IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 21 November 2010 (all times are UTC)
| 00:40 | alkisg has quit IRC | |
| 00:44 | sweetpi has joined #ltsp | |
| 01:12 | johnny has joined #ltsp | |
| 01:25 | itcharlie has quit IRC | |
| 01:25 | itcharlie has joined #ltsp | |
| 01:58 | itcharlie has quit IRC | |
| 01:58 | itcharlie1 has joined #ltsp | |
| 04:22 | gorkhaan has joined #ltsp | |
| 04:44 | komunista has joined #ltsp | |
| 04:49 | sweetpi has quit IRC | |
| 05:41 | yorick has joined #ltsp | |
| 05:41 | yorick has left #ltsp | |
| 05:46 | Trixboxer has joined #ltsp | |
| 06:00 | gorkhaan has quit IRC | |
| 06:04 | itcharlie1 has quit IRC | |
| 06:04 | itcharlie has joined #ltsp | |
| 06:19 | itcharlie1 has joined #ltsp | |
| 06:19 | itcharlie has quit IRC | |
| 06:30 | alkisg has joined #ltsp | |
| 06:46 | itcharlie1 has quit IRC | |
| 07:00 | alkisg has quit IRC | |
| 07:04 | alkisg has joined #ltsp | |
| 07:08 | alkisg has quit IRC | |
| 07:28 | litlebuda has joined #ltsp | |
| 07:39 | litlebuda has quit IRC | |
| 07:39 | dlezcano has quit IRC | |
| 07:40 | litlebuda has joined #ltsp | |
| 07:48 | dlezcano has joined #ltsp | |
| 07:58 | alkisg has joined #ltsp | |
| 08:23 | gentgeen__ has joined #ltsp | |
| 08:28 | litlebuda has quit IRC | |
| 08:30 | litlebuda has joined #ltsp | |
| 08:30 | gentgeen__ has quit IRC | |
| 08:31 | gentgeen__ has joined #ltsp | |
| 08:34 | gentgeen__ has quit IRC | |
| 08:34 | gentgeen__ has joined #ltsp | |
| 08:39 | gentgeen__ has joined #ltsp | |
| 08:43 | gentgeen__ has joined #ltsp | |
| 08:57 | gentgeen__ has quit IRC | |
| 09:01 | gentgeen__ has joined #ltsp | |
| 09:02 | gentgeen__ has joined #ltsp | |
| 09:06 | gentgeen__ has joined #ltsp | |
| 09:07 | gentgeen__ has quit IRC | |
| 09:12 | sweetpi has joined #ltsp | |
| 09:13 | gentgeen__ has joined #ltsp | |
| 09:15 | gentgeen__ has quit IRC | |
| 09:16 | gentgeen__ has joined #ltsp | |
| 09:16 | gentgeen__ has quit IRC | |
| 09:17 | gentgeen__ has joined #ltsp | |
| 09:18 | Kicer86 has joined #ltsp | |
| 10:02 | dgroos has joined #ltsp | |
| 10:04 | gentgeen__ has joined #ltsp | |
| 10:09 | dgroos has quit IRC | |
| 10:09 | dgroos has joined #ltsp | |
| 10:22 | gorkhaan has joined #ltsp | |
| 10:31 | Da-Geek has joined #ltsp | |
| 11:06 | sweetpi has quit IRC | |
| 11:14 | sweetpi has joined #ltsp | |
| 11:17 | litlebuda has quit IRC | |
| 11:20 | litlebuda has joined #ltsp | |
| 11:27 | litlebuda has quit IRC | |
| 11:28 | litlebuda has joined #ltsp | |
| 11:29 | litlebuda has quit IRC | |
| 11:30 | litlebuda has joined #ltsp | |
| 11:38 | Steve_the_Pirate has joined #ltsp | |
| 11:40 | * Da-Geek waves @ Steve_the_Pirate | |
| 11:45 | litlebuda has quit IRC | |
| 11:46 | litlebuda has joined #ltsp | |
| 11:59 | alkisg has quit IRC | |
| 13:16 | ltspbot has joined #ltsp | |
| 13:19 | alkisg has quit IRC | |
| 13:31 | bobby_C has joined #ltsp | |
| 13:48 | scottmaccal has quit IRC | |
| 13:48 | scottmaccal_ has quit IRC | |
| 14:02 | Da-Geek has quit IRC | |
| 14:05 | Steve_the_Pirate has quit IRC | |
| 14:22 | Kicer86 has quit IRC | |
| 14:28 | mistik1_ has joined #ltsp | |
| 14:32 | mistik1 has quit IRC | |
| 14:32 | mistik1_ is now known as mistik1 | |
| 15:30 | itcharlie has left #ltsp | |
| 15:51 | wideyes has joined #ltsp | |
| 15:53 | wideyes has joined #ltsp | |
| 16:00 | <wideyes> Hello all. I have an error related to the issue in this bug report: https://bugs.launchpad.net/ubuntu/+source/ltsp/+bug/457702
| |
| 16:01 | basically, thin clients having squashfs-related errors upon reboot, presumably because their network connections are cut off before other crucial shutdown procedures
| |
| 16:02 | the issue I'm trying to tackle is having thin clients successfully handle a server reboot. As it is now I have some thin clients handling the reboot just fine, with others suffering the squashfs errors
| |
| 16:03 | the bug report I showed is old - looks like these issues had been resolved, yet I am having (what I believe to be) the same issue and thought I'd check in here to see if there's any news on this
| |
| 16:07 | alkisg has joined #ltsp | |
| 16:14 | * Hyperbyte frowns | |
| 16:15 | <Hyperbyte> Sorry, happens whenever I see the words 'server' and 'reboot' in one line. :-)
| |
| 16:16 | <wideyes> well, they do have to happen from time to time...
| |
| 16:17 | looking to save me or the staff having to go around to restart 20 thin clients any time the server might go down for reboot
| |
| 16:20 | <Hyperbyte> If by "from time to time" you mean "once a year", then yes, I agree. :-)
| |
| 16:20 | Hehe, sorry.. I'm not helping am I? ;-)
| |
| 16:20 | <wideyes> heh, well, not with the task at hand, anyway ;)
| |
| 16:22 | <Hyperbyte> I'm sure someone can help you, but not me. I think the channel is usually busiest during European office hours
| |
| 16:22 | <wideyes> it's also a nice reassurance to have that, should the server crash, I can just reboot it remotely and have assurance that the workstations will be up and running
| |
| 16:22 | oh, sure
| |
| 16:23 | well, friendlier than a silent room, anyway :)
| |
| 16:23 | <Hyperbyte> Me - I just send out a memo to 'shutdown before you leave - or else!' and then plan an 'at' task to reboot a little before midnight. :P
| |
| 16:24 | <alkisg> Hyperbyte: reboot once a year? I get kernel upgrades 10 times a year.... :D
| |
| 16:24 | <Hyperbyte> alkisg: constant kernel updates are for pussies. :D
| |
| 16:24 | <wideyes> alkisg, I know I've been bugging you, but I've seen you comment on this issue before, and maybe you have a quick answer: I've been getting squashfs errors from some but not all thin clients upon rebooting the thin client server
| |
| 16:24 | <alkisg> And security updates too?
| |
| 16:25 | bobby_C has quit IRC | |
| 16:25 | <Hyperbyte> Production systems I really only upgrade when a) I have a problem or b) I need to restart for a hardware update
| |
| 16:25 | <wideyes> some handle it gracefully, other's will just throw out a long litany of shashfs errors
| |
| 16:25 | <alkisg> wideyes: distro/version?
| |
| 16:25 | <wideyes> sorry, typing is sloppy today :/
| |
| 16:25 | Ubuntu 10.04
| |
| 16:26 | 64 bit server edition with 32-bit chroot environment
| |
| 16:26 | <alkisg> wideyes: that bug report is only about rebooting clients, is that when you're getting the problem?
| |
| 16:27 | <wideyes> I figured it was somewhat related, but no, it's not the exact problem. I am getting the errors on the clients when I reboot the server
| |
| 16:27 | only some of them, though
| |
| 16:27 | <alkisg> Erm, sorry, what? :D
| |
| 16:28 | You expect the clients to work fine when you reboot the server?
| |
| 16:28 | They're connected to the network disk on the server...
| |
| 16:28 | <wideyes> oh, no, I just hope they will reconnect
| |
| 16:28 | so I don't have to restart 20+ clients manually
| |
| 16:28 | works like a charm on my old ltsp implementation
| |
| 16:29 | not with all of these clients, unfortunately
| |
| 16:29 | I understand the somewhat circular logic of what I'm asking
| |
| 16:29 | <alkisg> Well... supposedly with nbd-proxy that would work, or even with nbd-client -persist (which was dropped)
| |
| 16:29 | <wideyes> some clients have no trouble even with what I'm running now
| |
| 16:29 | <alkisg> I don't trust any of those for server reboots
| |
| 16:30 | I'd just make a script to reboot the clients when the server reboots..
| |
| 16:30 | <wideyes> any suggestions for a good way to do that?
| |
| 16:30 | <alkisg> Btw, if your clients are logged on, it won't work
| |
| 16:30 | <wideyes> ah
| |
| 16:30 | <alkisg> The ssh connection can't be reconnected afaik
| |
| 16:30 | <wideyes> shame
| |
| 16:32 | the environment is this: the terminals are basically dumb workstations, just for internet use and one server-based software. No data is potentially lost on a reboot. Unfortunately, nobody else in the building has a flying clue about how to reboot a pc
| |
| 16:32 | <Hyperbyte> alkisg: it would be a nice feature maybe if an LTSP client would automatically reboot whenever an SSH/NDB/NTFS connection is lost maybe?
| |
| 16:33 | <wideyes> when the server reboots, does it issue some kind of shutdown/logout procedure in the thin clients? Or do they just die mid-process?
| |
| 16:33 | <alkisg> The nbd and nfs connections could be reconnected, but I don't think that's the case with ssh
| |
| 16:33 | <wideyes> hm
| |
| 16:33 | <alkisg> wideyes: nope, it doesn't issue anything
| |
| 16:34 | <wideyes> ah, ok
| |
| 16:34 | <alkisg> But you can tell it to issue some commands, it would require a little scripting though
| |
| 16:34 | <wideyes> I'd be willing to look into that
| |
| 16:34 | <Hyperbyte> Shouldn't be too bad actually?
| |
| 16:34 | <wideyes> yeah, sounds like what I'd need
| |
| 16:34 | <Hyperbyte> Just make matching keys in ~/.ssh/ on both server and client chroot
| |
| 16:34 | <alkisg> There are some methods for automated shutdown/reboot etc at the ubuntultsp wiki: https://help.ubuntu.com/community/UbuntuLTSP/
| |
| 16:35 | <Hyperbyte> And from server run shutdown command via ssh on logged in clients?
| |
| 16:35 | <alkisg> and also sbalneav has described a script that listens on a port for reboot commands, on the ltsp mailing list
| |
| 16:35 | <Hyperbyte> Or am I saying something dumb now? :P
| |
| 16:35 | <alkisg> ssh would work too, but it's a little insecure to share the keys, so some people don't want to install ssh to the chroot
| |
| 16:36 | <wideyes> good resources. I will follow through, and thanks :)
| |
| 16:36 | <alkisg> Even a cron job on the client that pings the server, and reboots if it doesn't get an answer, will also do
| |
| 16:36 | Hmm no, not cron
| |
| 16:36 | <Hyperbyte> alkisg: question would be, doesn't cron crash when a client looses it's /?
| |
| 16:37 | <alkisg> Hyperbyte: right, that's why I thought it again and said not cron, it would need to be loaded in the client ram to work
| |
| 16:37 | <Hyperbyte> I think it's best to issue reboot command to clients before you reboot server, because once you reboot server, nothing on the clients should actually anymore... with a few exceptions maybe.
| |
| 16:38 | + work ^
| |
| 16:39 | Interesting stuff, anyway.
| |
| 16:39 | <wideyes> the thin client server is a virtual machine and reboots very quickly, so I'll bet issuing reboot to all clients right before server reboot would allow plenty of time
| |
| 16:41 | <alkisg> wideyes: how's your shell scripting?
| |
| 16:41 | <wideyes> so-so
| |
| 16:42 | I think this one's doable for me, unless there's some complications unforseen by me
| |
| 16:42 | <alkisg> E.g., on the thin client: while true; do command=$(nc -l 1234); if [ "$command" = "poweroff" ]; then poweroff; fi; done
| |
| 16:42 | On the server, you can grep for nbd connections and "echo poweroff | nc ip 1234"
| |
| 16:43 | <wideyes> huh. So the question would be, how do I get that script to run on the clients? Is that what I'm reading up on at the moment?
| |
| 16:43 | a good simple solution, though
| |
| 16:44 | <alkisg> !lts.conf
| |
| 16:44 | <ltspbot> alkisg: "lts.conf" :: http://manpages.ubuntu.com/lts.conf
| |
| 16:44 | <alkisg> See RCFILE_01 to 10 there
| |
| 16:44 | <wideyes> ok, good deal, that page you referred me to has the answers, I believe
| |
| 16:53 | alkisg has quit IRC | |
| 17:02 | dgroos_ has joined #ltsp | |
| 17:03 | dgroos_ has quit IRC | |
| 17:05 | dgroos has quit IRC | |
| 17:14 | komunista has quit IRC | |
| 17:50 | litlebuda has quit IRC | |
| 18:13 | wideyes has quit IRC | |
| 18:25 | yanu has quit IRC | |
| 18:25 | yanu has joined #ltsp | |
| 18:25 | yanu has joined #ltsp | |
| 18:57 | sweetpi has quit IRC | |
| 20:17 | chupacabra has joined #ltsp | |
| 21:01 | vagrantc has joined #ltsp | |
| 21:26 | daya has joined #ltsp | |
| 22:07 | mistik1_ has joined #ltsp | |
| 22:08 | mistik1 has quit IRC | |
| 22:08 | mistik1_ is now known as mistik1 | |