1161390459 J * mire ~mire@7-167-222-85.adsl.verat.net
1161390922 M * Bertl welcome mire!
1161391224 M * Hollow Bertl: http://linux-vserver.org/Related_Projects
1161391274 M * Hollow or do you mean sth different?
1161391524 M * Bertl I'm currently in progress of creating a new page labeled Mainline Kernel Virtualization
1161391528 M * Bertl From Linux-VServer
1161391530 M * Bertl Jump to: navigation, search
1161391548 M * Bertl hmm, copy and paste is strange with this wiki too :)
1161391563 M * Hollow ah, ok...
1161391970 M * Hollow Bertl: did the old wiki have some kind of anti-spam regex?
1161392002 Q * meandtheshell Quit: exit (0);
1161392195 M * Bertl Hollow: yep, I sent you that with the config/modifications
1161392277 M * Hollow Bertl: ok, will have a look, it seems there are some pre-made blacklists around too in the mediawiki metawiki
1161392295 M * Bertl yeah, can't hurt to combine that
1161392320 M * Bertl also an ip blacklist proved very helpful in certain cases :)
1161392501 M * Hollow Bertl: http://blacklist.chongqed.org/mediawiki/ seems quite "complete"
1161392511 M * Bertl perfect
1161392622 M * Hollow ok, extension installed...
1161392643 M * Bertl great!
1161392660 M * Hollow i also read about an extension which asks users to type in a random string generated with an image as soon as URLs appear in the edit.. do you think that would make sense?
1161392717 M * Bertl you mean a captcha?
1161392754 M * Bertl I think that is probably overkill and just annoying for those who _do_ the editing
1161392773 M * Hollow yeah, that's what i thought too...
1161392786 M * Bertl well, unless it is only active iif the user is not logged in, of course
1161392818 M * Bertl I think all folks doing serious editing are logged in .. maybe I'm wrong there
1161392866 M * Bertl but keep the link, it might get worse :)
1161392924 M * Hollow yep ;)
1161392947 M * Hollow maybe one thing we could add now, is an open proxy blocker..
1161392966 M * Bertl for editi
1161392973 M * Bertl +ng
1161392980 M * Bertl ?
1161392985 M * Hollow yup
1161392997 A * Bertl has clumsy fingers when eating :)
1161393010 M * Hollow it is already in mediawiki and uses the SORBS DNSBL
1161393045 M * Bertl okay, unless somebdoy complains, that sounds fine to me, but it should state the fact somehow to the editor
1161393072 M * Hollow it will only block anon users though... if you have an account it will not check against the BL
1161393086 M * Bertl ah, okay, works for me then
1161393126 M * Hollow ok, enabled :)
1161393151 M * Hollow let's see how good these two systems work .. 
1161394042 J * hardwire` ~hardwire@89-208-58-66.gci.net
1161394042 Q * hardwire Read error: Connection reset by peer
1161395751 M * Radiance hey guys, perhaps some one here experienced the same, i noticed that soft interrupts are taking alot of cpu time ... is there a way to prevent this ? Perhaps preemption model adjustment in the kernel ? 
1161395972 M * Bertl hmm, you mean, in general or vserver specifically
1161396046 M * Bertl what I'm trying to figure here is: do you observe unusual high softirqs on a Linux-VServer system or is that soemthing you observe in general?
1161396082 M * Radiance in general 
1161396121 M * Radiance i observed this on 2 systems, 1 is a normal server and the other is running a vserver
1161396129 M * Radiance so i think what would be best to do in general
1161396130 M * Bertl okay, well, most soft irqs are the result of workload offloaded by hardirqs and/or kernel threads
1161396138 M * Radiance perhaps io scheduler change ?
1161396148 M * Bertl yes, can help in certain cases
1161396163 M * Radiance i suspect this is because of many network connections going on, serving many users
1161396163 M * Bertl for servers, in general the following 'things' apply
1161396190 M * Bertl - a lower HZ value increases overall performance and 'smoothness' for sacrificing a little 'interactivity'
1161396208 M * Bertl - preemption is generally a thing you want to avoid
1161396228 M * Radiance the HZ value is set to 250 btw
1161396229 M * Bertl - swap should better be damn fast or unused :)
1161396254 M * Radiance swap is hardly used, the i/o with disk is low
1161396265 M * Bertl - avoid heavy I/O like 'unfortunate' raid setups on I/O bound servers
1161396267 M * Radiance prempt is set to voluntarily
1161396276 M * Radiance no raid used here
1161396305 M * Bertl - use netapi for network bound servers (and remove unnecessary tables/checks/etc)
1161396311 M * Radiance but if i change the HZ value and drop the preempt...won't this harm services like postfix, apache, proxy ?
1161396340 M * Bertl nope, the only services which could show effects are interactive things like ssh or telnet
1161396357 M * Bertl we are talking about tiny fractions of seconds
1161396372 M * Bertl you never even notice that on a web page or mail exchange
1161396402 M * Radiance ok so an interactive shell over ssh could suffer from it ?
1161396416 M * Radiance but how about ssh tunnels which only carry traffic for both sides ?
1161396443 M * Bertl if there is high load going on, interactive services could be affected (i.e. you should test that)
1161396478 M * Radiance ok thanks mate
1161396481 M * Bertl network performance mostly depends on the network cards and how many checks/iptables/bridge sthingies you have there
1161396520 M * Bertl badly chosen network cards can have heavy influence, e.g. most Realtek are crap
1161396528 M * Radiance i'm going to test now by changing some options in the kernel
1161396530 M * Radiance yeah true heh
1161396563 M * Radiance btw, message signaling interrupts and irq balancing by the kernel ... what is your opinion about these to manage load in general ?
1161396581 M * Bertl MSI is nice when it works :)
1161396593 M * Bertl unfortunately it is buggy on many systems
1161396611 M * Bertl irq balancing is interesting if you have CPU bound loads
1161396632 M * Bertl if the load is I/O (i.e. irq) bound, then you want
1161396642 M * Bertl to do fixed assignments (if you ahve several cpus)
1161396662 M * Bertl e.g. nic1 <-> cpu1, nic2 <-> cpu2 ....
1161396682 M * Bertl or put all timer irqs to cpu1, and all disk irqs to cpu2 ...
1161396691 M * Radiance ah ok heh
1161396714 M * Radiance ok bud, i'm gonna recompile and adjust the hz and drop the prempt,see if that helps 
1161396714 M * Bertl also a working apic is worth a lot there
1161396746 M * Radiance yeah, worst case is usually the acpi code being buggy causing trouble
1161396768 M * Bertl okay, let me know how it goes 
1161396883 M * Radiance ok, recompiling as we speak
1161397550 M * Hollow http://linux-vserver.org/ProcFS has been updated to 2.1.1
1161397595 M * Bertl nice work!
1161397616 M * Hollow Bertl: could you confirm the explanation of the scheduler? and fill the missing two fields?
1161397664 M * Hollow sched_pc->{token,idle}_time that is
1161397673 M * Bertl will try to do so .. please bug me again if I forget ...
1161398161 M * Radiance BertI, i have to wait about 2 hours before i can reboot to clear users, but i'll msg you when it's done :)
1161398185 M * Bertl okay, great!
1161398511 J * Aiken ~james@tooax6-157.dialup.optusnet.com.au
1161398526 M * Bertl morning Aiken!
1161398588 M * Aiken good morning
1161398809 M * Hollow off to bed now.. cu tomorrow!
1161398863 M * Bertl night!
1161399295 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1161399304 J * ensc ~irc-ensc@p54B4F1E0.dip.t-dialin.net
1161399629 J * Aiken_ ~james@tooax6-157.dialup.optusnet.com.au
1161399630 Q * Aiken Read error: Connection reset by peer
1161400206 J * matled_ ~matled@85.131.246.184
1161400237 Q * matled Write error: connection closed
1161400237 N * matled_ matled
1161402528 Q * Johnnie Remote host closed the connection
1161402555 J * Johnnie ~jdlewis@jdlewis.org
1161402947 M * Bertl okay, I'm off to bed for today ... back tomorrow! have a good one everyone! cya!
1161402957 N * Bertl Bertl_zZ
1161403819 P * Aiken_ Leaving
1161409215 Q * _are_ Ping timeout: 480 seconds
1161415257 Q * Johnnie Ping timeout: 480 seconds
1161415434 J * Johnnie ~jdlewis@jdlewis.org
1161415492 Q * gdm Remote host closed the connection
1161415501 J * gdm ~gdm@www.iteration.org
1161415570 J * hardwire ~hardwire@89-208-58-66.gci.net
1161415570 Q * hardwire` Read error: Connection reset by peer
1161416560 Q * Johnnie Remote host closed the connection
1161416968 J * _are_ ~are@gprs-pool-1-014.eplus-online.de
1161417120 J * Johnnie ~jdlewis@jdlewis.org
1161417996 J * dna_ ~naucki@177-239-dsl.kielnet.net
1161419974 M * yang derjohn: ?
1161420528 Q * Johnnie Quit: G'bye!
1161421617 J * Johnnie ~jdlewis@jdlewis.org
1161421909 J * coocoon ~coocoon@dslb-084-056-130-131.pools.arcor-ip.net
1161422011 M * coocoon morning
1161422084 J * joep ~Joep@c53753253.cable.wanadoo.nl
1161422222 Q * Johnnie charon.oftc.net oxygen.oftc.net
1161422222 Q * _are_ charon.oftc.net oxygen.oftc.net
1161422222 Q * hardwire charon.oftc.net oxygen.oftc.net
1161422222 Q * Osgiliath charon.oftc.net oxygen.oftc.net
1161422222 Q * mountie charon.oftc.net oxygen.oftc.net
1161422222 Q * pusling charon.oftc.net oxygen.oftc.net
1161422222 Q * WorkRoey charon.oftc.net oxygen.oftc.net
1161422222 Q * morfoh charon.oftc.net oxygen.oftc.net
1161422222 Q * vasko charon.oftc.net oxygen.oftc.net
1161422222 Q * romke charon.oftc.net oxygen.oftc.net
1161422222 Q * ex charon.oftc.net oxygen.oftc.net
1161422222 Q * Nam charon.oftc.net oxygen.oftc.net
1161422222 Q * kevinp charon.oftc.net oxygen.oftc.net
1161422222 Q * mugwump charon.oftc.net oxygen.oftc.net
1161422222 Q * besonen charon.oftc.net oxygen.oftc.net
1161422222 Q * duckx charon.oftc.net oxygen.oftc.net
1161422222 Q * fosco charon.oftc.net oxygen.oftc.net
1161422222 Q * bj charon.oftc.net oxygen.oftc.net
1161422222 Q * Wonka charon.oftc.net oxygen.oftc.net
1161422222 Q * trippeh charon.oftc.net oxygen.oftc.net
1161422222 Q * weasel charon.oftc.net oxygen.oftc.net
1161422222 Q * kaner charon.oftc.net oxygen.oftc.net
1161422222 Q * ray6 charon.oftc.net oxygen.oftc.net
1161422222 Q * tanjix charon.oftc.net oxygen.oftc.net
1161422222 Q * Hunger charon.oftc.net oxygen.oftc.net
1161422222 Q * FloodServ charon.oftc.net oxygen.oftc.net
1161422222 Q * daniel_hozac charon.oftc.net oxygen.oftc.net
1161422222 Q * mcp charon.oftc.net oxygen.oftc.net
1161422222 Q * Hollow charon.oftc.net oxygen.oftc.net
1161422222 Q * Greek0 charon.oftc.net oxygen.oftc.net
1161422222 Q * weeble charon.oftc.net oxygen.oftc.net
1161422222 Q * yang charon.oftc.net oxygen.oftc.net
1161422222 Q * samuel_ charon.oftc.net oxygen.oftc.net
1161422222 Q * cehteh charon.oftc.net oxygen.oftc.net
1161422222 Q * Loki|muh charon.oftc.net oxygen.oftc.net
1161422222 Q * wenchien charon.oftc.net oxygen.oftc.net
1161422222 Q * [PUPPETS]Gonzo charon.oftc.net oxygen.oftc.net
1161422222 Q * neuralis charon.oftc.net oxygen.oftc.net
1161422222 Q * nebuchadnezzar charon.oftc.net oxygen.oftc.net
1161422222 Q * derjohn charon.oftc.net oxygen.oftc.net
1161422222 Q * virtuoso charon.oftc.net oxygen.oftc.net
1161422222 Q * michal` charon.oftc.net oxygen.oftc.net
1161422222 Q * eGnarF charon.oftc.net oxygen.oftc.net
1161422222 Q * gluk charon.oftc.net oxygen.oftc.net
1161422222 Q * phedny charon.oftc.net oxygen.oftc.net
1161422222 Q * Zaki charon.oftc.net oxygen.oftc.net
1161422222 Q * kir charon.oftc.net oxygen.oftc.net
1161422237 J * virtuoso ~s0t0na@shisha.spb.ru
1161422239 J * kir ~kir@swsoft-mipt-nat.sw.ru
1161422239 J * Zaki ~Zaki@212.118.110.193
1161422239 J * phedny ~mark@phedny.vps.van-cuijk.nl
1161422239 J * gluk ~kvirc@breuss.ws.ehouse.ru
1161422239 J * eGnarF ~bartek@bk.crystone.se
1161422239 J * michal` ~michal@www.rsbac.org
1161422239 J * derjohn ~derjohn@80.69.37.19
1161422239 J * Johnnie ~jdlewis@jdlewis.org
1161422239 J * _are_ ~are@gprs-pool-1-014.eplus-online.de
1161422239 J * hardwire ~hardwire@89-208-58-66.gci.net
1161422239 J * Osgiliath ~abigor@kurzweil.no-ip.org
1161422239 J * kevinp ~kevinp@ny.webpipe.net
1161422239 J * fosco fosco@konoha.devnullteam.org
1161422239 J * daniel_hozac ~daniel@c-2c1472d5.010-230-73746f22.cust.bredbandsbolaget.se
1161422239 J * mcp ~hightower@wolk-project.de
1161422239 J * Hollow ~hollow@styx.xnull.de
1161422239 J * Greek0 ~greek0@85.255.145.201
1161422239 J * weeble ~weeble@81.52.144.1
1161422239 J * yang ~yang@cpe-213-157-253-172.dynamic.amis.net
1161422239 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1161422239 J * pusling pusling@195.215.29.124
1161422239 J * WorkRoey ~katz@h-69-3-4-130.mclnva23.covad.net
1161422239 J * mugwump ~samv@watts.utsl.gen.nz
1161422239 J * besonen ~besonen@dsl-db.pacinfo.com
1161422239 J * duckx ~Duck@tox.dyndns.org
1161422239 J * neuralis ~krstic@solarsail.hcs.harvard.edu
1161422239 J * Loki|muh loki@satanix.de
1161422239 J * Nam ~nam@S0106001195551ff0.va.shawcable.net
1161422239 J * weasel weasel@weasel.noc.oftc.net
1161422239 J * nebuchadnezzar ~nebu@zion.asgardr.info
1161422239 J * wenchien ~wenchien@59-105-176-11.adsl.static.seed.net.tw
1161422239 J * cehteh ~ct@pipapo.org
1161422239 J * Hunger Hunger.hu@213.163.11.138
1161422239 J * kaner kaner@strace.org
1161422239 J * trippeh atomt@x.vx.no
1161422239 J * ex ex@valis.net.pl
1161422239 J * bj ~bj@insanefactory.com
1161422239 J * romke ~romke@acrux.romke.net
1161422239 J * Wonka produziert@chaos.in-kiel.de
1161422239 J * vasko ~vasko@unreal.rainside.sk
1161422239 J * morfoh ~morfoh@kilo105.server4you.de
1161422239 J * samuel_ ~samuel@jupe.quebectelephone.com
1161422239 J * ray6 ~ray@194.126.159.45
1161422239 J * [PUPPETS]Gonzo gonzo@langweiligneutral.deswahnsinns.de
1161422239 J * tanjix ~tanjix@office.star-hosting.de
1161422239 J * FloodServ services@services.oftc.net
1161422270 Q * phedny Read error: Connection reset by peer
1161422270 Q * michal` Read error: Connection reset by peer
1161422282 J * phedny ~mark@phedny.vps.van-cuijk.nl
1161422290 J * michal` ~michal@www.rsbac.org
1161422387 Q * kir Read error: Operation timed out
1161422542 Q * _are_ Ping timeout: 480 seconds
1161422968 J * kir ~kir@swsoft-mipt-nat.sw.ru
1161423799 Q * joep Quit: Leaving
1161423914 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1161425301 M * derjohn2 Bertl_zZ, yesterday Iask what the DMA engine kernel option is. By random I found that news:http://www.golem.de/0610/48503.html 
1161425692 Q * wenchien Ping timeout: 480 seconds
1161425733 M * waldi hmm, each pci device can use dma, what is the advantage of this?
1161425748 M * waldi i hope they also implement an iommu
1161426201 J * Piet hiddenserv@tor.noreply.org
1161426202 M * derjohn2 Well, I just wondered what the new dma-engine option in 2.6.18.1 is
1161426221 M * derjohn2 (as DMA fir NICs should be around for some time already)
1161426440 M * doener I read it, that DMA engines work both ways NIC -> kernel and kernel -> userspace. While good old dma only does the former. Does that make sense?
1161426484 M * derjohn2 that would be a clear advantage!
1161426534 M * derjohn2 especially for the folks doing routing on Linux ... might be on reason less to buy cisco.
1161426581 M * derjohn2 hm, does the DMA Engine need interrupts? Or does itr simply run "autonomously"?
1161426765 J * meandtheshell ~markus@85-124-232-24.work.xdsl-line.inode.at
1161429258 J * matled_ ~matled@85.131.246.184
1161429391 Q * matled Remote host closed the connection
1161429391 N * matled_ matled
1161430989 J * bonbons ~bonbons@83.222.36.111
1161433958 M * doener anyone running a 2.6.17 vserver kernel?
1161433989 M * doener ah, nevermind, I still got one myself
1161434381 M * doener waldi: regarding your chroot related mail to lkml, how does the behaviour for /proc/1/root actually differ between 2.6.18 and previous kernels?
1161434582 M * waldi its the behaviour of /proc/*/{cwd,root}
1161434631 M * waldi old kernels checked if the location is within the root of the process and refused to read it if not
1161434636 M * doener your example used /proc/1/root, and that seems to be equal for 2.6.19-rc1 and 2.6.16.something
1161434672 M * waldi pid 1 is usualy running outside of a chroot and it was possible to check if the root of the current process is different
1161434807 M * doener http://paste.linux-vserver.org/532
1161434855 M * waldi exactly, this was different before .18
1161434871 M * waldi err
1161434885 M * doener hm? there are two things in that paste... .16.19 and .19-rc1
1161434935 M * waldi hrm
1161435028 M * waldi hmm, your .16 is a vserver kernel
1161435083 M * waldi root@lxdebian:/srv/chroot# chroot sid/buildd/active /bin/sh
1161435083 M * waldi sh-3.1# ls -al /proc/1/root
1161435083 M * waldi ls: cannot read symbolic link /proc/1/root: Permission denied
1161435083 M * waldi lrwxrwxrwx 1 root root 0 Oct 21 12:51 /proc/1/root
1161435098 M * waldi Linux lxdebian.bfinv.de 2.6.17-3-s390x #1 SMP PREEMPT Fri Sep 15 14:29:30 CEST 2006 s390x GNU/Linux
1161435109 M * doener interesting
1161435126 M * trippeh build-etch64:/proc# ls -l
1161435126 M * trippeh total 4259846
1161435126 M * trippeh ?--------- ? ?     ?              ?            ? 1
1161435135 M * waldi vserver kernel
1161435137 M * waldi ?
1161435141 M * trippeh Yep, 2.6.18
1161435145 M * waldi it behaves different anyway
1161435212 M * doener trippeh: inside a vserver? IIRC the latest patches have a new fakeinit thing, so that might not tell anything at all ;)
1161435235 M * trippeh I'm just spewing it out :P
1161435243 M * doener waldi: does that readlink change also allow the namespace crossing?
1161435256 M * waldi doener: possible
1161435274 M * doener I wrote a quick chroot++, but a plain cd /proc/1/root also gets me outta there...
1161435284 M * doener I'm pretty sure that this was impossible in the past
1161435289 M * waldi doener: readlink returns a different location that a chdir does
1161435317 M * doener after all, the corresponding mount does not even exist in that namespace
1161435328 M * waldi hihi
1161435390 M * trippeh Oh well. Never mount /proc inside plain chroots, at least if its used to constrain daemons and such.
1161435411 M * trippeh (and don't run the process as root) ;)
1161435482 A * doener wonders if his box will do weird things if he umounts the non-existant mounts...
1161435495 M * doener (expect me to be gone into panic soon *g*)
1161435503 A * waldi managed to umount the real / once
1161435516 M * waldi got weird effects
1161435529 M * trippeh I got the kernel to do that for me a few times, by having a raid controller drop off the pci bus ;)
1161435545 M * trippeh The apps ran fine for weeks though, from whatever was still in RAM
1161435552 M * waldi hihi
1161435587 M * nox my daily fool for a minute: what would happen if i would scp a /dev node into a guest? could i evade the mknod limitation? 
1161435622 M * trippeh Try it ;)
1161435632 M * doener gets me an -EINVAL... at least one check that works ;)
1161435669 M * nox trippeh: no idea for a quick test (maybe hda from a livecd)
1161435722 J * weichong78 ~weichong7@219.95.29.124
1161435726 M * trippeh I guess the ssh server would have to actually mknod itself, so I guess it'll get blocked.
1161435786 M * nox well cp -a from host works so i came to that question
1161435831 M * daniel_hozac because you're executing it in the host...
1161435832 M * waldi hmm, anyone knows the behaviour of /dev/tty? when can I use it?
1161435851 M * daniel_hozac waldi: doesn't it refer to the controlling tty?
1161435865 M * weichong78 hello, im a newbie in vserver, just compiled the kernel for 2.6.12.4-vs2.0 and util-vserver-0.30.210-0, testme.sh give me "chbind: kernel does not provide network virtualization", please help
1161435885 M * waldi daniel_hozac: possible
1161435886 M * daniel_hozac weichong78: why on earth did you get such an old kernel?
1161435901 M * daniel_hozac weichong78: and why old utils?
1161435916 M * doener waldi: yeah, namespace boundaries also broke in .18
1161435918 M * doener :(
1161435923 M * waldi doener: yeah
1161435937 M * waldi doener: can we see this as security problem?
1161435978 M * weichong78 daniel_hozac: because im still using my old FC3 and when i attempt to compile 2.6.17, vserver compiles fine but somehow my FC3 adsl rp-pppoe doesn't work at all, so i can't go online, so i was plannign to play around with this old kernel while waiting for FC6.
1161436003 M * waldi doener: quick fix: readd the root check for root/cwd?
1161436010 M * doener not sure... but it completely breaks my chroot++ (which doesn't chroot but creates a cleaned namespace, eg. chdir("..") attacks won't work)
1161436030 M * waldi doener: please provide the source
1161436070 M * waldi doener: but yes, its likely to break namespace boundaries
1161436077 M * doener daniel_hozac: http://people.linux-vserver.org/~doener/
1161436090 M * doener ehrm, s/daniel_hozac/waldi/
1161436102 M * doener and the wrong link on top of that... *sigh*
1161436106 M * doener http://people.linux-vserver.org/~doener/chroot++.c
1161436151 M * doener it's a little hackish, but it works (well, on pre .18 kernels, on .18 it's just a chroot with a nice /proc/mounts output)
1161436152 M * nox weichong78: pastebin  grep -i vserver /boot/config-`uname -r`  anywhere
1161436185 M * waldi doener: bah
1161436190 M * nox weichong78: i.e. http://paste.linux-vserver.org/
1161436231 M * doener waldi: is it such a bad quality hack? ;)
1161436249 M * daniel_hozac weichong78: you really should use a more recent kernel, even if it means you have to upgrade rp-pppoe.
1161436264 M * waldi doener: except pivot_root, but there is no other possibility
1161436289 M * waldi doener: initramfs have the same problem, you can't umount the inital root
1161436299 M * weichong78 nox: http://paste.linux-vserver.org/533
1161436317 M * nox weichong78: maybe forgot ppp support? could not imagine how kernel should brake pppoe
1161436319 M * weichong78 daniel_hozac: this is just a temporary measure ;) i will upgrade it in a few days time
1161436352 M * weichong78 nox: nope, in the new kernel, i tried everythign from enable all ppp to disable all ppp, none works with my FC3 
1161436477 M * nox weichong78: sry no idea
1161436493 M * weichong78 nox: no worries, thanks for the help anyway
1161436520 M * weichong78 however, chbind fail just means i can't use vserver's network right, but i can still play around with it otherwise?
1161436706 M * daniel_hozac well, chbind is required to get a guest up and running...
1161436718 M * weichong78 aiks
1161436725 M * daniel_hozac you could probably recompile your utils with --enable-apis=NOLEGACY to get it to work though.
1161436734 M * weichong78 oh ok
1161436739 M * weichong78 thanks for the tips
1161436741 M * daniel_hozac why did you enable CONFIG_VSERVER_LEGACY_VERSION=
1161436746 M * daniel_hozac s/=/?/
1161436778 M * weichong78 i didn't really know much about it yet
1161436789 M * weichong78 just enable everything that says vserver :P
1161436792 M * daniel_hozac didn't you read the help?
1161436844 M * weichong78 i only read the http://oldwiki.linux-vserver.org/FedoraCore3_HowTo.
1161436904 M * weichong78 oh you mean the help during menuconfig?
1161436908 M * daniel_hozac yep.
1161436912 M * weichong78 oops didn't do that, how foolish of me
1161436913 M * weichong78 :(
1161439135 P * weichong78 
1161439614 J * weichong78 ~weichong7@219.95.252.6
1161439627 J * Blissex ~Blissex@82-69-39-138.dsl.in-addr.zen.co.uk
1161439649 M * weichong78 daniel_hozac: recompiled kernel with that legacy thingy disabled, now testme.sh works.  thanks a hunderd
1161439658 M * weichong78 s/hunderd/hundred/
1161439697 M * daniel_hozac np
1161439928 P * weichong78 
1161440122 N * Bertl_zZ Bertl
1161440127 M * Bertl morning folks!
1161440159 M * Borg- afternoon Bertl 
1161440657 Q * dmax Ping timeout: 480 seconds
1161440740 M * Radiance BertI, afternoon :)
1161440779 M * Radiance about the changes, i dropped the prempt and it looks like performance is better, si percentage stays almost to nothing
1161440788 M * Bertl good
1161440796 M * Radiance i kept the hz at 250 ... just keeping it in mind in case it needs some more tuning
1161440812 M * Bertl any negative effects of the preempt removal?
1161440847 M * Radiance currently none, i will keep an eye on it for the next few days and see if things stay fine
1161440878 M * Bertl okay, let us know if ...
1161440925 M * Radiance also for the record, the option kernel does irq balancing is enabled...
1161440933 M * Radiance not sure if i should leave it like that or not
1161440980 M * Bertl well, if you don't want to 'assign' irqs, it's probably the best
1161441048 M * Radiance true indeed, i was thinking the userland irqbalance could be an alternative but it actually doesn't matter that much
1161441102 M * Radiance on a dual xeon box i had to disable the kernel option and use the irqbalance in userland else the cpu's and ht didn't spread the irq load equally
1161441118 M * Radiance (it kept them tied to only 1 cpu)
1161441139 M * goblin hmm, seems like vroot is oopsing my kernel
1161441146 M * Bertl Radiance: which (btw) could be perfectly fine
1161441158 M * Bertl goblin: what kernel/patches?
1161441179 M * Bertl goblin: do you have a stack trace?
1161441189 M * goblin yup one sec
1161441189 M * Radiance yeah, i still didn't figure out what is "best" ...let 1 cpu handle them or spread them hehe but for now it's working fine
1161441208 M * Bertl okay
1161441243 M * goblin Linux leviathan 2.6.17.11-grsec2.1.9-vs2.0.2 #4 SMP Wed Oct 4 19:22:03 BST 2006 x86_64 x86_64 x86_64 GNU/Linux
1161441247 M * goblin that's the uname
1161441266 M * Bertl hmm, an older kernel and grsec patches ...
1161441273 M * Bertl stack trace?
1161441287 M * goblin one sec, pasting
1161441338 M * goblin http://pastebin.co.uk/4451
1161441353 M * goblin well, the latest kernel to which vserver + grsec are available, afaik
1161441364 M * Bertl hmm? see topic
1161441398 M * daniel_hozac even the wiki is updated.
1161441442 M * Bertl goblin: do you have the x86_64? kernel at hand?
1161441443 M * goblin OHH, 2.6.17.13, blind me
1161441452 M * Borg- hmm?
1161441472 M * Borg- whats wrong w/ 2.6.17.13 ?
1161441477 M * goblin Bertl, yes, I have x86_64 arch, and I guess I do have the kernel at hand, depends what you mean by that
1161441490 M * goblin Borg-, nothing, it's just that I have 2.6.17.11
1161441491 M * Bertl the vmlinux not the bzImage
1161441505 M * Borg- ah ;) ok.. since I run 2.6.17.13
1161441506 M * goblin Bertl, I have the sources somewhere
1161441523 M * daniel_hozac 2.6.17.14 is the latest, btw.
1161441543 M * Borg- daniel_hozac: any important fixes 2.6.17.13 -> 2.6.17.14 ?
1161441546 M * Bertl goblin: I doubt that your kernel contains debug info, but it's worth a try
1161441566 M * Bertl goblin: let's check with a working addr2line or gdb on the vmlinux
1161441568 M * daniel_hozac Borg-: i haven't really checked.
1161441581 M * goblin Bertl, no, it doesn't contain debug info
1161441602 M * goblin Bertl, the problem occured after I set up a vroot device,. ..
1161441611 M * Bertl okay, do you think you could enable the various debug options and recreate this issue?
1161441614 M * daniel_hozac goblin: did you try to mount the vroot device?
1161441615 M * goblin I issued vserver test start, and it oopsed when mounting
1161441636 M * daniel_hozac goblin: or what makes you think the mount oops is related to the vroot?
1161441637 M * Bertl ah, yes, sounds like it ... interesting
1161441638 M * goblin daniel_hozac, yes, the vroot device was already mounted inside the vserver before I started it
1161441649 M * daniel_hozac ... you can't mount the vroot device.
1161441652 M * goblin daniel_hozac, because everything works fine without vroot
1161441660 M * goblin Bertl, quite difficult, I'm afraid
1161441695 M * goblin maybe on some other box, this one is my private shell server being used by various people
1161441741 M * daniel_hozac goblin: if you could tell us the exact steps you used to get the oops, that'd probably help.
1161441748 M * Bertl okay, well, I'm inclined to look into the vserver+grsec oops, if it has debug info attached (or you can figure the address to line mapping)
1161441766 M * Bertl usually, I'd suggest to try to recreate without grsec
1161441790 M * goblin hmm
1161441809 M * goblin Bertl, ok, I'll try to recreate it some other time
1161441825 M * goblin not today though ;-(
1161441839 M * Bertl okay, is the vroot compiled as module?
1161441841 M * daniel_hozac even just a function name might be helpful.
1161441842 M * goblin daniel_hozac, I can...
1161441847 M * goblin Bertl, it's not
1161441862 M * Bertl so into the kernel, and you configured it with vrsetup?
1161441877 M * goblin yes, I'm pretty sure I have
1161441881 M * goblin lemme go to the source
1161441914 M * goblin CONFIG_BLK_DEV_VROOT=y
1161441915 M * goblin yeah
1161441919 M * Bertl and you copied it into the guest, instead of the 'original' device
1161441958 M * Bertl what filesystem does the guest use?
1161441978 M * goblin well, I mounted the fs on the guest, and entered /dev/vroot/0 /mountpoint xfs usrquota,grpquota in the mtab for this vserver
1161441981 M * goblin xfs
1161441987 M * goblin everything uses xfs
1161442001 M * Bertl mtab not fstab, yes?
1161442022 M * goblin oh no, sorry, fstab
1161442031 M * daniel_hozac which fstab? the /etc/fstab inside the guest?
1161442032 M * goblin /etc/vservers/test/fstab
1161442039 M * goblin no, the one in host
1161442058 M * Bertl interesting :)
1161442093 M * Bertl might explain the oops, daniel_hozac: do you have a setup to test this?
1161442102 M * daniel_hozac i should be able to reproduce it.
1161442110 M * goblin oh, actually now I read it again, it should rather be in init/mtab, apparently
1161442119 M * Bertl yep
1161442140 M * Bertl nevertheless, the vroot should not oops on this 'misuse' :)
1161442151 M * goblin yup, definitely :-)
1161442160 M * goblin ok, can I help you to get function start points somehow?
1161442172 M * goblin from the kernel image?
1161442204 M * Bertl try with 'addr2line -e vmlinux ffffffff80356e2f
1161442222 M * Bertl if that just gives you ??:0
1161442245 M * Bertl try with 'gdb vmlinux' then l *0xffffffff80356e2f
1161442269 J * yarihm ~yarihm@84-74-17-6.dclient.hispeed.ch
1161442305 M * daniel_hozac Bertl: will vr_open work if the vroot device hasn't been initialized?
1161442313 M * goblin hang on, I have to install gdb first
1161442321 M * daniel_hozac i.e. won't vr be NULL?
1161442327 M * goblin (it gave me ??:0)
1161442338 M * Bertl daniel_hozac: hmm, good point, have to check that
1161442360 M * daniel_hozac hmm, no, that doesn't make sense, vr isn't NULL in the ioctl.
1161442595 M * daniel_hozac is there a way to get at least a function name from the addresses even without debug info? i imagine there has to be.
1161442616 M * Bertl yeah, from the map
1161442642 M * Bertl System.map check for the next lower address there
1161442660 M * goblin I don't get it then, if I am to put /dev/hdv1 to this mtab, how will the guest now which of the 8 vroot devices it should use?
1161442677 M * daniel_hozac because you copy the correct one to it.
1161442695 M * daniel_hozac each guest will only have one /dev/hdv1.
1161442696 M * goblin aaaah, here
1161442698 M * goblin blind me
1161442706 M * goblin right...
1161442718 M * goblin but in this case it means I can only have one vroot device per vserver
1161442725 M * Bertl nah
1161442725 M * daniel_hozac no it doens't.
1161442741 M * goblin hmm
1161442742 M * Bertl you can name them however you like
1161442748 M * goblin oh, true actually
1161442749 M * Bertl *whatever
1161442755 M * goblin ok, that's something new :-)
1161442768 M * Bertl so e.g. the following would be quite fine:
1161442786 M * Bertl guest A: /dev/hdv1 (vr0) and /dev/hdv2 (vr1)
1161442795 M * Bertl guest B: /dev/hdv1 (vr0) and /dev/hdv2 (vr2)
1161442805 M * goblin (gdb) l *0xffffffff80356e2f
1161442805 M * goblin No symbol table is loaded.  Use the "file" command.
1161442826 M * goblin yeah, I get it now. thanks, I'll try it soon
1161442830 M * Bertl okay, check the System.map file, for the 'next lower' address if possible
1161442861 M * Bertl goblin: btw, is your leviathan armed? :)
1161442874 M * goblin ffffffff80356ad0 T xfs_buf_get_flags
1161442891 M * goblin and straight after that, ffffffff80356f80 T xfs_buf_pin
1161442905 M * Bertl okay, and for this one:
1161442908 M * Bertl ffffffff8028366d
1161442909 M * goblin Bertl, it knows how to bite, yes
1161442934 M * goblin ffffffff80283650 T bio_add_page
1161442934 M * goblin ffffffff80283680 t bio_free_map_data
1161443061 J * s0undt3ch ~s0undt3ch@81.193.63.211
1161443263 M * goblin so if the / isn't quota-enabled, but only /home is, for example
1161443298 M * goblin I'd put something like hdv1 / ufs rw 0 0 and hdv2 /home ufs rw,usrquota,grpquota 0 0
1161443301 M * goblin ?
1161443321 M * Bertl yes, and make hdv2 a vroot device
1161443330 M * goblin cool
1161443342 M * Bertl no need to have a special hdv1 at all, just a file would be enough
1161443352 M * daniel_hozac even that isn't required.
1161443368 M * Bertl IIRC, some mount versions check that
1161443391 M * Bertl (i.e. it looks strange if missing, but it's fine, e.g. on mandriva)
1161443435 M * Hollow Bertl: two things that still may be missing but not realy needed right now: rlimit_mask syscall for accounted values, ctx_stat for nx
1161443492 M * Bertl okay, we should make a todo list (like on the old wiki)
1161443515 M * Bertl I'll check the old todo tonight and will migrate/add pages
1161443522 M * Hollow ok, thanks
1161443530 M * daniel_hozac Bertl: ok, reproduced it. vrsetup /dev/vroot0 /dev/loop0; mount -t xfs /dev/vroot0 /mnt seems to do it.
1161443543 M * daniel_hozac -t ext3 or no -t at all doesn't show it.
1161443556 M * daniel_hozac so i'd say it's an xfs bug.
1161443568 M * daniel_hozac (the filesystem is ext3, FWIW)
1161443648 M * Bertl okay, do you have a trace? maybe we can fix it
1161443667 M * daniel_hozac damn, wrong kernel booted... i'll reboot into a vanilla kernel and get a trace from that.
1161443677 M * Bertl lol
1161443774 M * daniel_hozac i forgot i was testing my new FC5 kernel, hehe.
1161443936 M * goblin :-)
1161444140 M * daniel_hozac ... and my vanilla kernel doesn't have xfs, please hold.
1161444194 M * goblin looks like only I use such dodgy fses ;-)
1161444216 M * harry xfs for president!
1161444238 M * harry (so much for useful contributions)
1161444263 M * goblin CRRAPP forgot to remove the entry from fstab, oopsed again ;-(
1161444272 M * daniel_hozac heh.
1161444451 M * Bertl okay, have to leave now ... will be back later ...
1161444457 N * Bertl Bertl_oO
1161444461 M * harry don't die Bertl_oO ! :)
1161444517 Q * weasel Quit: brb
1161444537 J * weasel weasel@asteria.debian.or.at
1161445158 J * coocoon ~coocoon@dslb-084-056-204-036.pools.arcor-ip.net
1161445665 J * bronson ~bronson@c-71-198-75-160.hsd1.ca.comcast.net
1161446156 M * goblin right. now I got to the point where running quotaon inside guest returns no such device or address for vr0
1161446184 M * goblin quotacheck ran fine (although I'm not sure what it did, and what it meant by writing aquota.{group,user} files)
1161446202 M * goblin and when I try to check the quota for a user, it erroneously states that there are no quotas for this user
1161446241 M * daniel_hozac are you running it on the device or the mount point?
1161446249 M * daniel_hozac IIRC the mount point doesn't work if you're using XFS.
1161446261 M * goblin I'm running quotaon on the vroot device
1161446567 M * daniel_hozac did you run vrsetup since rebooting?
1161446584 M * daniel_hozac Bertl_oO: http://paste.linux-vserver.org/534
1161446594 M * goblin a very good question
1161446623 M * goblin I haven't! :-)
1161446698 M * goblin now it says quotaon: using /srv/aquota.group on /dev/vroot/0 [/srv]: Function not implemented
1161446704 M * goblin bugger.
1161446715 M * daniel_hozac and you have ufs in /etc/mtab inside the guest?
1161446719 M * daniel_hozac (as the filesystem)
1161446729 M * goblin yes
1161446754 M * daniel_hozac if you strace it, which call is failing?
1161446755 M * goblin in mtab yes, in /proc/mounts it appears as xfs
1161446765 M * daniel_hozac of course.
1161446775 M * goblin quotactl(Q_QUOTAON|USRQUOTA, "/dev/vroot/0", 2, {8462652031284441903, 8243121640098067567, 0, 49, 93824993365008, 93824993365018, 3406201476451877717, 4922226899865108527}) = -1 ENOSYS (Function not implemented)
1161446791 M * goblin hmm, maybe that's because of --bind, I'll check...
1161446793 M * daniel_hozac well, uh, do you have quotas enabled in your kernel?
1161446811 M * goblin oh, definitely
1161446817 M * goblin they work in the host
1161446850 M * goblin no, this one was even without --bind
1161446891 M * goblin oh well. I guess I'll write a small service that will run on the host and that will pass the quotas over IP to the guest
1161446917 M * goblin and make a client that I'll call 'quota' which will connect to the server on the host rather than use sysctls
1161446967 M * daniel_hozac so what does grep Q..T .config give you?
1161447034 M * goblin CONFIG_CPU_FREQ_STAT=y
1161447034 M * goblin # CONFIG_CPU_FREQ_STAT_DETAILS is not set
1161447034 M * goblin CONFIG_XFS_QUOTA=y
1161447034 M * goblin CONFIG_QUOTA=y
1161447034 M * goblin # CONFIG_QFMT_V1 is not set
1161447036 M * goblin CONFIG_QFMT_V2=y
1161447038 M * goblin CONFIG_QUOTACTL=y
1161447041 M * goblin :-)
1161447101 Q * Blissex Remote host closed the connection
1161451362 Q * derjohn2 Ping timeout: 480 seconds
1161451398 J * derjohn2 ~aj@dslb-084-058-241-218.pools.arcor-ip.net
1161451711 M * Borg- guys.. why are you touching HZ ?? As far as I understand.. having it higher than 100 for x86 (not alpha or sparc) is nonsense..
1161451991 M * Hollow isn't 1000 Hz standard for 2.6?
1161452069 J * the_hydra ~a_mulyadi@202.59.168.29
1161452077 M * the_hydra Bertl_oO: ping
1161452976 M * the_hydra for Vserver, does various HZ settings bring any difference for guest's perfomance?
1161453235 M * doener the_hydra: should be the same as for a normal host that runs the same set of processes
1161453251 M * the_hydra humm
1161453295 M * doener the_hydra: but if you use the token bucket scheduler, the settings might need to be tweaked of match the HZ 
1161453304 M * doener s/of match/to match/
1161453377 M * the_hydra ehm, pardon me, token bucket scheduler is ... ?
1161453394 M * the_hydra something like resource management?
1161453441 M * doener cpu resources, yep. Each vserver has a bucket of tokens from which it takes a token whenever a process is running and if it runs out of tokens, it is stopped
1161453487 M * the_hydra oh, something like process scheduler implemented inside vserver...CMIIW
1161453502 M * doener there are some parameters to control how big the bucket is, how often it is refilled, how much tokens need to be in it to make a vserver start working after it has been stopped and some more
1161453542 M * doener it's just an extension on top of the vanilla scheduler
1161453550 M * the_hydra got it
1161453646 M * the_hydra ok...so since it is related with scheduling, changing HZ should be followed by tweaking the bucket calculation..
1161453671 Q * yarihm Quit: Leaving
1161453764 M * doener in general, it's not that critical, but because the granularity changes, the behaviour can change a little
1161453792 M * the_hydra ah yes, you're correct...granularity
1161453841 M * doener if you had token bucket settings for HZ=100 that gave you about 66% cpu for the vserver, that might be more like 70% if you change to HZ=250 (if I did the math correctly :)
1161453849 M * the_hydra doener: away from this bucket stuff, between HZ=100, HZ=250 and HZ=1000, which one do you think is the ideal one for virtualization?
1161453908 M * daniel_hozac there's nothing ideal for virtualization, it's just more or less ideal for your workload.
1161453920 M * the_hydra 66%, then 70%? are you sure? I think it should be lower due to finer granularity
1161453925 M * daniel_hozac i.e. it depends on what your virtualized guests and the host will be doing.
1161454031 M * doener the_hydra: yeah, might also tend towards 50% not sure about it. Anyway, my point was that you hit different intervals with 250 than with 100 (eg. HZ=100: 0, 10, 20, 30ms but HZ=250: 0, 4, 8, 12, 16ms)
1161454068 M * the_hydra doener: ok, i got your essential points anyway...thanks :)
1161454092 M * the_hydra btw, where is herbert? is he around?
1161454121 M * daniel_hozac he's oO :)
1161454123 M * doener _oO means sth. along the lines of "out of Office" or "out of Order" (his words ;) )
1161454137 M * the_hydra oh :)
1161454140 M * daniel_hozac haha
1161454150 M * the_hydra ok now I understand what's that double "O" means :))
1161454232 M * the_hydra anyway, the new Vserver website looks great
1161454248 M * the_hydra especially those developers' photos :)
1161454328 M * Radiance hmm i got something weird going on, have 1 vserver running perfectly on a box, i installed a second vserver but when this one is activated and i try to do ls/mv/cp and other stuff i get this error:
1161454332 M * Radiance mv: error while loading shared libraries: libacl.so.1: cannot enable executable stack as shared object requires: Permission denied
1161454351 M * Radiance now i either forgot what i did to the first vserver to make it work fine, or this is something new
1161454374 M * the_hydra grsec?
1161454388 M * the_hydra stack no exec?
1161454423 M * Radiance yes grsec is active, but why does the first vserver work just fine ?
1161454441 M * daniel_hozac different distribution?
1161454461 M * Radiance i used this for both:  -m debootstrap -- -d sarge
1161454491 M * Radiance http://bugs.donarmstrong.com/cgi-bin/bugreport.cgi?bug=321717
1161454500 M * daniel_hozac and you've done nothing different?
1161454509 M * Radiance found this, but i'm not sure if i understand since they don't list a solution just mumble something about fixed at x place
1161454516 M * Radiance nope, nothing different
1161454641 M * daniel_hozac umm, it says it's been fixed in glibc 2.3.5-4.
1161454661 M * the_hydra permission denied...hmm
1161454739 M * Radiance i'm trying to manually use dpkg to install it but it dies with the same error (ofcourse)
1161454759 M * Radiance and i'm kinda surprised that it's sarge which should have not done this
1161454762 M * the_hydra but the error message really seems like coming from the loader (ld.so)
1161454792 M * daniel_hozac and i really don't think you're hitting that bug.
1161454877 M * Radiance when i do a dpkg -l | grep glibc then i don't see this: ii  libglib2.0-0   2.6.4-1
1161454891 M * Radiance which does appear when i run that command on the host or on the first vserver
1161454959 M * daniel_hozac if you see libglib2.0-0 when you do | grep glibc, something is seriously wrong.
1161454970 M * daniel_hozac given that libglib2 doesn't contain glibc.
1161455014 M * Radiance ah sorry for that
1161455024 M * Radiance nvm the dpkg output
1161455051 M * Radiance this is what i get when looking for libc:
1161455052 M * Radiance iF  libc6          2.3.6.ds1-4    GNU C Library: Shared libraries
1161455074 M * the_hydra what happen if you shutdown the first vserver and just turn on the 2nd?
1161455081 M * Radiance and the other vserver shows this:
1161455099 M * Radiance ii  libc6          2.3.2.ds1-22sa
1161455107 M * Radiance there is the nastiness
1161455114 M * Radiance and looking at the bug report...
1161455114 M * the_hydra hmm
1161455123 Q * the_hydra Read error: Connection reset by peer
1161455145 M * daniel_hozac unless you're running on a 2.4 kernel, that bug report doesn't have anything to do with your problem.
1161455284 M * Radiance ok
1161455297 M * Radiance lemme shutdown the first vserver and try the other
1161455352 J * the_hydra the_hydra@61.5.104.89
1161455361 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1161455395 M * Radiance didn't make a difference
1161455401 M * Radiance during the start i already get this:
1161455403 M * Radiance Starting MTA: /usr/sbin/exim4: error while loading shared libraries: libgcrypt.so.11: cannot enable executable stack as shared object requires: Permission denied
1161455441 M * doener Radiance: hm, is one guest 32bit and the other 64bit?
1161455463 M * Radiance nope
1161455467 M * Radiance all 32bit
1161455520 M * Radiance i'm almost sure that it's some bug or so from where the image was downloaded by debootstrap
1161455527 Q * mire Quit: Leaving
1161455543 M * doener did you manually install the etch libc6?
1161455555 M * Radiance this is not etch but sarge
1161455570 M * Radiance i used:  -m debootstrap -- -d sarge
1161455580 M * doener 2.3.6.ds1-4 is etch's libc6 version
1161455587 M * Radiance that is really fucked up
1161455591 M * Radiance <sigh>
1161455593 M * doener hm, and that isn't installed correctly
1161455609 M * doener ah, you tried, but dpkg bailed out
1161455643 M * Radiance yup
1161456230 M * Radiance during the install i got this line:
1161456231 M * Radiance I: Retrieving debootstrap.invalid_dists_sarge_main_binary-i386_Packages
1161456246 M * Radiance but it continued as "usual"
1161456347 M * Radiance I: Base system installed successfully.
1161456351 M * Radiance ended with that
1161456357 M * Radiance i don't think it has even to do with libc
1161456563 J * konrad konrad@82.160.3.200
1161457035 M * the_hydra gtg
1161457037 Q * the_hydra 
1161457282 Q * konrad 
1161457853 Q * shedi Quit: Leaving
1161458447 J * rgl Rui@217.129.151.190
1161458451 M * rgl hello
1161458453 M * daniel_hozac hi
1161459051 J * shedi ~siggi@inferno.lhi.is
1161459063 M * Radiance ok
1161459074 M * Radiance i re-installed a thought about every step
1161459081 M * Radiance the vserver did start normal
1161459096 M * Radiance but i remember i did an apt-get upgrade...which installed libc
1161459100 M * Radiance then the issues started
1161459117 M * doener Radiance: check /etc/apt/sources.list
1161459120 M * doener (in the vserver)
1161459120 M * Radiance i'm now going to install locales and it looks like libc6 is going to be installed with it
1161459121 M * Radiance yeah
1161459124 M * Radiance lemme do that indeed
1161459133 M * doener does that really list sarge? I remember some problems with it
1161459155 M * Radiance oh shit
1161459157 M * Radiance it lists testing
1161459171 M * Radiance damn how is this possible they didn't see it 
1161459177 M * doener daniel_hozac: does uv do that?
1161459178 M * Radiance i used apt-setup to setup sources.list in trust
1161459252 M * Radiance corrected, gonna try now
1161459493 M * Radiance doener, it's all working as it should
1161459550 M * Radiance thanks for the tip, i'm still recovering from a flue :-)
1161460244 J * debugger ~Rui@217.129.151.190
1161460349 M * daniel_hozac doener: hmm, not that i know of.
1161460641 J * mire ~mire@7-167-222-85.adsl.verat.net
1161460711 Q * rgl Ping timeout: 480 seconds
1161462360 N * debugger rgl
1161466034 Q * ensc Remote host closed the connection
1161466253 Q * michal` Ping timeout: 480 seconds
1161466492 J * michal` ~michal@www.rsbac.org
1161466637 Q * mire Quit: Leaving
1161467340 Q * rgl Quit: Fui embora
1161467714 J * mire ~mire@7-167-222-85.adsl.verat.net
1161467759 J * DreamerC_ ~dreamerc@59-115-48-218.dynamic.hinet.net
1161468125 Q * DreamerC Ping timeout: 480 seconds
1161468609 Q * Borg- Ping timeout: 480 seconds
1161468777 Q * Johnnie Ping timeout: 480 seconds
1161468855 Q * meandtheshell Quit: exit (0);
1161469316 Q * matti Ping timeout: 480 seconds
1161469899 J * matti matti@linux.gentoo.pl
1161470467 J * Borg- borg@cube.benet.uu3.net
1161471269 J * _are_ are@dslb-084-057-193-068.pools.arcor-ip.net
1161471525 Q * matti Ping timeout: 480 seconds
1161471629 Q * bonbons Quit: Leaving
1161472382 J * Johnnie ~jdlewis@jdlewis.org
1161472424 N * Bertl_oO Bertl
1161472429 M * Bertl evening folks!
1161472457 M * Osgiliath :)
1161472497 M * Bertl how's going?
1161472510 M * Osgiliath fine
1161472538 Q * dna_ Quit: Verlassend
1161472992 Q * _are_ Ping timeout: 480 seconds
1161473006 J * _are_ ~are@62.112.159.81
1161473084 M * Bertl wb _are_!
1161473089 M * _are_ Hi