107.51, Ping timeout: 265 seconds)
17:10calisto has left IRC (calisto!~calisto@unaffiliated/calisto, Ping timeout: 252 seconds)
17:12adrianorg has joined IRC (adrianorg!~adrianorg@189.58.183.208.dynamic.adsl.gvt.net.br)
17:22adrianorg has left IRC (adrianorg!~adrianorg@189.58.183.208.dynamic.adsl.gvt.net.br, Ping timeout: 276 seconds)
17:24adrianorg has joined IRC (adrianorg!~adrianorg@177.204.157.184.dynamic.adsl.gvt.net.br)
17:25alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg)
17:38Phantomas1 has joined IRC (Phantomas1!~Phantomas@ubuntu/member/phantomas)
17:39Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas, Disconnected by services)
17:39Phantomas1 is now known as Phantomas
17:40adrianorg has left IRC (adrianorg!~adrianorg@177.204.157.184.dynamic.adsl.gvt.net.br, Ping timeout: 240 seconds)
17:47adrianorg has joined IRC (adrianorg!~adrianorg@179.183.66.242)
18:08adrianorg has left IRC (adrianorg!~adrianorg@179.183.66.242, Read error: Connection reset by peer)
18:10adrianorg has joined IRC (adrianorg!~adrianorg@179.180.169.251)
18:22adrianorg has left IRC (adrianorg!~adrianorg@179.180.169.251, Read error: Connection reset by peer)
18:26adrianorg has joined IRC (adrianorg!~adrianorg@177.132.222.166)
18:39calisto has joined IRC (calisto!~calisto@unaffiliated/calisto)
18:56alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.)
19:10calisto has left IRC (calisto!~calisto@unaffiliated/calisto, Ping timeout: 264 seconds)
19:18Markus_123 has joined IRC (Markus_123!~Markus123@194-166-98-210.adsl.highway.telekom.at)
19:35cstk421 has joined IRC (cstk421!~cstk421@99-20-229-143.lightspeed.brhmmi.sbcglobal.net)
19:35
<cstk421>
how would you compare ltsp to openthinclient ?
19:36
i ask here b/c i have been unable to find an openthinclient channel
19:39
<vagrantc>
well, this is the first I've heard of openthinclient, so LTSP must be more well known. :)
19:40
<||cw>
me too, and it's hard to tell since all their docs redirect to the same wiki page that's locked in german
19:41
but it looks like maybe it's strictly a thin client system, where you'd use rdp or vnc or citrix, where ltsp's client can also, and primarily, give you a linux desktop from the server, or run a full diskless linux desktop on the client
19:41
<cstk421>
yeah hard to navigate but it has quite a bit of documentation
19:42
i pulled the vm image and got it working in about 30 min not bad
19:42
<vagrantc>
plug "openthinclient vs. ltsp" into a search engine, you'll find some reviews
19:43
sounds like it's ubuntu-based, at least the client OS
19:43
<cstk421>
im reading about ltsp now and im trying to get an understanding of what i want. I am looking for a pxe boot to a windows os. As far as i can see ltsp and all the other solutions just have the thinclient rdp to a windows vm. Do i have this right ?
19:43
i didnt think thats how the citrix vdi environment worked
19:43
<||cw>
cstk421: with ltsp you can make a xfreerdp session that connects to windows
19:44
and yes, that's how vdi works
19:44
packaged vdi's will have their own client though
19:44
<vagrantc>
mostly seems like a management frontend
19:44
<cstk421>
||cw: so the citrix vdi model is the same ? and by that i dont mean xenapp but xendesktop
19:45
<||cw>
I'm not familiar with xen's products
19:45
<cstk421>
ltsp doesnt actually run the windows vm's does it ?
19:45
<||cw>
no
19:45
ltsp is not a vdi solution, but it can be used as one
19:45
<cstk421>
||cw: understood.
19:45
<vagrantc>
by and large, anything you can run on a linux system will run on LTSP, either locally on the thin client hardware, or remotely on the server... so it is as flexible as your OS.
19:46
<||cw>
you have to manage the vm's and the client->vm mapping yourself
19:46
<cstk421>
got u
19:46
are there more "total
19:46
sorry mistyped
19:47
do you guys know of a full "vdi" solution thats opensource ? or is ltsp the closest ill get ?
19:47
<||cw>
I have not seen a full OSS vdi system
19:48
<cstk421>
grrrr that figures
19:48* vagrantc isn't familiar with the "vdi" acronym
19:48
<cstk421>
what would be the benefit of the vdi solution handling both the terminal server and vm's ? aside from management ? do i lose functionality going the LTSP route ?
19:49
<||cw>
virtual desktop infrastructure
19:49
well, terminal server would not be vdi...
19:49
<vagrantc>
https://en.wikipedia.org/wiki/Desktop_virtualization#Virtual_desktop_infrastructure
19:49
<cstk421>
||cw: dont understand what you mean by that ?
19:49
<||cw>
but a lot of vdi can be done using a terminal server instead
19:50
vdi means you have 1 vm per user
19:50
<cstk421>
||cw: so your saying vdi solutions dont use a TS type environment correct ?
19:50
<vagrantc>
it seems like LTSP could be once piece of a VDI infrastructure
19:50
<||cw>
it's still 1:1 os oer user, just like traditional desktops, but the virtualization server handles all the workload
19:50
oer/per/
19:51
<cstk421>
||cw: so with ltsp each users vm will handle each users workload as opposed to the virtualization server doing all the work for all the vms ?
19:51
<||cw>
it uses rdp or vnc or whatever the OS native remote access is, just as a terminal server would
19:51
<vagrantc>
so VDI basically gives each user access to their own Virtual Machine?
19:51
<cstk421>
||cw: they dont get the benefit of all the server resources
19:51
<||cw>
no
19:52
vagrantc: yes, exactly.
19:52
<vagrantc>
cstk421: with LTSP's default setup, all users log in to the same server, virtual or physical...
19:52
<||cw>
cstk421: ltsp is a terminal server. vdi is not a terminal server.
19:52
<cstk421>
||cw: i thought in the vdi model they are still limited to the resources of their respected vm
19:52
||cw: understood
19:52
vagrantc: got u
19:53
<vagrantc>
cstk421: but there's nothing to prevent you from configuring the login process to spin up a virtual machine that you connect to or something like that.
19:53
<||cw>
a terminal server is ONE server and all users login to it, multiuser, multi-session. vdi is one use, one session, multiple virtual servers (but desktop OS)
19:53
<vagrantc>
cstk421: other than perhaps the amount of effort...
19:54
<cstk421>
either im not understanding both infrastructures or there really isnt a difference other then the communication method between client and server.
19:54
<||cw>
the communication method is the same, the server side is very different
19:55
a true VDI system would use a session broker, much like a cluster of terminal servers, to direct specific users to specific virtual machines, or a pool of virtual machines
19:55
<cstk421>
k
19:56
<||cw>
it also would use a modified client that understand this, it's still rdp, but it's not rdesktop or xfreerdp
19:56
<cstk421>
let me see if this is what i am understanding. in ltsp you have the terminal server and multiple users logging into the server. you have 1 vm running 7 for example. you then have multiple users using that same instance of 7 with their own desktops?
19:57
<vagrantc>
cstk421: it doesn't even have to be a virtual machine
19:57
<||cw>
ltsp does not do windows directly
19:57
<cstk421>
vm or not does the layout i explained make sense ?
19:57
<vagrantc>
"running 7" ?
19:58
<||cw>
ltsp does linux. if you want windows, you can set up the lts.conf to do a xfreerdp session to a windows ip
19:58
<cstk421>
||cw: ok exactly what i want to do. so in this example i need a winbox for each user needing a session correct ?
20:03
<||cw>
if you want 1:1 vdi, yes
20:04
<cstk421>
||cw: ok i understand that part what is the alternative of the 1:1 vdi ? again sticking with windows
20:04
<||cw>
buying windows terminal server, then all clients can point to the one server IP. but your applications will need be TS compatible
20:05
annoyingly, cost is about the same, give or take a couple %. MS has not made it an easy choice
20:06
I gotta run though
20:07
<cstk421>
||cw: hey thanks for your help
20:07
||cw: really appreciate it
20:31vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Quit: leaving)
20:38calisto has joined IRC (calisto!~calisto@unaffiliated/calisto)
20:43calisto has left IRC (calisto!~calisto@unaffiliated/calisto, Ping timeout: 240 seconds)
20:56cstk421 has left IRC (cstk421!~cstk421@99-20-229-143.lightspeed.brhmmi.sbcglobal.net, Remote host closed the connection)
20:58brianr has left IRC (brianr!~brianr@out.ewbc.com, Quit: Leaving)
21:16khildin has left IRC (khildin!~khildin@ip-213-49-86-81.dsl.scarlet.be, Quit: I'm gone, bye bye)
21:26vmlintu has left IRC (vmlintu!~vmlintu@83.145.240.143, Ping timeout: 258 seconds)
22:21lmds_ has left IRC (lmds_!~lmds@tui.pi-et-ro.net, Read error: Connection reset by peer)
23:15gbaman has left IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com, Remote host closed the connection)
23:26gbaman has joined IRC (gbaman!~gbaman@host81-130-112-2.in-addr.btopenworld.com)