IRC chat logs for #ltsp on irc.libera.chat (webchat)
Channel log from 21 March 2013 (all times are UTC)
| 00:05 | Parker955_Away is now known as Parker955 | |
| 00:33 | cpm__bis has left IRC (cpm__bis!~cpm@78.193.56.105, Remote host closed the connection) | |
| 00:43 | vagrantc has left IRC (vagrantc!~vagrant@freegeek/vagrantc, Ping timeout: 260 seconds) | |
| 01:34 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Ping timeout: 246 seconds) | |
| 02:13 | jammcq has left IRC (jammcq!~jam@c-69-245-75-255.hsd1.mi.comcast.net, Quit: leaving) | |
| 02:46 | staffencasa has left IRC (staffencasa!~staffenca@8-220.ptpg.oregonstate.edu, Ping timeout: 255 seconds) | |
| 03:17 | Parker955 is now known as Parker955_Away | |
| 04:08 | andygraybeal_ has left IRC (andygraybeal_!~andy@h7.213.22.98.dynamic.ip.windstream.net, Ping timeout: 272 seconds) | |
| 05:14 | alexqwesa has left IRC (alexqwesa!~alex@109.172.12.47, Quit: Хана X'ам !!!) | |
| 05:23 | alexqwesa has joined IRC (alexqwesa!~alex@109.172.12.47) | |
| 06:38 | ogra_ has left IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de, Ping timeout: 245 seconds) | |
| 06:39 | ogra_ has joined IRC (ogra_!~ogra_@p5098ed03.dip0.t-ipconnect.de) | |
| 07:07 | teacher has joined IRC (teacher!c23fe7c5@gateway/web/freenode/ip.194.63.231.197) | |
| 07:45 | epoptes_user9 has joined IRC (epoptes_user9!7b8a20da@gateway/web/freenode/ip.123.138.32.218) | |
| 07:46 | <epoptes_user9> hi. I've come with a question about epoptes. anyone here familiar with it?
| |
| 07:47 | khildin has joined IRC (khildin!~khildin@ip-80-236-227-61.dsl.scarlet.be) | |
| 07:49 | work_alkisg is now known as alkisg | |
| 07:49 | <alkisg> !ask | echo epoptes_user9:
| |
| 07:49 | <ltsp> epoptes_user9: ask: Don't ask to ask a question, simply ask it, and if someone knows the answer, they'll respond. Please hang around for at least a full hour after asking a question, as not everybody constantly monitors the channel.
| |
| 07:52 | <epoptes_user9> ltsp: thanks. just downloaded epoptes to opensuse 12.3. When I try and open it I get the message: "User jon must be a member of group epoptes to run epoptes." Question: how do I add user jon to group epoptes?
| |
| 07:53 | <alkisg> epoptes_user9: gpasswd -a epoptes jon
| |
| 07:53 | With sudo or su, whatever opensuse uses
| |
| 07:54 | <epoptes_user9> alkisg: i'll try it now thanks.
| |
| 07:55 | <alkisg> epoptes_user9: note that you need to logout/login again, in order for group changes to take effect
| |
| 07:55 | There should also be some GUI tool in opensuse for user management
| |
| 07:55 | <epoptes_user9> alkisg: i get the error message: "user 'epoptes' does not exist"
| |
| 07:56 | <alkisg> epoptes_user9: sorry, it's the other way around then: gpasswd -a jon epoptes
| |
| 07:56 | <epoptes_user9> alkisg: ok
| |
| 07:56 | looks good. i'll logout/in now
| |
| 07:57 | epoptes_user9 has left IRC (epoptes_user9!7b8a20da@gateway/web/freenode/ip.123.138.32.218, Quit: Page closed) | |
| 07:59 | wiki3 has joined IRC (wiki3!51ba184e@gateway/web/freenode/ip.81.186.24.78) | |
| 08:01 | test_ has joined IRC (test_!51ba184e@gateway/web/freenode/ip.81.186.24.78) | |
| 08:01 | test_ has left IRC (test_!51ba184e@gateway/web/freenode/ip.81.186.24.78) | |
| 08:13 | alkisg is now known as work_alkisg | |
| 08:29 | <elias_a> !lastseen vmlintu
| |
| 08:29 | <ltsp> Error: "lastseen" is not a valid command.
| |
| 08:29 | <elias_a> Oops.
| |
| 08:51 | <knipwim> !seen vmlintu
| |
| 08:51 | <ltsp> vmlintu was last seen in #ltsp 1 week, 1 day, 23 hours, 26 minutes, and 1 second ago: <vmlintu> alkisg: where can I find ltspd?
| |
| 08:57 | <elias_a> knipwim: Thank you!
| |
| 09:01 | anunnaki has left IRC (anunnaki!~christoph@c-174-54-115-236.hsd1.pa.comcast.net, Remote host closed the connection) | |
| 09:16 | vmlintu has joined IRC (vmlintu!~vmlintu@cs181214103.pp.htv.fi) | |
| 09:18 | <vmlintu> elias_a: how's it going?
| |
| 09:21 | teacher is now known as dorita | |
| 09:33 | <elias_a> vmlintu: Wow! The invisible man has returned! :D
| |
| 09:34 | <vmlintu> elias_a: just too busy coding a bettter world
| |
| 09:41 | Mo has joined IRC (Mo!c100f604@gateway/web/freenode/ip.193.0.246.4) | |
| 09:45 | work_alkisg is now known as alkisg | |
| 09:58 | dobber_ has joined IRC (dobber_!~dobber@89.190.199.210) | |
| 10:05 | vmlintu_ has joined IRC (vmlintu_!~vmlintu@cs181214103.pp.htv.fi) | |
| 10:07 | vmlintu has left IRC (vmlintu!~vmlintu@cs181214103.pp.htv.fi, Ping timeout: 245 seconds) | |
| 10:10 | dorita has left IRC (dorita!c23fe7c5@gateway/web/freenode/ip.194.63.231.197, Ping timeout: 245 seconds) | |
| 10:25 | Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas) | |
| 10:49 | adrianorg__ has joined IRC (adrianorg__!~adrianorg@177.132.221.56) | |
| 10:53 | adrianorg_ has left IRC (adrianorg_!~adrianorg@177.132.221.56, Ping timeout: 276 seconds) | |
| 11:08 | alkisg has left IRC (alkisg!~alkisg@plinet.ioa.sch.gr, Ping timeout: 245 seconds) | |
| 11:13 | work_alkisg has joined IRC (work_alkisg!~alkisg@150.140.47.54) | |
| 11:13 | work_alkisg is now known as alkisg | |
| 11:25 | ltspuser_69 has joined IRC (ltspuser_69!c82072f7@gateway/web/freenode/ip.200.32.114.247) | |
| 11:25 | Gremble has joined IRC (Gremble!~Ben@cpc35-aztw23-2-0-cust207.18-1.cable.virginmedia.com) | |
| 11:44 | alkisg is now known as work_alkisg | |
| 12:25 | bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) | |
| 12:32 | monteslu has left IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net, Read error: Operation timed out) | |
| 12:39 | bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Quit: Goin' down hard) | |
| 12:40 | bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) | |
| 13:04 | Phantomas has left IRC (Phantomas!~Phantomas@ubuntu/member/phantomas) | |
| 13:33 | dsugar100 has joined IRC (dsugar100!~dsugar@columbia.tresys.com) | |
| 13:40 | <sbalneav> Morning all
| |
| 13:57 | <Hyperbyte> Morning. :)
| |
| 13:59 | monteslu has joined IRC (monteslu!~monteslu@ip68-109-174-213.ph.ph.cox.net) | |
| 14:00 | <elias_a> Mourning!
| |
| 14:01 | <ogra_> !s
| |
| 14:01 | <ltsp> s: Scotty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
| |
| 14:03 | Gremble has left IRC (Gremble!~Ben@cpc35-aztw23-2-0-cust207.18-1.cable.virginmedia.com, Quit: I Leave) | |
| 14:03 | vnc786 has left IRC (vnc786!~chatzilla@49.248.129.178, Read error: No route to host) | |
| 14:05 | <sbalneav> Hey ogra_!
| |
| 14:05 | vnc786 has joined IRC (vnc786!~chatzilla@49.248.129.178) | |
| 14:06 | * ogra_ hands sbalneav a https://lh6.googleusercontent.com/-RGqGhbZCxmI/UUraE5SMbGI/AAAAAAAABWE/IRwgZvae-5c/s707/smooth_canadian.jpg | |
| 14:07 | garymc has joined IRC (garymc!~chatzilla@host81-139-87-129.in-addr.btopenworld.com) | |
| 14:08 | <sbalneav> That's what all us Canadians look like, you know.
| |
| 14:09 | http://iraffiruse.net/post/787221983
| |
| 14:10 | <ogra_> haha
| |
| 14:23 | ||cw has left IRC (||cw!~chris@68.187.0.66, Changing host) | |
| 14:23 | ||cw has joined IRC (||cw!~chris@phpgroupware/cw) | |
| 14:27 | staffencasa has joined IRC (staffencasa!~staffenca@8-220.ptpg.oregonstate.edu) | |
| 14:30 | Phantomas has joined IRC (Phantomas!~Phantomas@ubuntu/member/phantomas) | |
| 14:42 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 14:46 | cpm__bis has joined IRC (cpm__bis!~cpm@LPuteaux-151-43-17-232.w217-128.abo.wanadoo.fr) | |
| 14:49 | <cpm__bis> Hello!!! I am in a classroom in France where 20 pc are running well under LTSP. I am enjoy! :-)
| |
| 14:50 | (elementary school)
| |
| 14:50 | Question: if several students connect with the same user then Firefox won't launch twice or more.
| |
| 14:51 | Any idea to solve this case without create one user by student?
| |
| 14:52 | <laurense> I think you must somehow try to prevent firefox creating the lock file
| |
| 14:52 | or to let the clients create the lock file on a lokal temporary filesystem
| |
| 14:53 | Ik don't know how to implement it in ltsp, also a newbie on that area
| |
| 14:54 | but on starting firefox it will check for some kind of lock file
| |
| 14:55 | <cpm__bis> In this way, you hope that Firefox's file are sharable ^^
| |
| 14:56 | Thank's for the idea. :o)
| |
| 14:57 | alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg) | |
| 14:58 | <laurense> cpm__bis: alkisg helped me a lot a while ago, maybe he has a good idea about this problem?
| |
| 14:58 | <alkisg> Hi guys
| |
| 14:58 | * alkisg checks the irclogs... | |
| 14:58 | <laurense> unfortunatly he missed the conversation
| |
| 14:59 | <alkisg> cpm__bis: it's strongly recommented that you use a different user per pc. Not the same user for all PCs.
| |
| 15:01 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Ping timeout: 276 seconds) | |
| 15:02 | <laurense> cpm__bis: if it doesn't matter which user the students are working with you could look into auto login with a user per workstation
| |
| 15:03 | <cpm__bis> alkisg, Hi. Yes, It is the way I am working on. But I try to check other way. From what you tell me, there is no new way. Thank's a lot :-)
| |
| 15:03 | <laurense> or enabling the guest logon
| |
| 15:03 | that way creating a profile for the duration of the session
| |
| 15:04 | <alkisg> Local guest accounts is another thing, yeah, they don't even need to be server users
| |
| 15:04 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 15:05 | <cpm__bis> hooo, indeed, it is a solution but how configure it? I know how do it for one user on one computer.
| |
| 15:06 | I will check the lts.conf possibilities.
| |
| 15:06 | <alkisg> You don't want the users to be able to save data?
| |
| 15:06 | <cpm__bis> Of course YES :)
| |
| 15:06 | <alkisg> So guest accounts are NOT what you want :)
| |
| 15:06 | Go on with the "one user per client" plan
| |
| 15:07 | <cpm__bis> oki, many thanks.
| |
| 15:08 | Another point: have you got a URL explaining how configure default GNOME 2 desktop. I'd like configure same menu and icons for all students.
| |
| 15:08 | <alkisg> Gnome 2? Which distribution and version are you using?
| |
| 15:08 | <cpm__bis> Debian squeeze stable + backports.
| |
| 15:09 | <alkisg> Ah, ok. There are docs for that in gnome.org
| |
| 15:09 | <cpm__bis> I choose it because use Gnome 2 that is really good for student. ^^
| |
| 15:09 | jammcq has joined IRC (jammcq!~jam@c-69-245-75-255.hsd1.mi.comcast.net) | |
| 15:09 | <alkisg> But those are windows techniques
| |
| 15:09 | <jammcq> good morning friends
| |
| 15:09 | <alkisg> Morning jammcq
| |
| 15:09 | <cpm__bis> hello
| |
| 15:09 | <alkisg> cpm__bis: it's much easier in linux if you let each student have his own desktop, settings, files etc
| |
| 15:11 | <cpm__bis> Students are very young, the idea was to put selected icon on the desktop. But you are right, learn them to put icon on the desktop is a good idea. I will push it. :o)
| |
| 15:11 | <alkisg> Teachers here start with the same objectives, "how to lock everything", until they realize it's much easier to let students do whatever they want, and just reset stuff when needed
| |
| 15:12 | You can have a script put icons or symlinks to their desktops on login
| |
| 15:12 | That's quite easy
| |
| 15:13 | <cpm__bis> really? Good news. I will try this. I am goto the Gnome 2 documentation. Thank's you alkisg, laurense for your help. Have a good day from Paris (France)!
| |
| 15:13 | <alkisg> Good day from Greece to you too :)
| |
| 15:14 | <laurense> no problem
| |
| 15:14 | Good day form the Netherlands to you too
| |
| 15:14 | <cpm__bis> hahaah, excellent! :o)
| |
| 15:16 | ukfb has joined IRC (ukfb!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 15:17 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 15:17 | ukfb is now known as |Paradox| | |
| 15:25 | Mo has left IRC (Mo!c100f604@gateway/web/freenode/ip.193.0.246.4, Ping timeout: 245 seconds) | |
| 15:27 | <alkisg> cpm__bis: also, if you want to move to wheezy, this is similar to gnome2: http://packages.debian.org/search?keywords=gnome-session-fallback
| |
| 15:30 | <sbalneav> alkisg: So, this weekend, I'm going to try to glue together udisks-glue, sshfs, and some scripting I've been working on for auto-key generation into a replacement localdev implementation.
| |
| 15:31 | <cpm__bis> alkisg, thank, I tested gnome session fallback and I do not approve it. :-/
| |
| 15:38 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Ping timeout: 256 seconds) | |
| 15:42 | <alkisg> sbalneav: cool! we don't use localdevs at all here though so I won't be able to test them much, but it certainly is a good step towards ltsp6
| |
| 15:47 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 15:49 | <alkisg> So wait, no ldm, cdpinger, getltscfg, xatomwait, ltspfs, ... LTSP 6 won't have .c files at all?! :D
| |
| 15:50 | <jammcq> :(
| |
| 15:50 | <alkisg> ...we could also rewrite ldminfod in shell/inetd, and remove python too :P
| |
| 15:52 | dead_inside has joined IRC (dead_inside!~dead_insi@76.75.3.174) | |
| 15:55 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 15:56 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 15:56 | cpm__bis has left IRC (cpm__bis!~cpm@LPuteaux-151-43-17-232.w217-128.abo.wanadoo.fr, Remote host closed the connection) | |
| 16:17 | garymc has left IRC (garymc!~chatzilla@host81-139-87-129.in-addr.btopenworld.com, Remote host closed the connection) | |
| 16:19 | <Phantomas> alkisg: and ltspd? :p
| |
| 16:20 | Or you named ltspd ldminfod? :P
| |
| 16:21 | <sbalneav> alkisg: Well, it will have the pam module.
| |
| 16:25 | <alkisg> The pam module is a separate package
| |
| 16:26 | Phantomas: that's what I was saying "rewrite ldminfod in shell", but I was just joking about it
| |
| 16:27 | <Phantomas> alkisg: I thought we were calling it ltspd :P
| |
| 16:27 | <sbalneav> Plus, we may still need xatomwait, but if we're going to run an sshd down on the thin client, then maybe not.
| |
| 16:27 | <alkisg> Phantomas: ldminfod is an existing component of ltsp. ltspd aims to replace it, along with lts.conf
| |
| 16:28 | <Phantomas> alkisg: ah, ok then
| |
| 16:28 | dsugar100 has left IRC (dsugar100!~dsugar@columbia.tresys.com, Remote host closed the connection) | |
| 16:28 | Da-Geek has joined IRC (Da-Geek!~Da-Geek@exch.imparando.com) | |
| 16:32 | bdqsfo has joined IRC (bdqsfo!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 16:33 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 16:33 | bdqsfo is now known as |Paradox| | |
| 16:34 | vmlintu_ has left IRC (vmlintu_!~vmlintu@cs181214103.pp.htv.fi, Ping timeout: 256 seconds) | |
| 16:38 | dobber_ has left IRC (dobber_!~dobber@89.190.199.210, Remote host closed the connection) | |
| 16:39 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: No route to host) | |
| 16:39 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 16:41 | komunista has joined IRC (komunista!~slavko@87.244.209.121) | |
| 16:54 | hontmrldbbj has joined IRC (hontmrldbbj!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 16:54 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 16:55 | hontmrldbbj is now known as |Paradox| | |
| 17:01 | vagrantc has joined IRC (vagrantc!~vagrant@freegeek/vagrantc) | |
| 17:12 | mikkel has joined IRC (mikkel!~mikkel@80-71-132-15.u.parknet.dk) | |
| 17:13 | dsugar100 has joined IRC (dsugar100!~dsugar@columbia.tresys.com) | |
| 17:23 | <Enslaver> https://admin.fedoraproject.org/updates/FEDORA-EPEL-2013-0727/ltsp-5.4.5-4.el6
| |
| 17:23 | * vagrantc takes notes | |
| 17:24 | <warren> Enslaver: I didn't expect that you'd submit it to the actual repository so soon
| |
| 17:24 | Enslaver: it didn't pass the required package reviews
| |
| 17:25 | Enslaver: please don't push it to updates, updates-testing is OK for now.
| |
| 17:25 | <Enslaver> i submitted to testing
| |
| 17:26 | <warren> yeah, just that we're violating policy to be in testing
| |
| 17:26 | <Enslaver> hows that?
| |
| 17:26 | <warren> this technically requires a package review, which currenlyt it has no chance of passing.
| |
| 17:27 | <Enslaver> yeah there's no way all of this will make it into EPEL anyway
| |
| 17:28 | i mean heck, the way we're building the client is to call other repo's
| |
| 17:29 | <warren> that can be fixed later
| |
| 17:29 | we have approval (sort of) to put our kernel in EPEL
| |
| 17:29 | like I had it before
| |
| 17:29 | <Enslaver> calling it?
| |
| 17:29 | I might have a better solution actually
| |
| 17:29 | <warren> ?
| |
| 17:30 | <Enslaver> the other day i got a nbd dkms rpm working
| |
| 17:30 | so it will compile the nbd.ko module for whatever kernel the user is running, provided the kernel-devel package is installed, thats option 1
| |
| 17:31 | option 2 is to include the initrd.ltsp with the binary server rpms
| |
| 17:31 | and only keep it in the tftpboot directory
| |
| 17:32 | that way we can keep it all at the same level, which can be different from the actual client kernel
| |
| 17:32 | <warren> dkms?
| |
| 17:32 | or the newer akmod standard?
| |
| 17:32 | trouble with either is it requires having the compiler in the chroot?
| |
| 17:33 | <Enslaver> yup, gotta have it in the chroot :( but can be removed after it's complete, just takes ltsp-build-client a bit longer
| |
| 17:33 | <warren> Enslaver: EL6 kernel is supposed to have a stable ABI, meaning with a little hacking you can make a binary nbd.ko that loads in any version of EL6 kernel
| |
| 17:33 | <Enslaver> unsigned nbd.ko
| |
| 17:33 | which i don't mind
| |
| 17:34 | i just didn't think EPEL would want that being put in people's modules directory
| |
| 17:35 | bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Ping timeout: 245 seconds) | |
| 17:36 | <warren> Enslaver: not a problem, install it with ltsp-client
| |
| 17:37 | <Enslaver> ltsp-client rpm tho, which has to be hosted on EPEL as well, correct?
| |
| 17:37 | ideally anyway
| |
| 17:39 | komunista has left IRC (komunista!~slavko@87.244.209.121, Read error: Connection reset by peer) | |
| 17:40 | komunista has joined IRC (komunista!~slavko@87.244.209.121) | |
| 17:42 | faahpua has joined IRC (faahpua!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 17:43 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 17:43 | faahpua is now known as |Paradox| | |
| 17:45 | <Enslaver> To future proof this we may want to explore something other than nbd :( it seems rh just doesnt want people having it, maybe to convert us borgs to use spice?
| |
| 17:54 | JesseC has joined IRC (JesseC!~JesseCWor@wsip-98-175-20-126.br.br.cox.net) | |
| 18:13 | alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.) | |
| 18:22 | christopher has joined IRC (christopher!~christoph@c-174-54-115-236.hsd1.pa.comcast.net) | |
| 18:25 | mikkel has left IRC (mikkel!~mikkel@80-71-132-15.u.parknet.dk, Quit: Leaving) | |
| 18:35 | alexqwesa has left IRC (alexqwesa!~alex@109.172.12.47, Quit: Хана X'ам !!!) | |
| 18:36 | alexqwesa has joined IRC (alexqwesa!~alex@109.172.12.47) | |
| 18:38 | smd_ has joined IRC (smd_!c9d98964@gateway/web/freenode/ip.201.217.137.100) | |
| 18:47 | smd_ has left IRC (smd_!c9d98964@gateway/web/freenode/ip.201.217.137.100, Quit: Page closed) | |
| 18:47 | <sbalneav> Enslaver: NFS?
| |
| 18:48 | <Enslaver> For squashfs images
| |
| 18:49 | <warren> Enslaver: make the nbd.ko that can load into any EL6 kernel
| |
| 18:49 | <Enslaver> Ive played with dmsquash-live a little, it could be a good solution, if not, then iscsi
| |
| 18:49 | <warren> Enslaver: and fedora has nbd.ko already
| |
| 18:51 | <Enslaver> I have nbd.ko compiled for all arches, is there a specific way to compile them so they load with any kernel symbols?
| |
| 18:57 | ltspuser_69 has left IRC (ltspuser_69!c82072f7@gateway/web/freenode/ip.200.32.114.247, Quit: Page closed) | |
| 19:13 | suwwjj has joined IRC (suwwjj!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 19:13 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 19:13 | suwwjj is now known as |Paradox| | |
| 19:31 | ywiuwsk has joined IRC (ywiuwsk!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 19:31 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 19:31 | ywiuwsk is now known as |Paradox| | |
| 19:39 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Ping timeout: 264 seconds) | |
| 19:40 | |Paradox| has joined IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 19:45 | mmetzger has left IRC (mmetzger!~mmetzger@99-71-214-196.lightspeed.mdldtx.sbcglobal.net, Ping timeout: 264 seconds) | |
| 19:47 | <||cw> Enslaver: if only kernel modules worked that way...
| |
| 19:47 | dead_inside has left IRC (dead_inside!~dead_insi@76.75.3.174, Quit: Computer has gone to sleep.) | |
| 19:47 | <Enslaver> i know that you have to distribute the symbols with them but i know theres a way to compile them into the module, i just can't remember
| |
| 19:51 | atykxzvravk has joined IRC (atykxzvravk!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 19:53 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: Connection reset by peer) | |
| 19:53 | atykxzvravk is now known as |Paradox| | |
| 20:02 | andygraybeal_ has joined IRC (andygraybeal_!~andy@h7.213.22.98.dynamic.ip.windstream.net) | |
| 20:06 | alkisg has joined IRC (alkisg!~alkisg@ubuntu/member/alkisg) | |
| 20:09 | <alkisg> Enslaver: did you try mounting a squashfs image as root, over NFS instead of NBD? Maybe the performance hit isn't that bad...
| |
| 20:09 | <Enslaver> yeah i had it mounted as a loop device, mounted on /opt/ltsp/i386 and it still was painful
| |
| 20:10 | <jammcq> but where was it mounted as a loop device, on the server?
| |
| 20:10 | or on the client?
| |
| 20:10 | on the server would be like the worst of both worlds
| |
| 20:11 | not even sure you could do it on the client
| |
| 20:11 | llin has joined IRC (llin!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 20:11 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: No route to host) | |
| 20:11 | llin is now known as |Paradox| | |
| 20:12 | <vagrantc> you can loopback mount files served over NFS...
| |
| 20:14 | <alkisg> Enslaver: yeah I meant on the client, from the initramfs... e.g. nfsmount server:/opt/ltsp/images/i386.img /run/i386.img; mount -o loop,ro /run/i386.img /rootmnt; pivot-root /rootmnt or something
| |
| 20:16 | <Enslaver> so nfs mount the shared folder inside the initramfs t
| |
| 20:16 | then mount it as a loop device?
| |
| 20:16 | <alkisg> Yup
| |
| 20:16 | <Enslaver> hmm
| |
| 20:16 | that one I haven't tried
| |
| 20:16 | <alkisg> You can also export a file with NFS, not a folder, right?
| |
| 20:17 | But anyway if not, the whole /opt/ltsp/images folder, ok
| |
| 20:19 | <vagrantc> you'd save something having the sqaushfs compression, and disk caching in ram might be a little more efficient
| |
| 20:19 | <alkisg> Compression + lack of metadata
| |
| 20:20 | <jammcq> my guess is it would give performance somewhere between normal NFS and NBD
| |
| 20:20 | <alkisg> It could even be a bit more stable wrt disconnects/reconnets
| |
| 20:21 | <jammcq> hmm, could be
| |
| 20:21 | <alkisg> (than nbd)
| |
| 20:21 | <jammcq> interesting idea at least
| |
| 20:21 | * alkisg tries an hdparm comparison there... | |
| 20:24 | <alkisg> hdparm => inappropriate ioctrl for device, trying dd...
| |
| 20:25 | nbd=40 MB/sec, nfs=25 MB/sec
| |
| 20:26 | <jammcq> yeah, nbd is a pretty light protocol
| |
| 20:26 | very low overhead
| |
| 20:26 | nbd is pretty chatty
| |
| 20:27 | <alkisg> Still, it's much better than plain nfs, which is about 5 times slower then compressed nbd
| |
| 20:28 | * alkisg tries a tar... | |
| 20:29 | <jammcq> what are you dd'ing? something from inside the squashfs filesystem?
| |
| 20:29 | <alkisg> The first try was /dev/nbd0 vs /mnt/i386.img (where /mnt is server:/opt/ltsp/images over NFS)
| |
| 20:29 | Now I'll try some actual files, with tar
| |
| 20:30 | <jammcq> but you aren't using squashfs, right?
| |
| 20:30 | <alkisg> The images are squashfs'ed
| |
| 20:30 | <jammcq> but you are grabbing the whole image? or something inside the image?
| |
| 20:30 | <alkisg> The first try was the whole image
| |
| 20:30 | The second try now, will be tar'ring files from inside the image
| |
| 20:31 | Both images are squashfs images
| |
| 20:31 | <jammcq> so you are mounting the squashfs file on the client?
| |
| 20:31 | and then tarring something inside it?
| |
| 20:31 | <alkisg> For the second try, yes
| |
| 20:31 | <jammcq> k
| |
| 20:31 | <alkisg> tar > /dev/null
| |
| 20:31 | <jammcq> that's the good test
| |
| 20:32 | <Enslaver> well i've been playing with iscsi enterprise target for net booting squashfs root
| |
| 20:32 | boxmgehz has joined IRC (boxmgehz!~iamparado@c-71-206-132-62.hsd1.va.comcast.net) | |
| 20:32 | |Paradox| has left IRC (|Paradox|!~iamparado@c-71-206-132-62.hsd1.va.comcast.net, Read error: No route to host) | |
| 20:32 | <jammcq> iscsi sounds interesting, but I think there's a small audience for that
| |
| 20:32 | <Enslaver> which might also allow for not requiring a /var/lib/tftpboot directory
| |
| 20:33 | boxmgehz is now known as |Paradox| | |
| 20:33 | <alkisg> Enslaver: how are you going to netboot without pxelinux?
| |
| 20:33 | <Enslaver> small audience that uses iscsi?
| |
| 20:33 | <jammcq> yeah
| |
| 20:34 | <Enslaver> I haven't seen a place without iscsi
| |
| 20:34 | <ogra_> Enslaver, in schools ?
| |
| 20:35 | <sbalneav> I don't have it, all the schools in Brazil deploying LTSP probably don't have it, the libraries in the US that use LTSP probably don't have it....
| |
| 20:35 | <jammcq> is it just a matter of setting up a service ?
| |
| 20:35 | <ogra_> i wrote the nbd implementation at a time where iscsi wasnt existing btw ...
| |
| 20:35 | <sbalneav> Heck, WE don't have it.
| |
| 20:35 | <jammcq> I know it's been in the kernel for a while
| |
| 20:35 | <ogra_> thats the simple reason why we have nbd today
| |
| 20:35 | <alkisg> nbd: 10.413 secs, vs nfs: 10.127... /me wonders if his tar command line is wrong...
| |
| 20:35 | <jammcq> sbalneav: maybe you do :)
| |
| 20:35 | <Enslaver> schools?
| |
| 20:35 | <ogra_> if it had existed back then i would have looked at it
| |
| 20:35 | <Enslaver> ogra_: i'm targeting company's
| |
| 20:35 | <jammcq> Enslaver: good, many of us aren't
| |
| 20:36 | <ogra_> yeah
| |
| 20:36 | <jammcq> Enslaver: are you assuming they have a SAN ?
| |
| 20:36 | <alkisg> Does this seem ok? root@ltsp123:/tmp/nfs# time tar -cf /dev/null usr/
| |
| 20:36 | <jammcq> of will any old linux box act as a iscsi server?
| |
| 20:36 | <ogra_> its a kernel feature
| |
| 20:37 | <Enslaver> jamm: no
| |
| 20:37 | jammcq, i'm allowing for both, but assuming nothing
| |
| 20:37 | <jammcq> Enslaver: then I don't have iscsi
| |
| 20:37 | <ogra_> so yes, you could run iscsi on the ltsp server
| |
| 20:37 | * vagrantc notes that squashfs over NFS is what the debian-live and ubuntu-live netboot uses | |
| 20:37 | <jammcq> vagrantc: over nfs ?
| |
| 20:37 | <Enslaver> vagrantc, and fedora/redhat uses dmsquash-live
| |
| 20:38 | <alkisg> vagrantc: unless I'm doing something very wrong in my benchmark... I see no difference in the actual NFS vs NBD test
| |
| 20:38 | <Enslaver> i tried it like this: root=live:/opt/ltsp/images/i386.img
| |
| 20:39 | <vagrantc> jammcq: yes, just like the tests alkisg is performing.
| |
| 20:39 | <Enslaver> i did tests yesterday
| |
| 20:39 | * vagrantc suspects debian-live can support half a dozen network booting options, from a glance at the code | |
| 20:40 | <alkisg> It also has a vast range of configuration options for the live system, apart from the boot device
| |
| 20:40 | <Enslaver> lemme guess, iscsi, AoE, nbd, nfs, gfs?
| |
| 20:41 | <ogra_> its not a matter of options, its a matter of code :)
| |
| 20:41 | write an implementation for either of these that we can use upstream :)
| |
| 20:42 | nobody will refuse an iscsi AoE or gfs implementation ;)
| |
| 20:43 | the more choices the better
| |
| 20:43 | <Enslaver> yeah I'm almost done with the iscsi one
| |
| 20:44 | <ogra_> will it work on ubuntu and debian too ?
| |
| 20:44 | <Enslaver> what iscsi-target implementation does deb / ubuntu use?
| |
| 20:44 | <muppis> open
| |
| 20:45 | <ogra_> yeah, what muppis said
| |
| 20:45 | <Enslaver> they have a target?
| |
| 20:46 | last i checked open-iscsi only had the initiator
| |
| 20:46 | <vagrantc> there's both open-iscsi and iscsitarget packages in debian
| |
| 20:46 | as well as a couple others
| |
| 20:46 | that might be related
| |
| 20:47 | <alkisg> We could probably grow some ltsp.mount=xxx kernel options, to allow for arbitrary mounts in the initramfs... e.g. here's another use case: mount a local NTFS partition, then loop-mount the squashfs image C:\i386.img from that partition, and use that for root (+ a cow of course)
| |
| 20:47 | <ogra_> well, ubuntu only has the open-iscsi packages in main so i doubt anything else is needed
| |
| 20:47 | <alkisg> That way we could easily implement a true ltsp-client-setup.exe installer for windows :P
| |
| 20:48 | <Enslaver> if it's iet then it'll work
| |
| 20:48 | iscsi enterprise target
| |
| 20:48 | <jammcq> I would hope that NFS would be the common denomenator across all distros
| |
| 20:48 | but nbd has better performance, so we should try to include that everywhere possible
| |
| 20:48 | <ogra_> anyway, what jammcq said
| |
| 20:49 | <jammcq> and I would expect iscsi would have possibly better performance than nbd
| |
| 20:49 | and it offers some interesting things, like putting the images on a san
| |
| 20:49 | * ogra_ doubts that, at least in the single img context | |
| 20:49 | <Enslaver> I would love to stick with NFS :( my boss just won't go for it
| |
| 20:50 | <alkisg> NBD has had a lot of problems recently though... like all the nbd-server instances crashing because of a problematic negotiation with a single client...
| |
| 20:50 | <jammcq> yeah, that's a problem
| |
| 20:50 | <ogra_> i suspect you wont see many differences anymore between the different network filesystem options nowadays
| |
| 20:50 | <alkisg> From what I read, iscsi is a bit slower than nbd, it has more overhead
| |
| 20:50 | And aoe is a bit lighter
| |
| 20:50 | <jammcq> I would consider a distro's LTSP implementation to be non-compliant if it doesn't support NFS
| |
| 20:50 | <Enslaver> there was something i couldn't get working with AoE, can't remember what
| |
| 20:50 | <jammcq> but I dunno if i'd also think that if it didn't support nbd
| |
| 20:51 | <ogra_> alkisg, well, likely in margins you wont really notice on a typical ltsp setup
| |
| 20:51 | <alkisg> True, I can't even notice it with NFS in my benchmarks now :-/
| |
| 20:51 | <ogra_> yeah
| |
| 20:51 | <alkisg> I'll measure boot times tomorrow
| |
| 20:52 | If they're the same, I see no reason to continue supporting nbd at all...
| |
| 20:52 | <ogra_> that shoould be more intresting
| |
| 20:52 | simply because yu wil have a lot parallel I/O
| |
| 20:52 | <muppis> I'm not fully pleasant about open-iscsi. It might be cause by server, which is ubuntu 12.04 based and cluster. When failover happens, iscsi connections are remounted as ro which causes little problems in vm.
| |
| 20:52 | <jammcq> i'd be surprised if you didn't see a big difference in boot times between NBD and NFS
| |
| 20:53 | <alkisg> jammcq: I'm talking about the squashfs file served over NFS, of course...
| |
| 20:53 | <jammcq> ah
| |
| 20:53 | <ogra_> jammcq, well, the last time we measured was what ... 2006 ?
| |
| 20:53 | or 2008
| |
| 20:53 | <alkisg> Plain NFS vs squashfs NBD was 5 times slower
| |
| 20:53 | <jammcq> ok, i'd be interested in a normal NBD vs normal NFS boot
| |
| 20:53 | <Enslaver> has NFSv4 been tested?
| |
| 20:53 | <ogra_> i would actually expect nfs to be a lot better nowadys
| |
| 20:53 | compared to back then at least
| |
| 20:54 | <jammcq> better, like 3 times slower than nbd ?
| |
| 20:54 | <ogra_> .... though a wet sponge is better than nfs was back then :)
| |
| 20:54 | so that shouldnt be hard to beat
| |
| 20:55 | i would still think nfs is slower ... but not by such a margin
| |
| 20:55 | <jammcq> btw, how does Oct-31 through Nov 3rd sound for BTS-2013 ?
| |
| 20:55 | <alkisg> In my tests (copying files from the chroot to /dev/null), NFS was about 2 times slower than NBD. But NBD was also compressed 2.5 times, so in total that was 5 times.
| |
| 20:55 | <ogra_> jammcq, any time is fine ... my UDS is virtual :)
| |
| 20:56 | * ogra_ will definitely make it this year | |
| 20:56 | <Enslaver> it appears as though ubuntu and debian both can use IET
| |
| 20:56 | <ogra_> IET ?
| |
| 20:56 | * jammcq is definitely happy to hear ogra will make it | |
| 20:56 | <Enslaver> iscsi enterprise target
| |
| 20:56 | <ogra_> ah
| |
| 20:56 | sure
| |
| 20:56 | <jammcq> Enslaver: is a target the service, (where the data is) ?
| |
| 20:57 | and an 'initiator' the client?
| |
| 20:57 | <ogra_> canonical lives from its enterprise cloud business ... would be funny if it didnt support that
| |
| 20:58 | <Enslaver> yes, target = server, initiator = client
| |
| 20:59 | <ogra_> yes, i get that
| |
| 20:59 | <jammcq> so a linux box could be a target, and the thin client would be the initiator
| |
| 20:59 | <Enslaver> yup
| |
| 21:00 | <ogra_> didnt stgraber play with iscsi back in the days with ltsp-cluster ?
| |
| 21:00 | <Enslaver> or they could have a separate open filer vm or net app NAS box acting as the target
| |
| 21:00 | <alkisg> vagrantc: have you noticed any advantages to NFS vs NBD, wrt reconnects? E.g. if you disconnect an NBD client for 1 minute, it won't come back, but if you disconnect an NFS client for 1 minute, it will?
| |
| 21:01 | <jammcq> it should
| |
| 21:01 | nfs is stateless
| |
| 21:01 | <vagrantc> alkisg: sometimes i've seen clients disconnected for over 24 hours and recover fine
| |
| 21:01 | <alkisg> NFS only, right?
| |
| 21:01 | <vagrantc> yes
| |
| 21:01 | * alkisg will try to verify tomorrow that squashfs file over NFS is as fast as squashfs file over NBD | |
| 21:02 | <Enslaver> what is "canonical" i keep hearing that word a lot
| |
| 21:02 | <ogra_> Enslaver, the company driving ubuntu
| |
| 21:02 | <Enslaver> i see
| |
| 21:02 | they trying to become the next red hat in a way?
| |
| 21:03 | <ogra_> similar to what redhat is to fedora, just completely different
| |
| 21:03 | <alkisg> vagrantc: and, any specific reason for not doing swap over NFS? (sparse files etc)
| |
| 21:03 | <Enslaver> gotcha, so almost exactly the same except not?
| |
| 21:03 | <jammcq> heh, you could say they're already there
| |
| 21:03 | <ogra_> since canonical doesnt sell an OS
| |
| 21:03 | <Enslaver> red hat doesn't either
| |
| 21:03 | <ogra_> and canonical doesnt make the rules for ubuntu
| |
| 21:03 | <Enslaver> red hat doesn't make the rules for fedora
| |
| 21:03 | <ogra_> (else we would have rolling releases by now)
| |
| 21:03 | <vagrantc> alkisg: swap over NFS caused kernel deadlocks
| |
| 21:04 | <alkisg> Ah
| |
| 21:04 | <jammcq> yeah, swap over nfs is troublesome
| |
| 21:04 | <ogra_> Enslaver, because they take it, modify and sell it
| |
| 21:04 | dsugar100 has left IRC (dsugar100!~dsugar@columbia.tresys.com, Quit: dsugar100) | |
| 21:04 | <ogra_> Enslaver, canonnical doesnt do such a thing
| |
| 21:04 | <vagrantc> there were ancient patches for swap over NFS that never got integrated into the kernel ... for good reason.
| |
| 21:04 | <Enslaver> I don't remember every having to buy an OS from redhat
| |
| 21:04 | <ogra_> so it is 100% depending on the community steering the distro
| |
| 21:05 | well, you buy licenses for an enterprise product they make
| |
| 21:05 | <Enslaver> you can buy support licenses
| |
| 21:05 | * vagrantc isn't so sure about 100% ... | |
| 21:05 | <jammcq> heh
| |
| 21:05 | <Enslaver> which also buys access to use rhn
| |
| 21:05 | <alkisg> https://bugzilla.novell.com/show_bug.cgi?id=678472 "Swap over NFS deadlocks under memory pressure" ==> fix committed... /me is wondering if that ended up upstream...
| |
| 21:06 | <ogra_> vagrantc, ?
| |
| 21:06 | <Enslaver> ogra_: he's saying money = power
| |
| 21:06 | <ogra_> vagrantc,, ubuntu is governed by the technical board and community council ... while canonical has some seats in both, its the minority
| |
| 21:07 | <Enslaver> they might not steer it, but i'm sure they can navigate it
| |
| 21:07 | <vagrantc> ogra_: i'm not directly experienced in the issue, but i've seen a number of folks leaving ubuntu lately because of how much sway canonical has...
| |
| 21:08 | <Enslaver> like how everyone left RH
| |
| 21:08 | right war war?
| |
| 21:08 | <ogra_> vagrantc, right, because rickspencer wrote a werid mail saying hey, well switch to rolling releases ... and people took that as an order from canonical
| |
| 21:08 | <Enslaver> he's probably on the beach
| |
| 21:08 | <jammcq> I think part of that is perceived sway
| |
| 21:09 | <alkisg> So if the community didn't like unity, ubuntu wouldn't have unity? /me doubts that...
| |
| 21:09 | <ogra_> vagrantc, the result of all this was that the TB cut the support cycle in half for the non LTS releases ... and there is something similar to a unstable symlink on archive.u.c now
| |
| 21:09 | vagrantc, what you saw was mainly overreaction and utter nonsense
| |
| 21:09 | <vagrantc> ogra_: it's not that one thing ... but i'm not personally invested either way.
| |
| 21:09 | * jammcq stopped being pissed at Ubuntu when he figured out that they did exactly what they said they'd do: Linux for Human Beings | |
| 21:10 | <jammcq> and most of us just aren't normal human beings
| |
| 21:10 | <ogra_> well, there was the rollong release thing and Mir ...
| |
| 21:10 | i cant remember anythign else recently
| |
| 21:10 | and Mir is a replacement for SurfaceFlinger atm, nothing more
| |
| 21:11 | * vagrantc can't keep track of all the things people grumble about, legitimately or not | |
| 21:11 | <ogra_> heh
| |
| 21:11 | <jammcq> it's not a replacement for Xorg ?
| |
| 21:11 | <ogra_> long term it is ...
| |
| 21:11 | by 14,04 or so
| |
| 21:11 | <jammcq> there's plenty of choices out there, and still plenty of reason to love whichever distro you love
| |
| 21:12 | <ogra_> yeah
| |
| 21:12 | * vagrantc tips a hat to jammcq | |
| 21:12 | * jammcq is re-kindling his old love ov debian | |
| 21:12 | <jammcq> s/ov/of/
| |
| 21:12 | <vagrantc> distro fanaticism is boring :)
| |
| 21:13 | * jammcq thinks ubuntu will be great for the masses, while debian is great for the geeks | |
| 21:13 | <ogra_> vagrantc, well, you wouldnt want to have to package rpms either, do you
| |
| 21:14 | <jammcq> I just wish I could get my dual 30" displays to play well with some distro
| |
| 21:14 | <ogra_> probably not fanaticim ... but we all swing one way or the other
| |
| 21:14 | <jammcq> it's not a happy experience with ubuntu atm
| |
| 21:14 | <ogra_> whats your issue ? my triple head works out of the box here
| |
| 21:15 | with and without binary drivers
| |
| 21:15 | any error message ?
| |
| 21:16 | <alkisg> Safe swapping over NFS and NBD in kernel 3.6: http://kernelnewbies.org/Linux_3.6#head-90a9988ae90f3481e4e6ddcae356bb8c614b531e
| |
| 21:16 | <jammcq> I think i need a geForce card, cuz my ati and quattro cards suck
| |
| 21:16 | 30" monitors are a different beast cuz they need either displayPport or dual-link dvi
| |
| 21:16 | and 2 of them means dual-dual-link
| |
| 21:16 | <ogra_> well i have a radeom 6990 and a gtx 650 or so
| |
| 21:17 | <vagrantc> alkisg: crazy!
| |
| 21:17 | <ogra_> alkisg, i did swapping over nbd ages ago ... works really nice if your swapfile lives in a tempfs in ram on the server ;)
| |
| 21:18 | jammcq, heh, but thats a challenge for every distro i fear
| |
| 21:19 | <jammcq> yeah, but it works flawlessly on OSX
| |
| 21:19 | <ogra_> though all of them support xorg.conf still
| |
| 21:19 | <jammcq> i'm just gonna run that
| |
| 21:19 | <ogra_> hah
| |
| 21:19 | following miguel ?
| |
| 21:19 | <jammcq> and it works in Win-7, but i'm not gonna run that either
| |
| 21:19 | heh
| |
| 21:19 | he makes a good point
| |
| 21:19 | <ogra_> he surely does
| |
| 21:20 | dead_inside has joined IRC (dead_inside!~dead_insi@76.75.3.174) | |
| 21:21 | <ogra_> but seeing nvidia using unity on stage at their presentation and not OSX or Win8 says something as well i'd say
| |
| 21:22 | komunista has left IRC (komunista!~slavko@87.244.209.121, Quit: Leaving.) | |
| 21:22 | * alkisg waves | |
| 21:22 | alkisg has left IRC (alkisg!~alkisg@ubuntu/member/alkisg, Quit: Leaving.) | |
| 21:25 | <ogra_> (or that linux game sales on steam are cuttenly on their way to surpass games on mac after only a few months)
| |
| 21:29 | <vagrantc> meh.
| |
| 21:29 | debian's now in deep freeze and there were a few "important" fixes that i don't know if i can get in.
| |
| 21:30 | <ogra_> when is release due ?
| |
| 21:31 | ||cw has left IRC (||cw!~chris@phpgroupware/cw, Ping timeout: 256 seconds) | |
| 21:31 | <vagrantc> same as always
| |
| 21:31 | <ogra_> haha
| |
| 21:32 | <vagrantc> the bug count is getting pretty low now, though.
| |
| 21:32 | <ogra_> nice, so shouldnt be long
| |
| 21:33 | <vagrantc> need to open a couple for libnss-sshsock and libpam-sshauth, since we don't really want to support those versions for a whole release cycle...
| |
| 21:36 | * ogra_ is sadly doing more android than ubuntu work lately | |
| 21:37 | <ogra_> and its no fun at all ...
| |
| 21:38 | handling tens of gigabyte big branches just to get a 50M image out on the rear end
| |
| 21:48 | simpoir has left IRC (simpoir!~simpoir@209.141.57.61, Ping timeout: 252 seconds) | |
| 21:48 | simpoir has joined IRC (simpoir!~simpoir@209.141.57.61) | |
| 21:56 | bobby_C has joined IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at) | |
| 22:02 | leio has left IRC (leio!~leio@gentoo/developer/leio, Read error: Operation timed out) | |
| 22:02 | gvy has left IRC (gvy!~mike@altlinux/developer/mike, Read error: Operation timed out) | |
| 22:03 | gvy has joined IRC (gvy!~mike@altlinux/developer/mike) | |
| 22:04 | leio has joined IRC (leio!~leio@gentoo/developer/leio) | |
| 22:07 | dead_inside has left IRC (dead_inside!~dead_insi@76.75.3.174, Quit: Leaving...) | |
| 22:48 | bobby_C has left IRC (bobby_C!~bobby@85-124-22-227.teleworker.xdsl-line.inode.at, Ping timeout: 245 seconds) | |
| 22:50 | jammcq has left IRC (jammcq!~jam@c-69-245-75-255.hsd1.mi.comcast.net, Quit: leaving) | |
| 22:55 | <Enslaver> so i got a question, when building the client i can specific the name of the chroot by doing ltsp-build-client --chroot, but how do i specify what chroot to use when doing ltsp-chroot ?
| |
| 22:55 | Also, in the ltsp-chroot code it says it reads from a config file /etc/ltsp/ltsp-chroot.conf but where in the code does it look at that file?
| |
| 22:57 | <ogra_> vagrantc, ugh ... http://lists.freedesktop.org/archives/systemd-devel/2013-March/009797.html ... what will debian do ?
| |
| 22:58 | no more dbus without systemd sound pretty bad
| |
| 22:58 | *sounds
| |
| 23:04 | ||cw has joined IRC (||cw!~chris@68.187.0.66) | |
| 23:06 | <vagrantc> ogra_: something
| |
| 23:06 | <ogra_> heh
| |
| 23:07 | well, i guess it wont itch you much for the next few months ...
| |
| 23:08 | but no more dbus upstream means its effectively dead unless your distro switches the init system
| |
| 23:10 | <vagrantc> ubuntu going systemd?
| |
| 23:11 | <ogra_> well, for parts ...
| |
| 23:11 | <vagrantc> Enslaver: the configuration file handling for most tools should now be in common functions
| |
| 23:11 | <ogra_> consolekit is dead already ... so you need logind
| |
| 23:11 | udev is dead
| |
| 23:11 | * vagrantc blinks | |
| 23:11 | <warren> systemd replaces all that?
| |
| 23:11 | <ogra_> so you need to build it from the systemd source
| |
| 23:12 | warren, yeah, i'm waiting for it to grow an input system and a display server soon
| |
| 23:12 | <warren> ogra_: what? no device drivers too?
| |
| 23:12 | <Enslaver> I'm thinking systemd is whats coming in EL7 :(
| |
| 23:12 | <ogra_> so in ubuntu we have the systemd source package but build only some individual tools to keep the pieces working
| |
| 23:13 | <warren> Enslaver: already need to support it if you support Fedora 18+
| |
| 23:13 | <Enslaver> warren: Yeah i saw, not looking forward to that
| |
| 23:13 | <warren> is upstart getting lots of attention?
| |
| 23:13 | <ogra_> well, as long as canonical is around it will ... its less and less thanks to lennart
| |
| 23:13 | <warren> Enslaver: at least most of the old tools work as passthrough
| |
| 23:13 | <vagrantc> now that systemd is required for nearly everything, i wonder how long it'll take for lennart to switch to something else
| |
| 23:13 | <ogra_> upstart is what chromeos uses
| |
| 23:14 | <Enslaver> <3 chromeos
| |
| 23:14 | <ogra_> and i would suspect android too soon ... with the recent movements inside google
| |
| 23:14 | <warren> google can do whatever the hell they want, the user experience is abstracted from all that
| |
| 23:14 | <ogra_> yeah
| |
| 23:14 | <Enslaver> chromeos is soo pretty, after i use it i just want to go in the bathroom, light some candles and touch myself
| |
| 23:14 | <ogra_> vagrantc, ++
| |
| 23:15 | <warren> I'm still a lot more pissed off about pulseaudio than systemd, personally
| |
| 23:15 | <ogra_> wayland already has its systemd equivalent in Mir now, so he wont do a display server next
| |
| 23:15 | <Enslaver> they realized pulse audio is lame and are going back to alas? yay
| |
| 23:16 | alsa*
| |
| 23:16 | <ogra_> warren, pulse got great once lennart left
| |
| 23:16 | <vagrantc> it's amazing how many technologies i've never had to bother with by simply working with a distro that releases once every two years...
| |
| 23:16 | <ogra_> its a fairly good thing nowadays
| |
| 23:16 | haha
| |
| 23:16 | <warren> oh, then I should hope lennart leaves systemd?
| |
| 23:16 | <Enslaver> vagrantc: debian only releases once every two years?
| |
| 23:16 | <vagrantc> Enslaver: approximately
| |
| 23:17 | <Enslaver> vagrantc, hmmm, maybe i'll give it another look
| |
| 23:17 | <ogra_> warren, depends what he turns to, what vagrantc said :)
| |
| 23:17 | * warren hopes Windows Phone. Please please. | |
| 23:17 | <vagrantc> Enslaver: it's not on a predictible release cycle
| |
| 23:17 | <ogra_> LOL
| |
| 23:17 | <warren> vagrantc: are Debian releases modern by the time they happen?
| |
| 23:18 | * warren has never used Debian or Ubuntu ever. | |
| 23:18 | <Enslaver> I tried the latest ubuntu, wouldn't boot on any of my systems
| |
| 23:18 | <ogra_> debian has a pretty good backports story
| |
| 23:18 | just got better recently even
| |
| 23:19 | but its pretty outdated at release day if you look at upstreams
| |
| 23:19 | <vagrantc> https://wiki.debian.org/DebianReleases#Release_statistics
| |
| 23:19 | <Enslaver> debian still interests me, also checked out suse, it's got a nice model
| |
| 23:20 | <ogra_> ubuntu is pretty recent ... but for your mom ...
| |
| 23:20 | <warren> ChromeOS
| |
| 23:20 | <vagrantc> if you honestly can't bear to use software that's a year or two old, i question the quality of the software in the first place :)
| |
| 23:20 | <ogra_> and indeed not as well tested as a distro that takes 2 years of QA
| |
| 23:21 | <warren> I enjoy seeing twice a year if they managed to seize making GNOME worse.
| |
| 23:21 | <vagrantc> GNOME does a great job of that
| |
| 23:21 | <ogra_> heh, yeah ...
| |
| 23:23 | <Enslaver> is chromeos open source? i kinda want to look at the kernel it uses
| |
| 23:23 | * ogra_ had to explain to someone today that you can only change the font size in the a11y settings nowadays ... and it only offers small/normal/large now | |
| 23:24 | <warren> that's all you need!
| |
| 23:24 | <ogra_> Enslaver, the underlying layer is full OSS ... on top it runs chrome in Xorg
| |
| 23:25 | <Enslaver> yeah i have one with developer mode enabled but wheres the source? code.google.com probably?
| |
| 23:25 | <ogra_> and chrome as well as the chormeos desktop pieces arent open afaik
| |
| 23:25 | yeah
| |
| 23:25 | <Enslaver> k ill check
| |
| 23:26 | <ogra_> http://www.chromium.org/chromium-os/developer-guide
| |
| 23:27 | its kind of like gentoo ... with a bit debian and ubuntu mixed in
| |
| 23:28 | <Enslaver> nah, its more fedora and altlinux with a hint of slackware
| |
| 23:29 | <ogra_> well, it uses the gentoo build scripts, and is fully based on upstart (they even hired the ubuntu upstart maintainer for that)
| |
| 23:29 | the rest comes from debian sources afaik
| |
| 23:30 | its a pretty weird hybrid
| |
| 23:30 | but it works :)
| |
| 23:31 | <Enslaver> nah its all based on a kernel built by a guy name linus, purchased by SCO and sold to red hat
| |
| 23:31 | therefore, its slackware
| |
| 23:31 | <ogra_> then it would use a strange tarball based package management system
| |
| 23:31 | <Enslaver> slackware invented tar :P
| |
| 23:32 | well, them and MIT
| |
| 23:32 | <ogra_> :)
| |
| 23:32 | khildin has left IRC (khildin!~khildin@ip-80-236-227-61.dsl.scarlet.be, Quit: I'm gone, bye bye) | |
| 23:34 | highvoltage has left IRC (highvoltage!~highvolta@ubuntu/member/highvoltage, Ping timeout: 252 seconds) | |