i-R2D247 (~deed@tor-irc.dnsbl.oftc.net) left irc: Ping timeout: 480 seconds
i-R2D247 (~deed@tor-irc.dnsbl.oftc.net) joined #fai.
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
MT (~MT@dove.informatik.tu-muenchen.de) left irc: Remote host closed the connection
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
ErKa (keryell@m5.wifi.enstb.org) joined #fai.
looperonfire_ (~steffen@141.18.9.101) left irc: Remote host closed the connection
ErKa (keryell@m5.wifi.enstb.org) left irc: Ping timeout: 480 seconds
ErKa (keryell@m5.wifi.enstb.org) joined #fai.
Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai.
rcy (~rcy@209.90.176.221) left irc: Ping timeout: 480 seconds
08:13 stockholmhuhu!
Action: stockholm hat netzwerk in spanien
rcy (~rcy@209.90.176.221) joined #fai.
sep (~sep@40.211.jostedal.no) left irc: Remote host closed the connection
sep (~sep@89.130.6.136) joined #fai.
Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Quit: leaving
Mrfai (~lange@suenner.informatik.uni-koeln.de) joined #fai.
Nick change: MT -> Guest2034
Guest2034 (~MT@dove.informatik.tu-muenchen.de) left irc: Read error: Connection reset by peer
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
09:57 stockholmMT: hast du was an shdd2 getan seit neulich?
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
10:07 sobersabrehi!
10:08 sobersabreguys, I am trying to configure NIS on install. I understand I need 2 things to do: 1. install and configure package nis with debconf
10:08 sobersabre2. modify system files ( /etc/nsswitch.conf, /etc/passwd, /etc/shadow, /etc/group, /etc/gshadow )
10:09 sobersabreis there a "ready to read" part of documentation for this ?
10:11 h01gerstockholm, deine frage laesst fragen offen...
10:11 h01gerneulich? +bischn unspezifisch
10:15 Mrfaisobersabre: http://paste.debian.net/40039
10:16 MrfaiThis is a script that I use for setting up NIS
10:16 eartoastmoin
10:21 sobersabreMrfai: you're a life saver. I'll need to edit it a bit, but it's completely clear!
10:21 sobersabreI thought to mess with debconf
10:21 sobersabre:)
10:22 h01gerMrfai is a life saver by having written fai anyway :)
10:23 sobersabreh01ger: I know, but still. it
10:23 sobersabre's great to have help like this.
10:23 h01gersure
10:23 h01geri just wanted to make that point
10:30 stockholmMT: das "zeroing out" worked, i could install and boot with root on lvm
10:30 stockholmMT: i need to check in the syntax error fix though
10:32 stockholmMT: i think if the "preserving" works like this it is overkill to do the long way around with detecting existing volume groups
10:33 sobersabreMrfai: where do I define the variable YPDOMAIN ( I could do it inside the script itself... maybe there's th proper way for it )
10:33 stockholmMT: i think there is no added value doing it the other way, really
10:34 MrfaiYPDOMAIN can be defined in dhcpd.conf or in class/*.var
10:34 sobersabreOK! class/*.var is the best
kugg (~jerkeby@89.130.6.136) joined #fai.
kugg (~jerkeby@89.130.6.136) left irc: Read error: Connection reset by peer
12:09 sobersabreanyone in here ?
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) left irc: Quit: Lost terminal
ErKa (keryell@m5.wifi.enstb.org) left irc: Ping timeout: 480 seconds
MT (~MT@dove.informatik.tu-muenchen.de) left irc: Remote host closed the connection
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
13:17 MTsorry stockholm, wasn't at my computer till now
13:17 MTah, you already committed something
kugg (~jerkeby@89.130.6.136) joined #fai.
ErKa (keryell@m5.wifi.enstb.org) joined #fai.
14:25 stockholmMT: i have something else to commit without dd
14:25 stockholmthat works even
14:26 stockholmMT: please check it out
14:26 stockholmMT: just committed
sep (~sep@89.130.6.136) left irc: Ping timeout: 480 seconds
14:28 stockholmMT: now can you tell me... will it work to set a :preserv tag on a partition or a volume group or so and preserve all related items?
sep (~sep@89.130.6.136) joined #fai.
14:29 stockholmwhere IS sep?
ErKa (keryell@m5.wifi.enstb.org) left irc: Ping timeout: 480 seconds
14:39 stockholmMrfai: we think that it would make sense to have an LVM class
14:39 stockholmMrfai: would it be ok to commit that to fai head?
14:40 h01gerstockholm, file a bug with patch first?
14:43 Mrfaia wishlist bug please. I do not understand why FAI needs an LVM class.
14:47 stockholmMrfai: the lvm class would contain for example the dependency on lvm2
14:54 MrfaiThis is up to every user. FAI cannot define classes for all environments. a LVM class is an example and should g ointo the wiki
14:59 MTstockholm?
14:59 MTDu hast u.a. folgendes committed
14:59 MT-      $val *= ( 1 / 1024 );                                                                                                                
14:59 MT+      $val = $1 ( 1 / 1024 );                
14:59 MTda fehlt wohl ein *
15:05 h01gerMrfai, i do think that LVM is used at >80% of all installations today... thats not a corner case
15:11 lazyb0yMrfai: die slac im winter ist erst die zweite?
15:14 stockholmMT: ja, stimmt
15:15 lazyb0yMrfai: if you want more examples, you must take some into the simple examples...
15:15 MTstockholm, ich bin mir nicht sicher ob das preserve sich so fortsetzt
15:15 MTich glaube, das könnte unvollständig sein an der Stelle
15:16 lazyb0yif there's always many reasons not to take examples into the fai-doc package (cornercase, nobody maintains them, ...) then no additional example will get in
15:17 stockholmlazyb0y: perhaps that is the point?
15:17 lazyb0ystockholm: what?
15:17 stockholmMrfai: can you elaborate?
15:17 stockholmlazyb0y: to have less examples?
15:18 stockholmlazyb0y: to have a as small as possible starting config?
15:18 lazyb0ystockholm: hmm, no, aehm, I don't want to offend thomas now, but it's the same person saying we could use some more examples :)
15:18 lazyb0yand not taking in proposed ones :)
15:18 stockholmah, i get it
Action: h01ger still thinks fai could use a default configspace instead of an example one...
15:20 stockholmah, i guess you had this discussion before :-)
15:20 h01gerand instead of letting them rot in a wiki, i'd rather put them in my branch ;)
15:21 lazyb0yh01ger: hmm, I guess the problem is, even the very simple examples(or, then: default) has quite a lot of files, which then are all config files for dpkg?!
ErKa (keryell@keryell.pck.nerim.net) joined #fai.
15:23 h01gerlazyb0y, nope. conffiles are not allowed to be edited. but its fine to have files in /etc which are not conffiles
15:24 h01ger(and debian-edu needs an out of the box working fai configspace too, btw. and best its installed as a package)
i-R2D247 (~deed@tor-irc.dnsbl.oftc.net) left irc: Remote host closed the connection
i-R2D247 (~deed@tor-irc.dnsbl.oftc.net) joined #fai.
15:30 stockholmMT: with the pvremove i consider the problem solved in the proper way and it even works.
15:31 MTbut we can't preserve any logical volumes, right?
15:31 stockholmwe cant?
15:31 stockholmwhy not?
15:31 MTdon't you pvremove anything you find?
15:32 stockholmi only remove devices that should be created anyway
15:32 MTah
15:32 MTok
15:32 stockholmi only pvremove on devices that should be created anyway
15:33 stockholmplease check the code if i misunderstood it this correctly
15:34 stockholmwow, that sentence is broken
15:35 MTI will definitely look at all of the code at the dev weekend
15:36 stockholmit is not much code :-)
Barbarossa (~max@rfc2324.org) joined #fai.
15:43 Barbarossagun amnd
15:43 BarbarossaWie funktioniert denn das partitionieren in detail?
15:44 BarbarossaIch hab bisher rausgefunden, dass es nen hook partiotion.DEFAULT (o.ae.) gibt, der dann setup_harddisks aufruft
15:44 BarbarossaAber wer sucht sich das Partitionsschema aus?
15:49 BarbarossaAh Source lesen bildet
15:50 BarbarossaScheint nen Setup-Problem $hier zu sein
15:52 stockholmBarbarossa: huhu!
15:52 BarbarossaHi Andrea
15:52 Barbarossas
15:53 Mrfailazyb0y: ja ist erst die zweite slac
15:53 stockholmslac?
15:55 Mrfaione reason not to take a LVM class right now, is that the new partition tool is too new and not yet included into FAI.
15:57 BarbarossaMrfai: Can I tell FAI in the disk_config file to create an PV or a VG?
ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds
16:00 MTBarbarossa, take a look at http://faiwiki.debian.net/index.php/Setup_harddisks_2
16:00 MrfaiBarbarossa: for raid and lvm use the new partition tool. And ask mt and stockholm. there are mor familiar with it
16:00 Mrfais/there/they/
16:02 lazyb0yMrfai: that's a good and acceptable reason, in this case :)
16:02 lazyb0yI agree, no classes are needed that depend on features not yet committed into the trunk
16:02 MTlazyb0y, Mrfai, what about creating further examples in some branch
16:03 MrfaiMT: we should discuss that in Essen
16:03 lazyb0yif essen happens
16:03 lazyb0y(I must write a mail to the linuxhotel these days, asking if our group is too small)
16:03 lazyb0yif it's O.K. for them, we should do it
16:04 MTyes, otherwise we could also meet up somewhere else
16:04 Mrfaido they have a minimum of participants?
16:04 MTguys, I'll leave to catch the underground, I'll be back in an hour
16:06 lazyb0yI don't know, maybe they have other things that could happen there...
16:06 lazyb0yI don't believe they will send us away - otherwise we could revoke their license :)
16:06 Mrfai:-)
jair_ (~jair@adsl-64-108-201-216.dsl.chcgil.ameritech.net) joined #fai.
jair (~jair@adsl-64-108-201-216.dsl.chcgil.ameritech.net) got netsplit.
kriebly (~moho@wisdom.Stanford.EDU) got netsplit.
MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds
16:16 stockholmmt: we should consider changing the configuration parser to accept lists in lines seperated with any of the following: " ", "," and ", "
16:16 stockholmof course you could comment easier if you were here
16:20 h01gerlazyb0y, _if_ essen happens? uh. i'm considering not going to the next edu meeting and instead go to fai. if essen doesnt happens, it like to go to edu. when will you know`
16:20 h01ger?
kriebly (~moho@wisdom.Stanford.EDU) got lost in the net-split.
jair (~jair@adsl-64-108-201-216.dsl.chcgil.ameritech.net) got lost in the net-split.
16:24 stockholmMrfai: do you have a checklist of what needs to happen and what needs to be tested before shdd2 will get merged?
16:26 lazyb0yh01ger: the probability is wuite low, as can be read above
16:26 lazyb0y_very_ low
16:26 h01gerlazyb0y, ic
16:26 lazyb0yI just thought abput the possibility of this, and in private mail, thomas and i dicussed if we should do it at all
16:26 MrfaiI will do a look at the source code, and then I think talking to mt face to face should do the rest. I hope do do this in Essen
16:27 lazyb0ybut we decided we want not cancel it
16:29 h01gerlazyb0y, so you tell me now, it will very probably not take place, but you still dont want to cancel it now, but rather lastminute?
16:37 stockholmMrfai: btw, you could extend ainsl with a -p flag that creates missing directories
16:37 stockholmlike this:
16:37 stockholmmkdir -p $target/etc/initramfs-tools/
16:37 stockholmainsl -a $target/etc/initramfs-tools/modules dm_mod
16:40 Mrfaistockholm: yes. Please send me a patch or file a wishlist bug.
16:40 lazyb0yh01ger: no, I tell you it will very probably not be cancelled :)
16:41 lazyb0ythe only possibility are the linuxhotel people, but we have measures of pressure against them, as seen above
16:42 stockholmi guess a patch is easier
16:43 h01gerlazyb0y, then i got you wrong
Action: Mrfai .o0( make the BTS much more complicated, so people will send more patches) ;-)
16:43 h01ger.oO( never send patches to Mrfai. always send them to the bts )
16:43 h01gerwe also want to have fun
16:44 lazyb0yMrfai: if you discourage sending a bugreport without a patch, you discourage people of telling you what goes wrong, even if they don't know how to fix it
16:45 Mrfailazyb0y: it was a joke to stockholm. He prefers sending patches only because sending a bug report is more work.
16:46 Mrfaieartoast: does your fai-classes-config also detects classes used in script after the command ifclass?
16:47 lazyb0yMrfai: ah, I got you wrong - sorry
Action: lazyb0y should read context
16:47 Mrfailazyb0y: no roblem
16:47 Mrfailazyb0y: here's the missing 'p'
16:47 stockholmMrfai: if i have to use your perl coding style for writing the patch i wont do it
16:48 stockholmMrfai: can i do proper coding style, too? :-)
16:48 Mrfaistockholm: try to use my style. If I apply a patch, I will also reformat and recode it, if it does not match my style
16:55 stockholmMrfai: i will switch to use strict; regardless :-)
mfl (~mfl@adonis.fnb.maschinenbau.tu-darmstadt.de) joined #fai.
allee (~ach@allee.mpe.mpg.de) left irc: Ping timeout: 480 seconds
CXO_Tobi (~therziger@195.122.130.2) left irc: Ping timeout: 480 seconds
Mrfai (~lange@suenner.informatik.uni-koeln.de) left irc: Quit: leaving
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
17:53 sobersabrehi
17:54 sobersabreI am trying to install a client with 2 disks.
17:54 sobersabreand I am getting an error stating: no config file for setup_harddisk found.
17:55 sobersabreI indeed don't have such folder under config.
17:55 sobersabreQ1: why could I install other machines with 2 hard disks.
17:55 sobersabre?
17:55 sobersabre(it was /dev/hda and /dev/sda )
17:55 sobersabrenow it is /dev/hda and /dev/hdb
17:55 sobersabreany ideas ?
17:56 sobersabreand, I do have this disk_config folder, with supposedly good configurations.
17:56 sobersabreplease don't ask me to pastebin anything, I am ssh'ing from this client, to the machine with internet.  
17:57 sobersabreI can peek the config of course.
Lin (~igor@200.179.57.57) joined #fai.
18:01 sobersabrehmmm. ok. I reboot.
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) left irc: Quit: leaving
18:03 oz_what is the current fai kernel?
18:04 oz_2.6.18?
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
kriebly (~moho@wisdom.Stanford.EDU) joined #fai.
18:05 sobersabrehi. again.
18:05 sobersabrereducing the size of auto resizable partition didn't work.
18:05 sobersabreis anybody alive ?
18:09 sobersabrenoone. ok. I'
18:09 sobersabrem off home, good luck all.
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) left irc: Quit: leaving
18:11 Robinayor_.oO(...)
18:12 Robinayor_are we all dead now?
18:16 oz_no
Action: oz_ working hard
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
18:23 sobersabrehi back again.
18:23 sobersabre:(
18:24 sobersabreplease take a look at config
18:24 sobersabredisk:
18:24 sobersabrehttp://pastebin.ca/741435
18:25 sobersabre1st disk has 1st partition ntfs. I wrote: preserve partition 1
18:25 sobersabredidn't give it any mounting points.
18:26 oz_the ATI SB600, is it known to anyone?
18:26 sobersabrewhat is it ?
18:37 eartoast18:46 < Mrfai> eartoast: does your fai-classes-config also detects classes used in script after the command ifclass?
18:37 eartoastMrfai: yes
18:40 sobersabremust be off, bye all.
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) left irc: Quit: leaving
kugg (~jerkeby@89.130.6.136) left irc: Ping timeout: 480 seconds
ErKa (keryell@keryell.pck.nerim.net) joined #fai.
ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds
ErKa (keryell@keryell.pck.nerim.net) joined #fai.
MT (~MT@dove.informatik.tu-muenchen.de) joined #fai.
20:34 stockholmMT: das ganze partitionieren ist etwas riskant wegen der instabilen/asynchroen device detection
20:34 MTinwiefern?
20:35 MTalso wenn die Platten unterschiedlich sind, ist das natürlich richtig
20:35 stockholmwenn man zwei scsi controler hat mit disks dran kann es sein dass die reihenfolge in der die treiber geladen sind sich ändert
20:36 BarbarossaDoes anyone have a Xen Dom-U running with vanilla Kernel 2.6.23.1?
20:36 BarbarossaMy Dom-U freezes after 'xm create <conffile>'
20:36 stockholmoder dass die reihenfolge in der die kontroller die platten entdecken von boot zu boot variiert
20:37 MTja, aber die UUIDs sollten das ja beheben, jedenfalls wenn das System mal läuft
20:37 stockholmMT: aber du kannst keine uuids in der shdd2 config verwenden
20:37 stockholmdas skaliert nicht für viele rechner
20:38 MTja, und ich glaub auch gar nicht, dass eine Platte eine UUID hat
20:38 stockholmMT: das wird interessant wenn du "preserve" benutzen willst, z.b.
20:38 stockholmMT: und in software raid können beide platten (raid1) die gleiche uuid haben
20:39 MTich hab keine Ahnung, wie da eine Lösung dafür aussehen könnte?
20:40 MTalso ich hab hier das Problem immer dadurch gelöst, dass ich die externen Platten (die ich keinesfalls partitionieren möchte :-) ) abgesteckt hab
20:43 stockholmMT: ja, das ist die eine lösung die aber für fai und grosse mengen disks und rechner nicht funktioniert
20:44 MTschon klar, aber ich habe einfach keine Ahnung, wie so eine Lösung aussehen könnte
20:44 MTich meine, was ist die Eigenschaft, die eine Platte eindeutig identifiziert?
20:44 MTaber auch nur so eindeutig, dass die Konfiguration trotzdem noch auf vielen Rechnern passt
20:45 stockholmgenau
20:45 stockholmgute frage
20:45 stockholmich habe mal auf #linuxfs gefragt. da hängen die kernel hacker rum die das verursachen :-)
Lin (~igor@200.179.57.57) left irc: Quit: Ex-Chat
sep (~sep@89.130.6.136) left irc: Quit: Leaving
21:05 h01gerseriennummer. eartoast erzaehlte mir, dass bei ihm im institut die rechner die platten relativ gleiche (="wildcardbare") seriennummern haben... dazu einenn kleinen hack geschrieben und voila :)
21:06 stockholmund damit kannst du dann disk1 und disk2 unterscheiden?
mfl_ (~mfl@adonis.fnb.maschinenbau.tu-darmstadt.de) joined #fai.
mfl (~mfl@adonis.fnb.maschinenbau.tu-darmstadt.de) left irc: Write error: connection closed
21:26 MTwie kommt man denn an die Seriennummerß
21:26 MT?
21:26 MTich meine, das könnte man schon reinbauen
21:26 MTmeinetwegen sogar mit regexp
21:27 MTalso statt disk1 oder so dann einfach SER:nummer
21:39 h01gerdas skaliert nich
21:39 h01geralso weil es sehr selten ist, dass du 300 maschinen mit relativ gleichen seriennummern hast
21:40 MTvorhin meintest Du aber doch gerade noch, dass eartoast das Gegenteil behauptet hätte :-)
21:40 MTandererseits auch: was, außer Seriennummern, bleibt zur Identifikation einer Platte?
21:47 h01gerMT, ja. eartoast institut hat ne reihe von rechnern mit derselben plattenseriennummer. aber ich behaupte das ist sehr selten
21:47 h01gerMT, plattentyp/plattengroesse
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
21:48 sobersabrehi.
21:51 sobersabreh01ger, what's up ?
21:51 sobersabreWie geht's ?
21:53 h01gerfain
21:53 h01gery tu?
21:54 sobersabretudo bem, obrigado.
21:55 sobersabreexept what's not tudo bem.
21:55 sobersabre;)
21:57 sobersabreI've just read the guide, part 7.7
21:58 sobersabreand I wonder: is there some more material on this subject of harddisk setups
21:58 sobersabre?
22:01 h01gerprobably ;) dunno really. you _might_ want to look at setup_harddisc2 in the faiwiki
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) left irc: Quit: Leaving
22:01 h01gernarf :)
sobersabre (~bilbo@89-138-78-195.bb.netvision.net.il) joined #fai.
22:02 sobersabresorry: disconnected.
22:02 h01gerprobably ;) dunno really. you _might_ want to look at setup_harddisc2 in the faiwiki
22:02 h01gersobersabre, re: your question..
22:03 sobersabreoh, thanks1
22:05 sobersabreh01ger, I haven't found this string.
22:11 MTsobersabre, http://faiwiki.debian.net/index.php/Setup_harddisks_2 is the way to go
22:12 sobersabreI tried to search "hard disk"
22:13 sobersabrethanks for the link. I think there's something wrong with the search.
22:13 h01ger"recent changes" works well to ;)
22:14 sobersabreh01ger, hm... ok.
22:14 h01ger(i was mostly joking. look at recent changes in the wiki to understand why..)
22:16 sobersabreoh, is this you who makes all the changes ?
22:16 sobersabrewith this page ?
ErKa (keryell@keryell.pck.nerim.net) left irc: Ping timeout: 480 seconds
MT (~MT@dove.informatik.tu-muenchen.de) left irc: Ping timeout: 480 seconds
--- Fri Oct 19 2007

Generated by irclog2html.pl 2.1_h01ger1 by Jeff Waugh - find it at freshmeat.net! Customized by Holger 'h01ger' Levsen